Untitled

Download as pdf or txt
Download as pdf or txt
You are on page 1of 448

MERCHANTS

&----- . . . - - - . . . . . . . . . . . . . . ------------
This page intentionally left blank
Commerce, Science, and Art
in Early Modern Europe

~dtuxlllJf
Pamela H. Smith & Paula Findlen

Routledge
New York London
Published in 2002 by
Routledge
711 Third Avenue,
New York, NY 10017

Published in Great Britain by

Routledge
2 Park Square, Milton Park,
Abingdon, Oxon, OX14 4RN

Routledge is an imprint of the Taylor & Francis Group.

Copyright © 2002 by Routledge

All rights reserved. No part of this book may be reprinted or reproduced or uti-
lized in any form or by any electronic, mechanical, or other means, now known
or hereafter invented, including photocopying and recording, or in any infor-
mation storage or retrieval system, without permission in writing from the
publisher.

10 9 8 7 6 5 4 3 2

Library of Congress Cataloging-in-Publication Data

Merchants and marvels: commerce, science, and art in early modern Europe
! edited by Pamela H. Smith and Paula Findlen.
p. cm.
Includes bibliographical references and index.
ISBN 0-415-92815-X (hb) - ISBN 0-415-92816-8 (pbk.)
1. Art and science. Nature (Aesthetics). 3. Commerce-Europe-
History. I. Smith, Pamela H., 1957- II. Findlen, Paula.
N72.S3 M47 2001
704.9'43'094-dc21 2001016008

Publisher's Note
The publisher has gone to great lengths to ensure the quality of this reprint
but points out that some imperfections in the original may be apparent.
Contents

Acknowledgments IX

Introduction
Commerce and the Representation of Nature
in Art and Science
PAMELA H . SMITH AND PAULA FINDLEN

t&Part f STRUGGLING WITH REALITY


Visualizing Nature and Producing Knowledge

f Splendor in the Grass


The Powers of Nature and Art in the Age of Durer
LARRY SILV E R AND PAMELA H. SMITH

2 Objects of Art/Objects of Nature


Visual Representation and the Investigation of Nature
PAMELA O. LONG

.3 Mirroring the World


Sea Charts, Navigation, and Territorial Claims
in Sixteenth-Century Spain
ALISON SANDMAN

ij From Blowfish to Flower Still Life Paintings


Classification and Its Images, circa 1600
CLAUDIA SWAN

5 "Strange" Ideas and "English" Knowledge 137


Natural Science Exchange in Elizabethan London
DEBORAH E. HARKNESS

v
r.sSiS1r£ 2 NETWORKS OF KNOWLEDGE
Commerce and the Representation of Nature

6 Local Herbs, Global Medicines


Commerce, Knowledge, and Commodities
in Spanish America
ANTONIO BARRERA

7 Merchants and Marvels 182

Hans Jacob Fugger and the Origins of the Wunderkammer


MARK A. MEADOW

8 Practical Alchemy and Commercial Exchange 201

in the Holy Roman Empire


TARA E. NUMMEDAL

.9 Time's Bodies 223

Crafting the Preparation and Preservation of Naturalia


HAROLD J. COOK

10 Cartography, Entrepreneurialism, and Power


in the Reign of Louis XIV
The Case of the Canal du Midi
CHANDRA MUKERJI

11 'Cornelius Meijer inventor et fecit'


On the Representation of Science in Late
Seventeenth-Century Rome
KLAAS VAN BERKEL

0i\~£n,) CONSUMPTION, ART, AND SCIENCE

12 Inventing Nature 297


Commerce, Art, and Science in the Early
Modern Cabinet of Curiosities
PAULA FINDLEN

VI Contents
1:3 Nature as Art 324
The Case of the Tulip
ANNE GOLD GAR

14 Inventing Exoticism 347


The Project of Dutch Geography and the Marketing
of the World, circa 1700
BENJAMIN SCHMIDT

15 Shopping for Instruments in Paris and London 370


JAMES A. BENNETT

EPILOGUES

A World of Wonders, A World of One 399


LISSA ROBERTS

Questions of Representation 412


THOMAS DACOSTA KAUFMANN

Contributors

Index

Contents VII
This page intentionally left blank
Acknowledgments

This volume of essays began life in two conferences: a session held at the 1997
History of Science Society meeting in San Diego, and a 1999 workshop at the
Clark Library hosted by the Center for Seventeenth- and Eighteenth-Century
Studies at the University of California, Los Angeles, and convened by the two
editors. We are grateful to all the participants of those two conferences.
Peter Reill, director of the Center for Seventeenth- and Eighteenth-Cen-
tury Studies at UCLA, encouraged and fostered this conference as he has
done so many others over the years. We are deeply grateful for all he has
brought to the Center; his presence as Director has made the Clark Library
and southern California into a perhaps unlikely, but, most importantly,
lively and stimulating intellectual center for early modern studies. Without
the help of his staff, primarily Candis Snoddy, Nancy Connolly, Kathryn
Sanchez, and Marina Romani, the conference could not have taken place.
The volume would have come far more slowly into existence without the
enthusiasm and efficiency of our editor, Brendan O'Malley at Routledge,
and the capable assistance of Emily Klancher and Stacey Loughrey.
By all reports, edited volumes are the source of many tribulations. Our
experience has been otherwise; we feel richly rewarded by the conversations
this volume has spurred and by the opportunity the contributors have
afforded us to collect their thoughts into this volume. Our thanks to all of
them for helping us to bring merchants and marvels together.

IX
This page intentionally left blank
Introduction

Commerce and the Representation


of Nature in Art and Science
PAMELA H. SMITH AND PAULA FINDLEN

O n May 20, ISIS, an Indian rhinoceros (Rhinoceros unicornis)


arrived in Lisbon aboard the Nostra Senora de Ajuda, a Portuguese
trading ship from India. It was the first rhinoceros to reach
Europe alive since the third century. A diplomatic gift to Alfonso d' Albu-
querque, governor of Portuguese India (1509-15) by Sultan Muzafar II
(1511-26), ruler of Cambaia, or Gujarat, the animal was passed by Albu-
querque to his king, Don Manuel I (1495-1521) in ISIS. Later in the same
year, Manuel sent the rhino as a diplomatic gift to the Medici Pope Leo X
(1513-21) via Marseilles, where it was seen by King Francis I of France.
Dressed with a gilt-iron chain and a green velvet collar decorated with gilt
roses and carnations, the rhinoceros drowned when its ship sank off the
coast of Italy. It followed in the wake of an elephant named Hanno that
actually made it all the way to the papal city in 1514.'
The rhinoceros, circulating in Europe as part of a gift economy-not yet
commodified-was brought to Europe within the framework of global
commerce. Large gifts of nature were tangible signs of European overseas
expansion, through trade or conquest, into regions where the "marvels of the
East" could be found. If the rhinoceros itself was never fully commodified, its
portrayal in a drawing of ISIS sent to an unknown acquaintance by Albrecht
Durer did circulate as a commodity of sorts (fig. Intro I). Durer had trans-
ferred the visual and verbal description of his drawing to a woodcut that
served as the model for numerous reproductions in the following centuries
and was disseminated very widely. Increasingly the rhinoceros became the
artist's image of it. Durer participated in shaping the verisimilitude of his rhi-
noceros by describing it enthusiastically although he never saw it himself:2

In the year 15[113 [this should read 15151 on I May was brought to our
king of Portugal in Lisbon such a living animal from India called a
a*.
4drn.i Albrecht Diirer, Rhinoceros, 1515. Copyright O The British Museum.

Rhinocerate. Because it is such a marvel I considered that I must send this


representation. It has the colour of a toad and is covered all over with
thick scales, and in size is as large as an elephant, but lower, and is the
deadly enemy of the elephant. It has on the front of the nose a strong sharp
horn: and when this animal comes near the elephant to fight it always first
whets its horn on the stones and runs at the elephant with its head
between its forelegs. Then it rips the elephant where its skin is thinnest
and then gores it. The elephant is greatly afraid of the Rhinocerate; for he
always gores it whenever he meets an elephant. For he is well armed, very
lively and alert. The animal is called rhinocero in Greek and Latin but in
Indian, gomda.

Durer used the term abkuntevfet to describe his portrayal of the animal, a
term that often implied "copied from life,"4 and thus left open to his corre-
spondent the question of whether he had portrayed this particular rhinoc-
eros from the living model. In reality, he probably copied his description
from a report and drawing (now lost) sent to Nuremberg by a member of
the German community in Lisbon, yet Diirer's image of this wonder, which
went on to have an extraordinary life in copies, was probably taken by most
viewers to be an accurate rendering of a strange and exotic beast.5 It had the
descriptive, realist "mark of truth" that became signature in the work of
Durer and his followers, and also appeared in works of scientific illustration

2 Pamela H. Smith and Paula Findlen


at about the same time. 6 By the mid-sixteenth century naturalists would
commonly refer to their illustrations of nature as being done ad vivum,
which they associated strongly with making a "true portrait" of nature.
The story of Durer's rhinoceros points to several important aspects of the
relationships among art, science, and commerce in Europe that will be consid-
ered in this volume. Renaissance and post-Renaissance western Europe played
a unique role in the development of various arts and sciences devoted to the
imitation of nature. The fact that they did this during the same time that Euro-
peans expanded overseas had profound consequences for Europe and the
world. Early modern Europeans sought to master nature through technology
on an unprecedented scale, making the conquest of nature a political impera-
tive from the sixteenth through eighteenth centuries. 7 Their activities resulted
in the development of various arts and sciences devoted to the imitation of
nature, the emergence of new conceptions of nature that responded to political
and material changes, and a new discourse about nature that would become a
central cultural force in Western society. But these results did not emerge sui
generis, and we need to understand better the interconnections among diverse
aspects of the project of understanding, describing, and conquering nature in
order to appreciate fully the significance of these new developments.
During the early modern period a profound transformation in attitudes
toward the natural world, the material environment, and their artistic rep-
resentation occurred within a new environment of global trade and imperial
ambitions in which commodities were produced, accumulated, consumed,
and exchanged. The essays that make up this book explore many intersec-
tions between these developments, such as those suggested by Durer's rhi-
noceros. One main conclusion of these essays concerns the importance of the
coexistence of patronage and commerce in the early modern period and the
way in which these overlapping social and economic systems of establishing
value and significance resulted in an expansion of cultural production that
greatly encouraged the investigation of and familiarity with nature. In addi-
tion, patronage of and commerce in the representations of nature in both art
and science raised the status of individuals who claimed to imitate nature,
such as many of the artist-artisans, medical practitioners, and other investi-
gators of nature discussed in this book. This volume argues that these indi-
viduals helped lay the foundations of the new philosophy, which eventually
would come to be called "science." This new natural philosophy, pursued
with increasing enthusiasm in the late sixteenth and seventeenth centuries,
emphasized practice, the active collection of experience, and observation of
nature. One of the central goals of this volume is to invite readers to consider
how a greater awareness of the importance of commerce in relation to scien-
tific and artistic representations of nature and the growth of new technolo-
gies transforms the conventional story of the Scientific Revolution.

Introduction 3
COMMERCE AND PATRONAGE

The genesis of Durer's drawing of the rhinoceros illustrates the interpene-


tration of gift and commodity exchange in this period. As Europeans sailed
down the coast of Africa, around the Cape of Good Hope, and finally across
the Atlantic, the world changed. In 1601, as Deborah Harkness notes in her
essay here, John Wheeler wrote that "all the world choppeth and changeth,
runneth & raveth after Marts, Markets and Merchandising, so that all things
come into Commerce ... [and] all that a man worketh with his hand or dis-
courseth in his spirit is nothing els but merchandise." Like many early mod-
ern Europeans, Wheeler was acutely aware of the fact that political,
economic, and material changes were also remaking the world of knowl-
edge and culture. How did commodification affect objects and attitudes to
nature in early modern Europe? Traditional accounts of the new world of
wealth in early modern Europe have drawn a stark contrast between the
nobility and the bourgeoisie, seeing them as enmeshed in two different sys-
tems of production and as representing two contrasting types of economic
actors who inhabited divergent cultural environments. Norbert Elias 8 and
Otto Brunner,9 for example, adduced evidence to show how attitudes to
money wealth on the part of the nobility contrasted with that of the bour-
geoisie. Marcel Mauss lo distinguished between the feudal gift economy, per-
meated with notions of natural values, and the burgeoning new commercial
economy, which assigned value to things based on their labor cost, giving
them, in Marxian terms, an "unnatural," "fetishistic" value. In this view,
nobility and bourgeoisie were enmeshed in two separate systems of
exchange-one a world of gifts and favors, and the other a world of cash
and commerce.
These categories have been used particularly in examining the activity of
collecting. For example, Kryzstof Pomian places the "semiophore-man"
(whose objects are "priceless" and bear only mystical significance because
they point to the invisible) at one end of the spectrum and the "thing-man"
(whose objects are worth only their economic value and thus are not price-
less because they reflect only what is visible) at the other end. 11 We might
conclude from these arguments that noble objects were taken out of eco-
nomic circulation and thus did not belong to the same category as the bour-
geois commodities. We might even conclude with Max Weber that the
material accumulation of Calvinist Holland was a completely different
enterprise from the material accumulation of Catholic princes. An initial
examination of the activity of collecting appears to confirm this conclusion.
The Habsburg Kunstkammern of the sixteenth and seventeenth centuries,
for example, were intended to establish the fama of the ruler, representing
his majesty (being shown to the most important "public" at court-visiting

4 Pamela H. Smith and Paula Findlen


nobles and ambassadors-and being displayed at Schauessen, or banquets)
and wealth. 12 As an encyclopedia and memory theater of all nature, art, and
knowledge, the Kunstkammer represented a theater of the world, demon-
strating the ruler's mastery of nature. The cosmic themes of these Kunst-
kammern are clear from the decorative scheme in the Prague Kunstkammer,
for example, which included the seasons, the months, the elements, and the
planets, with Jupiter reigning over all in the center. Like many of the paint-
ings commissioned by the Habsburg, their Kunstkammern can be read as
imperial allegories, representing the ruler's symbolic mastery of the world. I3
Collections in mercantile Holland in the seventeenth century, on the other
hand, like many collections in Italy in the late sixteenth century, were filled
with natural things. A survey of ninety Amsterdam citizens reveals that their
collections were overwhelmingly made up of naturalia, followed closely by
prints and paintings,14 many of which were representations of natural and
artificial objects. These collections grew largely out of the "professional"
activities of doctors, apothecaries, and faculties of medicine at various univer-
sities. By the seventeenth century, however, much of these naturalia were nat-
ural objects turned commodities, such as shells whose price increased in
direct proportion to the rarity of their shape and color. In Amsterdam, exotic
plants were traded, collected, and sold in special shops that carried only East
and West Indian curiosities. Even paintings could be used to buy "real" prop-
erty such as land and houses. Neither curiosities nor paintings existed only in
the realm of pure representation; they also played an active role in the world
of commerce. Everything, it appears, could be bought and sold and given an
equivalent value in relation to other goods in the marketplace.
But if we look more carefully at practices of collecting, we find that the
situation is far less polarized between noble and bourgeois than appears at
first glance. The "commodity collections" of the Dutch burghers did not
possess only commercial significance, and in fact they were inferior to the
princely collections as investments, for they seldom even fetched their calcu-
lated value at auction, 15 while noble collections were far more convertible
into commodity value. Nobles collected the old-fashioned commodities of
gold and silver, and pieces were not infrequently melted down in times of
need, as the famous story of the bronze for Leonardo's equestrian monu-
ment of Francesco Sforza reminds us on an even larger scale. Perhaps the
important thing to keep in mind is the very instability of these categories
since the same object could serve multiple purposes. And both the Dutch
commodities and the noble "potential commodities" (to borrow a phrase
from the very stimulating work of Arjun Appadurai on commodities) could
function as gifts at certain times and as commodities at others. Bourgeois
collections functioned perfectly well within a gift economy,16 and both types
of collections had representational value.

Introduction 5
Furthermore, although much has been made of the practical bent of
Dutch scientific and artistic activity, as of that of their Italian predecessors,
even the Habsburg Kunstkammern in the sixteenth and seventeenth cen-
turies possessed a utilitarian and practical side (and had a number ofleading
northern European artists and naturalists employed on the emperor's behalf
when they were not in places such as Leiden). Collections contained instru-
ments used to survey noble lands; and painters were often sent to record
aspects of forests, villages, and towns in a ruler's domains because they had
special skills in representing their patron's dominion. The gardens, ponds,
menageries, workshops, observatories, and libraries were used by artisans,
artists, humanists, and natural philosophers for models, for the raw materi-
als of knowledge and theories, and as part of an effort to classify and catego-
rize. I7 Moreover, princes frequently showed as much interest in the new
natural philosophy as citizens of republics and patronized some of the most
important natural philosophers of the sixteenth and seventeenth centuries.
Some nobles might have been more interested in display than others, but
Landgrave Moritz of Hesse-Kassel, to name one, was more interested in
arcane knowledge and processes. I8 At the same time, some of the encyclope-
dic collections of the northern Netherlands were not so much "professional"
collections but results of the activities of rich merchants playing the honnete
homme, such as Jan Six and Joachim de Wicquefort, and in certain ways,
Rembrandt van Rijn, and, before him, Albrecht Durer himself. I9 Indeed,
Mark Meadow's essay places the merchant Hans Jacob Fugger squarely
within the genesis of the central European Kunstkammer.
Whether noble or bourgeois, however, the objects in each collection
pointed to their context: the items in Dutch collections represented objects
available to moderately wealthy individuals living in a society enmeshed in a
network of global trade. This fact, however, did not obscure other meanings:
religious, moral, philosophical, and professional meanings intermingled
with the commercial significance of a collection, and the alchemy among
them depended to a large extent upon the intent of the collector. 2o While the
social structure thus left its traces upon the collection, it did not determine
entirely the value or meaning of the object. We must conclude along with
more recent literature that commodification of natural objects was neither a
sudden nor an absolute process, nor one that caused incommensurable dis-
junctions in society and culture. 2I In this vein, John Prow has suggested that
commodification can be both enabling and productive as well as limiting
and destructive. According to his argument,

The commodity form does three things. First, it channels resources of cap-
ital into an area of production in order to expand it to its fullest capacity,
at the same time destroying all productive activities which are not them-

6 Pamela H. Smith and Paula Findlen


selves commodified. Second, it transforms the purpose of production
away from the particular qualities of the thing produced and toward the
generation of profit; production is the indifferent medium for capital val-
orisation, and the qualities of the thing produced are incidental to this
end. Third, it transforms previously or potentially common resources
(both raw materials and final products) into private resources; the alloca-
tion of these resources normally takes place according to economic criteria
(ability to pay rather than moral or civic entitlement), and it may be either
restrictive or expansive in its effects. In the case of most cultural produc-
tion-for example of books, perhaps the oldest of all commodities ...
which would not have come into being without extensive capital invest-
ment-the effects of commodification have been massively expansiveY

Historians could argue that at least the first and the third of his effects of
commodification applied to patronage in early modern Europe. Recent
scholarship on Renaissance consumption, particularly by Richard Goldth-
waite and Lisa Jardine, has begun to document the way in which patronage,
the gift economy, and commodity exchange all contributed to the expansion
of cultural production beginning in the Renaissance.23 The intersection
between patronage and commerce in this period seems to be one key to the
amplification of value and cultural significance that certain types of natural
and artificial objects and modes of representing nature underwent.

VISUALISING NATURE

Durer's rhinoceros also points to a new mode of description, which, even


when not entirely precise, appeared to be the authentic result of a new
emphasis on first-hand observation. Commerce clearly brought to light nov-
elties, and novelty helped occasion new modes of description. The Spanish
civil servant Gonzalo Fernandez de Oviedo (1478- 1557) lamented that he
did not have a famous artist with him to record the things that he described
in his natural history of the Indies in the early sixteenth century: "It needs to
be painted by the hand of a Berruguete or some other excellent painter like
him, or by Leonardo da Vinci or Andrea Mantegna, famous painters whom
I knew in ltaly."24 In so saying, he articulated the importance of art in ren-
dering knowledge visible for an audience fascinated by nature. Words alone
could not adequately represent the marvels of the New World for a Euro-
pean audience. They needed to see these strange and different things in
order to comprehend them fully. Fernandez de Oviedo was not alone in
believing that the skills possessed by Italian Renaissance painters versed in
the new science of painting, as articulated by theorists such as Leon Battista

Introduction 7
Alberti, would convey aspects of the Indies that verbal representation alone
could not capture.
In 1543, the same year that Andreas Vesalius's De Humani Corporis Fab-
rica revolutionized anatomical imagery, Leonard Fuchs published 512
woodcuts of naturalistic plant specimens in De Historia Stirpium. He
defended his departure from ancient practices:

Though the pictures have been prepared with great effort and sweat we
do not know whether in the future they will be damned as useless and of
no importance and whether someone will cite the most insipid authority
of Galen to the effect that no one who wants to describe plants would try
to make pictures of them. But why take up more time? Who in his right
mind would condemn pictures which can communicate information
much more clearly than the words of even the most eloquent men? Those
things that are presented to the eyes and depicted on panels or paper
become fixed more firmly in the mind than those that are described in
bare words. 25

But naturalistic representation, appearing first in northern Italy and then


in Flanders, was more than pictorial description. In the course of the fif-
teenth century, it became a fashion. The particularistic naturalism of Flem-
ish painters, such as Jan van Eyck and Robert Campin, that emerged out of
the forced integration of French court culture and the indigenous urban
artisanal style of the rich trading cities of the Lowlands, spread rapidly
to the noble courts and burgher homes of other parts of Europe. 26 Likewise
the peculiar version of naturalism that we associate with Tuscan art
between the age of Brunelleschi and Leonardo also found its following in
many other parts of Italy in the course of the fifteenth and early sixteenth
centuries, and influenced important northern artists such as DiirerY Natu-
ralistic images and natural objects became sought-after objects of patronage
and commerce among European nobility and wealthy burghers. In many
cases, it is not clear that the object was necessarily preferred over the image.
A series of images created an aesthetically pleasing archive of the world that
the diversity and fragility of actual artifacts never seemed quite able to
capture-which is why "paper museums" such as Ulisse Aldrovandi's col-
lection of thousands of naturalistic drawings and John White's illustrations
of North America continue to be fundamental resources for our under-
standing how early modern Europeans saw and understood their world. 28
For this reason, we should take seriously the role of images in generating a
new sense of the natural world, while at the same time preserving traditional
images of nature that were less easily rendered as objects because they were
acts of imagination.

8 Pamela H. Smith and Paula Findlen


Durer's nature studies, which both continued and extended the Flemish
innovations of the previous century and benefited from parallel develop-
ments in Italy, helped fuel a fashion in the description of nature. In the gen-
eration following Durer, we find "amateur" botanizers and artists who
self-consciously became his followers in cities such as Nuremberg, just as we
can find communities of artists deeply influenced by Leonardo's style in the
vicinity of Milan. They painted insects, small animals, and flowers in imita-
tion of Durer, but interestingly, they also went out into nature to find sub-
jects for their visual descriptions. 29 The passion for a new kind of naturalistic
art had indeed inspired artists, naturalists, and patrons to invest in the
objects they sought to represent. It brings to mind the comment of the
Veronese apothecary Francesco Calzolari, collector of art and nature and a
famous botanist in the late sixteenth century, who told his fellow naturalists
that when they climbed to the top of Monte Baldo near Verona the land-
scape that they would see resembled "a most beautiful Flemish painting."30
In the sixteenth and seventeenth centuries, the things of nature and their
naturalistic representations were collected, studied, sold, and consumed all
over Europe; objects and images claiming to portray and describe nature
became a desired fashion, hard to obtain, and sought by princes and scholars
for their cabinets of curiosities. The correspondence of the Augsburg mer-
chant Philipp Hainhofer (1578- 1647), an agent to several German princes,
provides numerous examples of the passion with which naturalistic objects
and images were sought. He often praised the "lifelikeness" or "naturalness"
of an image or sculpture to his noble patrons. 3l Small animals cast from life
in which "nothing is carved, but all is cast after life [nach dem leben]" were
especially highly prized. 32 In one case, he even called in medical doctors to
attest to the verisimilitude and aesthetic effect of a corpse sculpted in wax. 33
In his correspondence, Hainhofer both responded to and consciously trained
the taste of his patrons (who, as his increasingly frantic attempts to obtain
payment indicate, were simultaneously his clients).34
The fascination with natural goods and artificial objects made after
nature that Hainhofer and others both ministered to and fostered, had the
effect of disseminating images of nature, increasing familiarity with nature,
and training a taste for naturalism. As Hainhofer was selling naturalism to
his northern patrons, Galileo Galilei used naturalistic images in his attempts
to convince his "image-friendly" patrons of his claims that the surface of the
moon was irregular and imperfect (fig. Intro. 2).35 We see here another of the
effects of the fashion for naturalism that is discussed in the papers in this vol-
ume: the formation of a new visual language that would eventually become
an auxiliary tool of proof in the natural sciences.
The natural and artificial objects and their images that were brought to
the fore by the lively trade in early modern Europe constituted a sustained

Introduction 9
&&mm dntm 2
Galileo Galilei, "Moon
/ Surface" preparatory
drawing for Szdereus
Nunnus Ms. Gal. 48,
'. fol. z8r. Courtesy of the
Bibl~otecaNazionale,
Florence.

meditation on the boundary between art and nature and on the human capa-
bility of transcending this boundary.36 This meditation went on in artists'
workshops, natural philosophical cabinets and academies, and in noble col-
lections and studioli. Ultimately, this taste for and interest in nature and nat-
uralism trained the senses. By means of this circulation of naturalistic objects
and images, some people learned to observe, record, compare, and, above all,
value positively such acts of description, comparison, and recording. T h e
patronage and the commodity value of these goods created a climate favor-
able to the investigation and representation of nature and helped to raise the
status of the imitators and knowers of nature. As Claudia Swan has written

Throughout the sixteenth century and well into the seventeenth, observa-
tion, description, and accumulation were the means by which nature came
to be ever more systematically encountered, cataloged, published, col-
lected, and studied. The epistemological objectives sponsored a particular
kind of image, an image done ad v i v ~ r n . ~ '

10 Pamela H. Smith and Paula Findlen


We might simply say that the peculiar confluence of events made realistic
naturalism a model of representation and a measure of the actual objects
that were drawn and sculpted in various media. But the important thing to
bear in mind is the lesson of Durer's rhinoceros: such images were deeply
enmeshed in political and commercial networks that looked at nature in
new ways because it was a tangible sign of one's mastery of the world.

PRODUCING KNOWLEDGE

The turn to naturalism in art reflected a fairly sudden and dramatic transfor-
mation of the ways in which European artists represented their world. Its his-
tory, particularly in northern Italy and Flanders in the 1400s, has often been
recounted in an heroic mode that tends to assume the value of naturalism
without exploring what values it reflected at the time. In discussing the
remarkable naturalism of the illustrations of the manuscript Carrara Herbal of
around 1400 (fig. Intro 3), for example, Otto Pacht spoke of the illustrator's
"courage to turn his back on all patternbooks and to look nature straight in the
face."38 Similarly, Erwin Panofsky commented on the principle, new in the
Renaissance, that a work of art should be a faithful representation of a natural
object, writing, "Treatises on sculpture and painting, therefore, could no
longer be limited to supplying generally accepted patterns and recipes but had
to equip the artist for his individual struggle with reality."39 This heroic tone
might remind us of old narratives of the Scientific Revolution, in which the
protagonists finally looked at nature and saw what was "really" there. More
recent scholarship has reminded us that naturalistic images did not always
imply a theoretical commitment to observing nature for its own sake. The
meditation on the boundary between nature and art sometimes produced
witty counterfeit images, such as Joris Hoefnagel's imaginary insects that
played with viewer's presuppositions about verisimilitude. 40 Just as the history
of art has developed a more sophisticated approach to the idea of naturalism,
likewise the history of science no longer confines itself to a simple story of see-
ing what was "really" there. These heroic narratives, however, both speak to
the ways in which both modern techniques of representing reality and the
advent of a new science traditionally have symbolized the break with the past
that observers have discerned between the "premodern" and the "modern."
The similarity of these two stories and the coincidence of their arrival make us
want to see a connection between a deeper interest in nature that seems to be
manifested in naturalistic art and the investigation of nature by the "new phi-
losophy." What connections can we draw between the emergence of the new
practices vis-a-vis the natural world and the new naturalistic representation?
• • •

Introduction II
( !,.%c,:qono.~r;tn.l c clrli11r.irlcma.p~['.i!tcplrcl rrt 15 f i l m . ( , 11
n4;.1 f l i rhi.li11.1cl:.lc~iS~ I l ~ t ~ ? l o .!in 1x1 h i m ~ * i t o . l i ) ~ l ~ Cb.1
~ l Il ~ l
r.11111llIC1111.C IC fOlC 111~1111.~ IlCxm plrlc I\.C.lnml1 .1111.llr cgurc
4.. 311~~1:.1 FCIICm o l i l 1111.1r l m i Cpma d.rI : C I I C ~ J I . ~ I ~ ~.Icn!n.
.I I
nicnblc cI?i IC f;311!:1F I R f } w c . ~ . u!~.iic i l\~Icf i ~ it~iiclc
c .\LCfore
m l . l - ~ l . l lyllC!?llcl1o:
l L 1.1 I 1111!1 1 ~c1ll.llcC c I?lllOl~
304711 coc:~c Jqcglo.iltmc clxtrCc H(0.c -I.% i ~ i f c ul l c c ti1t1
b f i l i n i n c tili1;.1171y. r ~.,lca.* ~ l ounttr
c iill.1.11X1 fl)rn.l I
r l r CC C'*I.~!?I.I X!?t', cqclclt.1c 1.1 t ~ e l l u l r l l c1!1cllurc
c .ifo*:~a:'c rli I E f:>:ca!at~O(UMIc ~octr.l.rio~n.lu ztr I c f i ~ ~ n:c4
OII,-*~ c :I!CIIO:C c tun!owcr.c ~;*IIX! nuct.1 c y1lr.c I?.lLr r.lilil IIW
Y
v:: P- : c x ~ ~ c~ . Ci L . I ~ U C
\2i(in1 o i i c r & f1.c dr tlcrl Atin u u ~ i . i r n *
ti: crlc ciil'r ir fmc c Ic fulc i l l o h ? CILIPIC~ V I I C ~ CcO71 lfcc CIIOP Y I
~:ctim i ;,i(n. t i l v d ? : i c ( h ~ . > ':. V I <t1711S( L'.I~:c

&$~COZ&/%0..3 Carrara Herbal, Conuoluulus, ca. 1400, Egerton 2020, f. 33.


By permission of the British Library.
If we look again to Durer's rhinoceros and recognize the central role he
played in disseminating its representation along with other naturalistic
images, as well as in establishing a standard and technique of naturalism, we
can begin to understand the central role that artisans in the fifteenth and six-
teenth centuries played in shaping new modes of description of and new
attitudes toward nature. The artisan, after all, was neither wholly an artist
nor a technologist in a modern sense but a person with skills that traversed
these seemingly different domains. The history of many scientific instru-
ments, such as the telescope, the camera obscura, and the microscope, and
their relationship to new techniques of observation indicate similar conclu-
sions. All of these instruments emerged from artisanal and mercantile
milieus. Their application and dissemination also reflected the abilities of
individuals in early modern society to cross boundaries that we have since
installed by creating disciplines in areas which were predominantly defined
by skills and techniques rather than specific kinds of knowledge.
Historians of art and of science have treated the convergence between
artists-artisans and natural philosophers at length. 41 The subject of perspec-
tive, for example, and its mathematical construction has proved to be a locus
classicus for such interdisciplinary discussions. James Ackerman has made
clear the importance of the interaction between craftspeople and scholars in
the matter of scientific illustration; Svetlana Alpers has proposed a more dif-
fuse shared sensibility of descriptive analysis that functioned as a "two-way
street" between which natural philosophers and artisans traveled in the six-
teenth and seventeenth centuries;42 and Thomas DaCosta Kaufmann has
shown in depth the overlapping worlds of humanists and artists in the Holy
Roman Empire. 43 In The Science of Art, Martin Kemp reconstructs a rich
dialogue between artists and naturalists across several centuries, particularly
in the matter of perspective construction and the use of lenses, although he
concludes that there was no parity and no causal link between realist "art"
and realist "science" in the early modern period because science seeks expla-
nation and art seeks to create illusion. 44 Kemp expresses a loosely held con-
sensus among historians of art and science that the investigators of nature
and artists shared affinities, but not more. Recent research suggests that we
should push this argument further. While the opposition between art and
science as two different modes of engaging nature may be true in some sec-
tors today, in the fifteenth and sixteenth centuries, when the methods of the
new science were being constructed, artist-artisans were engaged in a kind
of philosophizing about nature, and, while so engaged, they articulated a
body of claims about nature and about the nature of authority that helped
form the basis of the new science. 45
Panofsky's phrase "the struggle with reality" resonates in this context,
because early modern artists, artisans, and natural philosophers were indeed

Introduction
grappling with the problem of what the world really was and how they
ought to see it, understand it, and represent it. How could they wrest knowl-
edge, sometimes by bodily force, from nature? How could the information
they gained through their senses give them reliable knowledge of nature?
And how could they turn that knowledge to the production of works (of art,
of technology, or of natural knowledge)? Both artisans and natural philoso-
phers needed such knowledge; they needed to learn by the repeated experi-
ence of apprenticeship the characteristics of the material they worked and
thus to produce masterpieces, or, in the case of architects and engineers, to
build fortresses, canals, and weapons, or, in the case of physicians and nat-
ural philosophers, to produce medicaments and knowledge from nature.
Moreover, both groups were engaged in a struggle to make sensory knowl-
edge of nature authoritative; that is, they themselves had to gain social
authority in order to formulate a picture of the rea1. 46 Early modern artists
and naturalists alike required this social authority and self-consciousness in
making the products that resulted from their efforts attractive and persua-
sive in a marketplace of goods and services that had no obligation to privi-
lege their vision of reality. Practitioners of art and science, then, were both
engaged in a quest for new methods of representing nature, as well as repre-
senting themselves as representers of nature. 47
Several essays in the first section of this volume explore the relationship of
artists and artisans to nature, and the way in which their values were
absorbed into the new philosophy of nature. They make the point, however,
that this was not in any way a linear and uncontested process. In the world
of fifteenth-century artisans, artists, and natural philosophers we can see
how much figures such as Leonardo and Durer helped to shape subsequent
discussions of representing and understanding nature without in any way
determining the outcome of the conversation. Moving into the sixteenth
century, in a world of learned naturalists such as Leonhart Fuchs, Conrad
Gessner, and Ulisse Aldrovandi who created communities of artists around
their projects of nature, we see a new set of emerging relations between the
artistic, artisanal, and philosophical communities, as each of these groups
underwent significant transformation. By the seventeenth century, an artist-
naturalist such as Maria Sibylla Merian could travel all the way to Surinam,
following the trade routes that had created this Dutch colony in the Indies,
in order to pursue the insects that she wished to represent. 48 None of the
groups that created the intersections among art, nature, and commerce
enjoyed a well-defined or even very stable role. If artists played a dramatic
role in ushering forth a new image of the world, naturalists took credit for
cataloging the many variations of nature and using the medium of the
printed natural history to combine words and images. Likewise the mer-
chants and princes who understood the economic value of nature and of new

Pamela H. Smith and Paula Findlen


technologies shaped the material environment in which it became possible,
even necessary, to create a new art and science of description. 49

NETWORKS OF KNOWLEDGE: THE MARKET FOR ART,


THE COMMERCE OF SCIENCE

Art historians have argued that the rise of commerce, as well as the com-
merce in paintings themselves, brought about changes in style and subject
matter. Michael Baxandall famously showed the ways in which the "dancing
bankers" of Renaissance Florence changed the face of art. 50 Following some
of Baxandall's fundamental insights, Richard Goldthwaite has argued that
mercantile and patrician demand for certain kinds of art and architecture
played an important role in the development ofItalian Renaissance art. John
Michael Montias has argued that market demand for paintings in the Dutch
Republic led to workshop innovations that brought about greater produc-
tion, such as the move from fine, detailed, smooth work to quick, rough,
painterly techniques in the northern Netherlands in the 1630S,51 and recent
work on the art market in sixteenth-century Antwerp has greatly expanded
our understanding of the way in which works of art became regular items of
international trade. 52 The issues raised by these works have spawned a liter-
ature far too extensive to cover here. 53 Strangely enough, no similar litera-
ture on science and commerce exists. It is instructive to explore why this
might be so.
Max Weber first linked the rise of science to the rise of capitalism early in
this century, regarding such a connection as a by-product of the relationship
between Calvinism and commerce. 54 According to Weber, Calvinism was
one factor in the emergence of a psychological habit that led to an inner-
worldly asceticism that valorized the creation of wealth (as the work of
Providence) and eventually resulted in a new rationality, thus tying both sci-
ence and capitalism to the emergence of a religion of the bourgeoisie. The
most famous result of Weber's link in the anglo phone world was Robert K.
Merton's 1938 monograph, Science, Technology, and Society in Seventeenth-
Century England, which posited a link between Puritanism and the rise of
modern science in the last half of the seventeenth century. Almost two-
thirds of Merton's book dealt with the economic factors that made science a
"socially countenanced, even esteemed" activity in seventeenth-century Eng-
land. 55 The link between Puritanism and science came to be known as "the
Merton thesis," but for Merton, it was not only Puritanism, but also the rise
of capitalist economic activity and social structures that made, as he put it,
the "cultural soil of seventeenth-century England peculiarly fertile for the
growth and spread of science."56 The subsequent debates over the "Merton

Introduction IS
thesis," which occupied so many historians of science, dealt only with the link
between Puritanism and science, and did not explore the link between capi-
talism and science. 57 The reasons for this neglect in anglophone history of sci-
ence could well be the subject of an entire essay, but important factors include
the prevalent influence in the United States and England of Alexandre
Koyn?s conceptualization of the scientific revolution as a revolution in
theory - primarily in astronomy and physics. The professionalization of his-
torians of science that occurred in the United States separated business histo-
rians and historians of technology into separate professional societies and
discouraged greater consideration of technology and science in their common
relation to commerce. No doubt the distaste for examining the social and
material relations of scientific knowledge owed something as well to the Cold
War. 58 The result was that important scholars of the Scientific Revolution
distinguished sharply between the classical, mathematical sciences and the
experimental sciences, or the mechanical arts, which they regarded as tied to
commercial ends and un influential in the Scientific Revolution. 59
This bifurcation has begun to break down as historians of science, influ-
enced by cultural history and its emphasis on practices, have focused on
"practitioners." The old story of the Scientific Revolution was largely the
story of theoretical change, in which the story of the invention of "experi-
ment" was also important but was written as the intellectual history of a
practice. 60 This story left out the large numbers of individuals who began to
show interest in and to practice the "new philosophy," and whose actions
brought about the institutionalization of the new philosophy, and, more
important, made the new method of pursuing knowledge part of the habits
of mind and action of European culture. One of the central components of
the Scientific Revolution is now seen to be the emergence of a whole new set
of beliefs about, and practices involving, nature. The pursuit of natural
knowledge became active and began to involve the body; the investigator of
the natural world had to observe, record, and engage with nature. This new
active element of philosophizing about nature led Francis Bacon to call it a
"New Philosophy, or Active Science." "Active science" was for Bacon an
oxymoron. 61 Furthermore, one of the aims of this new pursuit of the knowl-
edge of nature consisted in the production of effects-tangible objects or
observable phenomena. An important source for this new epistemology was
the entry into the knowledge-making process of a new group of people,
practitioners, drawn from all social strata, but often occupying a middle
ground between university-trained scholars, immersed in texts, and work-
shop-trained artisans, immersed in a world of technique. These practitioners
often saw natural knowledge as an arena in which they could gain new
authority and legitimacy. It brought them the attention of humanists, gave
them access to the republic of letters, and bestowed upon them the favor of

Pamela H. Smith and Paula Findlen


princes, all predicated not upon their birth nor upon their knowledge of elite
learning, but instead upon their ability to undertake particular practices and
produce tangible effects or objects. All these practitioners are relative new-
comers to the historiography of science, or perhaps a shadowy presence that
has always been there but without a clearly delineated role. When we view
them as part of the story of the emergence, or, better, the construction of sci-
ence, the connections between early modern natural philosophy and com-
merce jump out at us with striking clarity. Recent historiography of science
has found natural philosophy in new places, and has advanced the argument
that entrepreneurial doctors, Spanish juntas, foreign artisans in England,
and scholarly merchants all helped shape the habits of mind and action that
became the new science. 62 This volume brings together many examples of
this recent literature and places it next to a similar set of intersections that
have been articulated in the history of art and science.
It is particularly illuminating that for many of these practitioners, atten-
tion to nature and to the market sprang from their interest- in both its
intellectual and financial senses-in medicine. Recent work on natural his-
tory demonstrates the importance of medical practitioners in formulating,
articulating, and disseminating the new philosophy.63 Nor should this sur-
prise us. Medicine always had some manual component to it, and physicians
were always involved with the preparation of medicaments, combining the-
ory and practice in a way that would eventually characterize all natural
philosophical activity. Moreover, apothecaries and herbalists had long
claimed direct access to nature and were among the first to concern them-
selves with the relationship between word, image, and nature in herbals and
other health-related texts. 64
The essays in the second section of this volume show us just how many
individuals in sixteenth- and seventeenth-century Europe were engaged
quite straightforwardly both in commerce and the pursuit of natural knowl-
edge, and how many of their scientific ideas and practices had perfectly clear
and calculable economic value that could be offered on the market. They
make the point that, in some contexts, considerations of use and profit were
the main stimulus to the investigation that yielded new "matters of fact"
from the sixteenth to the late seventeenth century.65
The chapters in this section also show how pursuit of imperial power-
and the desire for domination over nature and peoples and their exploitation
for income-is integral to the construction of a new mode of rationality;
one that stressed eyewitnessing, close observation, group judgment and eval-
uation of information, and the disciplining of subjects. Anthony Pagden has
written very suggestively about the formation of this rationality, making the
point that a variety of strategies were used to give authority to these new
methods. For example, authors created new genres, often modeled on the

Introduction 17
natural history of Pliny, they used a new plain prose style, and they often
employed the legal category of "fact."66 The ancient genre of natural history
turned out to be a remarkably accommodating rubric under which to ascer-
tain the truth of nature in the service of empire. As Chandra Mukerji shows
in this volume, strategies of domination in the seventeenth-century French
kingdom also brought to the fore certain practices in relation to nature.
These essays give us the material to conceive of early modern science in
new ways and to construct a new social and cultural narrative about the
transformation in attitudes to the material world that is sometimes called
"the" Scientific Revolution. The story told in the pages that follow includes
many more people-ranging from indigenous New World artisans and
informants, to Spanish and English bureaucrats, Jesuit missionaries, coun-
terfeiters, and a multitude of others. It brings to light some of the early and
explicit connections between political power and knowledge of nature,
between the universalist claims of science and those of empire, and between
social practices and modes of thought. In contrast to the scholarship that fol-
lowed Merton, the essays in this volume suggest that it is in fact now diffi-
cult to overestimate the importance of the link between science and
commerce.

CONSUMPTION, ART, AND SCIENCE

Fernandez de Oviedo's desire, quoted above, that an artist would supple-


ment his verbal description of the wonders he encountered, is one obvious
link between art, science, and commerce, but a more profound connection
between these three areas can be found in their intersection in the body: in
sensory observation and sensual consumption, in the use of the senses to
know nature, and in the sensory enjoyment of the things of nature and art.
Sensuality, desire, and the passions were intimately linked to commerce, and
the passions played a part in sensory perception. The first of the passions was
curiosity, said by Descartes to be the beginning of knowledge, and it was also
the engine that drove the collection and trade of the objects of nature and
art. 67 One key to the relationship between commerce, art, and science lies
here, in the relation of human bodies and their sensory organs to the mater-
ial world. The essays in the final section of this volume discuss the connec-
tions between consumption and the representation of nature.
Curiosity and the search for profit were insatiable and omnivorous, and
the investigation of nature, the creation of art, and commercial exchange
were all deeply affected by a new valuation of curiosity in the early modern
period. As Lorraine Daston has shown, curiosity moved from an alignment
with lust and pride to one with avarice in the early modern period, and this

18 Pamela H. Smith and Paula Findlen


new sensibility oriented scientific investigation toward commodity-like
objects, as well as into the hidden secrets and small things of nature. Curios-
ity drove collection and consumerism, as well as serious scientific inquiry, as
was attested to by Isaac Newton, Robert Hooke, and Marin Mersenne. 68
Durer too, felt his portrayal of the rhinoceros was compelled by curiosity
and wonder: "Because it is such a wondrous thing, I had to send you a por-
trayal made after it (Das hab jch dir van wunders wegen musen abkunterfet
shicken)."69
Much work remains to be done in writing a history of the body and the
passions in relation to scientific investigation, and the relationship of art, sci-
ence, and commerce is just one of the areas that will be clarified by such a
history. The history that this volume seeks to recapture is not a neatly uni-
fied account of the early modern world, but a series of uneven develop-
ments-fascinating intersections that emerge when different disciplines
discuss the meaning of nature and experience, knowledge and profit, money
and materialism in the early modern period. Each contribution highlights a
different kind of intersection that defines the world of "merchants and mar-
vels" in the late fifteenth through late seventeenth centuries. They remind us
of the importance of careful comparative analysis, lest we be too quick to
offer a global definition of any of these categories. Finally, they reflect a new
geographic sensibility about the study of science and culture in early modern
Europe, one that places as much emphasis on countries such as Spain and the
Netherlands as it does on more traditional sites for the Scientific Revolution,
such as Italy, Germany, and, above all, England. New questions come from
historians' and art historians' interest in new materials. And they also arise
from our interaction with old debates - the incomplete vision of science and
technology offered by Merton, for example, or the suggestive account of
artistic naturalism proposed by Pacht and revised by Baxandall, Alpers, and
Kemp. Let us imagine Durer as he traveled through Europe, carrying the
image of a rhinoceros that never arrived. But of course it did -and that is
the point of the story.

J. Silvio A. Bedini, The Pope's Elephant (New York: Penguin, 1997).


2. T. H . Clarke points out that prior to the rhino's arrival in Lisbon in 1515, only a few
classical scholars and humanists would have been familiar with the animal, from Roman
coins, cameos, and Pliny's account. T. H. Clarke, The Rhinoceros from Durer to Stubbs,
1515-1799 (London and New York: Sotheby's Publications, 1986), 16. It is most likely that
Durer's drawing and woodcut are based upon a sketch that accompanied a newsletter sent to
Nuremberg in 1515 by the Moravian printer Valentin F ernandes, who worked in Lisbon.

Introduction
Shortly after the original drawing was made and certainly before his death in 1528, the first
edition of the woodcut was published. Second and third editions were published in the
1540S- 1550S, and it seems that these editions were the ones widely circulated and icono-
graphically influential to later artists. The drawing and woodcut are characterized by deco-
rative and imaginative patterning upon the separate plates of the rhinoceros's body. In 1515,
Durer lived in the street next to the armorers' quarter (the Schneidgasse) in Nuremberg and
was actively producing designs for armor. Clarke notes that this patterning bears strong
resemblance to Durer's armor designs of the same period and compares it particularly to his
sketch Visor for a Jousting Helm, ca. 1515.
3. This inscription written below the image in Durer's hand is likely a transcription of
Fernandes's newsletter account. John Rowlands, The Age of Durer and Holbein: German
Drawings 1400 - 1550 (London: British Museum Publications, 1988),94.
4. Peter Parshall, "Imago Contrafacta: Images and Facts in the Northern Renaissance,"
Art History 16 (1993): 554-579 .
5. Durer very early on recognized the commercial potential of the new medium of the
printing press, writing, "From now on, I shall concentrate on engraving. Had I done so all the
time I should today be richer by a thousand guilders." Quoted in Erwin Panofsky, The Life
and Art of Albrecht Durer (Princeton, N.J.: Princeton University Press, 1955),44.
6. Martin Kemp, "'The Mark of Truth': Looking and Learning in Some Anatomical
Illustrations from the Renaissance and Eighteenth Century," in Medicine and the Five Senses,
ed. W. F. Bynum and Roy Porter (Cambridge: Cambridge University Press, 1993): 85-121.
7. Anthony Pagden, European Encounters in the New World: From Renaissance to Romanti-
cism (New Haven, Conn.: Yale University Press, 1993).
8. Norbert Elias, The Development of Manners, Vol. I of The Civilizing Process, trans.
Edmund Jephcott (New York: Urizen Books, 1978 [1939]); idem, Power and Civility, Vol. 2 of
The Civilizing Process, trans. Edmund Jephcott (New York: Pantheon Books, 1982 [1939]).
9. Otto Brunner, Adeliges Landleben und europiiischer Geist. Leben und Werk Wolf
Helmhards von Hohberg, 1612-1688 (Salzburg: Otto Muller, 1949).
10. Marcel Mauss, The Gift, trans. W. D. Halls (New York and London: W. W. Norton,
1990 [1950]).
I!. Krzysztof Pomian, Collectors and Curiosities: Paris and Venice , trans. Elizabeth Wiles-

Portier (Cambridge: Polity Press, 1990 [1987]). See also Antoine Schnapper, Le Giant, La
Licorne et la Tulipe: Collections et Collectionneurs dans Ie France du XVIIe Siecle (Paris: Flam-
marion, 1988).
12. Thomas DaCosta Kaufmann has written eloquently on the political and representa-
tional goals of the central European Kunstkammern in The Mastery of Nature: Aspects of Art,
Science, and Humanism in the Renaissance (Princeton, N.J.: Princeton University Press, 1993),
especially the chapter, "From Mastery of the World to Mastery of Nature: The Kunstkam-
mer, Politics, and Science," using the examples of the sixteenth- and early seventeenth cen-
tury Kunstkammern of Emperor Rudolf II (I576- I6I2) and Archduke Ferdinand II of Tyrol
(I5 29-95)·
I3. The collections of Ferdinand II of Tyrol in Schloss Ambras and those of the emperors
in Vienna and Prague, the latter most famously enlarged by Rudolf II, were apparently
arranged according to the material of the objects and grouped vaguely under naturalia (rep-
resenting nature), artificialia (representing human art and the human place in the scheme of
cosmic history as well as the mastery of nature by art), and instrumenta (representing human
knowledge). See, among others, Christian Gries, "Erzherzog Ferdinand II. Von Tirol und
die Sammlungen auf Schlofl Ambras," Fruhneuzeit-Info 5 (I994): 7-37.
I4- Jaap van der Veen, "Dit klain Vertrek bevat een Weereld vol gewoel: Negentig Ams-
terdammers en hun kabinetten," in De wereld binnen handbereik: Nederlandse kunst-en rariteit-

20 Pamela H. Smith and Paula Findlen


enverzamelingen, 1585-1735, ed. Ellinoor Bergvelt, Renee Kistemaker (Zwolle: Waanders
Uitgevers, 1992), 232-258. See also T. H. Lunsingh Scheurleer, "Early Dutch Cabinets of
Curiosities," in The Origins of Museums, ed. Oliver Impey and Arthur Macgregor (Oxford:
Clarendon Press, 1985), II6- 120.
15. As was the case at the auction of the collection of Jan Swammerdam's father.
16. Arjun Appadurai, ed., The Social Life of Things: Commodities in Cultural Perspective
(Cambridge: Cambridge University Press, 1986).
17. Kaufmann, The Mastery ofNature.
18. Bruce Moran, The Alchemical World of the German Court: Occult Philosophy and
Chemical Medicine in the Circle of Moritz of Hessen (1572-1632), Sudhoffs Archiv, Supple-
ment 29 (Stuttgart: Franz Steiner Verlag, 1991), and idem, "German Prince-Practitioners:
Aspects in the Development of Courtly Science, Technology, and Procedures in the Renais-
sance," Technology and Culture 22 (1981): 253-274; Heiner Borgrefe, Vera Lupkes, Hans
Ottomeyer, eds., Moritz de,. Gelehrte. Ein Renaissancefunt in Europa (Eurasburg: Edition
Minerva, 1997)'
19. Dagmar Eichberger, "Naturalia and artefacta: Durer's Nature Drawings and Early
Collecting," in Durer and His Culture, ed. Dagmar Eichberger and Charles Zika (Cambridge:
Cambridge University Press, 1998), 13 -37.
20. See So wijd de wereld strekt (The Hague: Mauritshuis, 1979- 80).
21. See Appadurai, Social Life; Chandra Mukerji, From Graven Images: Patterns of Modern
Materialism (New York: Columbia University Press, 1983); and Craig Clunas, Superfluous
Things: Material Culture and Social Status in Early Modern China (Urbana: University of Illi-
nois Press, 1991).
22. John Frow, "The Signature: Three Arguments about the Commodity Form," in Aes-
thesia and the Economy of the Senses, ed. Helen Grace (Kingswood, N.S.W.: UWS Nepean,
1996), 151 -200, 192- 193.
23. Richard A. Goldthwaite, Wealth and the Demandfor Art in Italy, 1300-1600 (Baltimore:
Johns Hopkins University Press, 1995), and Lisa Jardine, Worldly Goods: A New History of the
Renaissance (New York: W.W. Norton, 1996). See also the AHR Forum on the Renaissance,
The American Historical Review !O3, no. I (1998).
24. Fernandez de Oviedo, Historia General, ii, 7, quoted in J. H. Elliott, The Old World and
the New 1492- 1650 (Cambridge: Cambridge University Press, 1970),21.
25. Leonhart Fuchs, De historia stirpium comentarii insignes .. . accessit iis succincta
admodum difficilium obscurarum passim in hoc opere occurrentium explicatio '" (Paris, 1543),
Preface, pp. x-xi, quoted in James S. Ackerman, "Early Renaissance 'Naturalism' and Scien-
tific Illustration," in The Natural Sciences and the Arts: Aspects of Interaction fi'om the Renais-
sance to the Twentieth Century: An International Symposium, ed. Allan Ellenius (Uppsala:
Almqvist & Wiksell, 1985), 1-17, 17.
26. Wim Blockmans and Walter Prevenier, The Promised Lands: The Low Countries Under
Burgundian Rule, 1369-1530, trans. Elizabeth Fackelman (Philadelphia: University of Penn-
sylvania Press, 1999).
27. The classic study of this subject remains Michael Baxandall, Painting and Experience in
Fifteenth -Century Italy (Oxford: Oxford University Press, 1972).
28. Giuseppe Olmi, L'inventario del mondo. Catalogazione della natura e luoghi del sapere
nella prima eta moderna (Bologna: II Mulino, 1993); and Roger Schlesinger and Arthur P. Sta-
bler, Andre Thevet's North America: A Sixteenth-Century View (Kingston, Ont.: McGill-Queen's
University Press, 1986).
29. Heidrun Ludwig, Nurnberger naturgeschichtliche Malerei im 17. und 18. Jahrhundert
(Marburg an der Lahn: Basilisken-Presse, 1998).
30. Francesco Calzolari, II viaggio di Monte Baldo (Verona, 1565), I I.

Introduction 21
31. A brieflook at the correspondence of Philipp Hainhofer, the Augsburg merchant and
princely factor, confirms this. For excerpts of his correspondence relating to art, see Oscar
Doering, "Des Augburger Patriciers Philipp Hainhofer Beziehungen zum Herzog Philipp II
von Pommern-Stettin. Correspond en zen aus den Jahren I61O-I6I9," Quellenschriften fur
Kunstgeschichte und Kunsttechnik des Mittelalters und del' Neuzeit, NF Bd. 6 (Vienna: Carl
Graeser, I894).
32. Ibid., 79 - 82.
33· Ibid., 91.
34- From I6I7, Hainhofer's demands for payment became urgent.
35. Mario Biagioli, "Picturing and Convincing: The Discovery and Illustration of
Sunspots," unpublished paper read at the Clark Library, Los Angeles, October 2000.
36. Martin Kemp, "Wrought by No Artist's Hand: The Natural, the Artificial, the Exotic,
and the Scientific in Some Artifacts from the Renaissance," in Reframing the Renaissance, ed.
Claire Farago (New Haven, Conn.: Yale University Press, I995), chapter 9, shows the various
and interpenetrating values assigned to objects that teetered on the edge of the nature lart divide.
See also Peter Parshall "Art and Curiosity in Northern Europe," Word and Image I I, no. 4 (1995):
327-33I.
37. Claudia Swan, "Ad vivum, naer het leven, from the life: Defining a Mode of Repre-
sentation," Word and Image, I I, no. 4 (I995), 352-372.
38. Otto Pacht, "Early Italian Nature Studies and the Early Calendar Landscape," Journal
of the Wal'burgand Courtauld Institutes I3 (I950): I3-46, 3I.
39. Panofsky, Life and Art of Durer, 243·
40. In a study of herbal illustrations, however, Sachiko Kusukawa has pointed out that
we should not assume that "naturalistic depiction implies a theoretical commitment to
observing nature for its own sake." "Leonhart Fuchs on the Importance of Pictures," Journal
of the History of Ideas 58 (I997): 403-427, esp. 427. Majorie Lee Hendrix, "Joris Hoefnagel
and the 'Four Elements': A Study in Sixteenth-Century Nature Painting" (Ph.D. diss.,
Princeton University, I984). Thomas DaCosta Kaufmann and Virginia Roehrig Kaufmann,
"The Sanctification of Nature: Observations on the Origins of Trompe L'Oeil in Netherlan-
dish Book Painting of the Fifteenth and Sixteenth Centuries," ]. Paul Getty Museum Journal
I9 (I99 I): 43 - 64.
4I. Historians of science who have treated art and science include Giorgio Santillana,
"The Role of Art in the Scientific Renaissance," in Critical Problems in the History of Science,
ed. Marshall Clagett (Madison: University of Wisconsin Press, I959), 33-65; Samuel Edger-
ton, J r., The Heritage of Giotto's Geometry: Art and Science on the Eve of the Scientific Revolution
(Ithaca, N.Y. and London: Cornell University Press, I99I); J. v. Field, The Invention of Infin-
ity: Mathematics and Art in the Renaissance (Oxford: Oxford University Press, I997). Art histo-
rians include James S. Ackerman, "The Involvement of Artists in Renaissance Science," in
Science and the Arts in the Renaissance, cd. John W. Shirley, F. David Hoeniger (Washington,
D.C.: Folger Shakespeare Library, I985), 94- 129; idem, "Early Renaissance 'Naturalism"';
Svetlana Alpers, The Art of Describing: Dutch Art in the Seventeenth Century (Chicago: Uni-
versity of Chicago Press, I983); Michael Baxandall, The Limewood Sculptors of Renaissance
Germany (New Haven, Conn.: Yale University Press, I980); Thomas DaCosta Kaufmann,
The Mastery of Nature; Martin Kemp, The Science ofArt (New Haven, Conn.: Yale University
Press, I990); David Freedberg, The Power of Images: Studies in the History and Theory of
Response (Chicago: University of Chicago Press, I989); idem, "Science, Commerce, and Art:
Neglected Topics at the Junction of History and Art History," in Art in History, History in Art:
Studies in Seventeenth Century Dutch Culture, ed. David Freedberg and Jan de Vries (Santa
Monica, Calif.: Getty Center for the History of Art and the Humanities,1991); Caroline Jones
and Peter Galison, Picturing Science Producing Art (New York: Routledge, I998).

22 Pamela H. Smith and Paula Findlen


42. "There is a two-way street here between art and natural knowledge. The analogy to
the new experimental science suggests certain things about art and artistic practice, and the
nature of the established tradition of art suggest a certain cultural receptivity necessary for the
acceptance and development of the new science .... Didn't northern viewers find it easier to
trust to what was presented to their eyes in the lens, because they were accustomed to pictures
being a detailed record of the world seen?" Alpers, The Act, 25.
43. Kaufmann, The Mastery of Nature, and Thomas DaCosta Kaufmann, Court, City, and
Cloister: The Art and Culture of Central Europe 1450-1800 (Chicago: University of Chicago
Press, 1995).
44- See the coda to Kemp, The Science ofArt, esp. 340-341. James Ackerman takes a more
developmental view, in which the aims of science and art diverged sometime after Leonardo
da Vinci. James Ackerman, "Science and Visual Art," in Seventeenth Century Science and the
Arts, ed. Hedley Howell Rhys (Princeton, N.J.: Princeton University Press, 1961): 63-90.
45. If we understand art to include other kinds of artisans (as it would have been under-
stood in the early modern period), then there have been numerous studies of the way in
which the new science rested on the work of artisans. This view has its origins in early history
of science, including the essay by Edgar Zilsel, "The Sociological Roots of Science," American
Journal of Sociology 47 (1942): 544-562; and the work of Paolo Rossi, particularly Philosophy,
Technology, and the Arts in the Early Modern Era, trans. Salvator Attanasio (New York: Harper
& Row, 1970); as well as Reijer Hooykaas, "The Rise of Modern Science: When and Why?"
British Journal for the History of Science 20 (1987): 453-473. Arthur Clegg, "Craftsmen and the
Origin of Science," Science and Society 43 (1979): 186-201, holds a similar view. See also Paolo
Rossi, "Hermeticism, Rationality and the Scientific Revolution," in Reason, Experiment, and
Mysticism in the Scientific Revolution, ed. M. L. Righini Bonelli and William R. Shea (New
York: Science History Publications, 1975): 247-273. More recent historians pursuing this
approach are James A. Bennett, "The Mechanics' Philosophy and the Mechanical Philoso-
phy," History of Science 24 (1986): I -28; idem, "The Challenge of Practical Mathematics," in
Science, Culture, and Popular Belief in Renaissance Europe, ed. Stephen Pumfrey, Paolo Rossi,
Maurice Slawinski (Manchester: Manchester University Press, 1991), 176-190; William
Eamon, in Science and the Secrets of Nature: Books of Secrets in Medieval and Early Modern Cul-
ture (Princeton, N.J.: Princeton University Press, 1994); Michael Hunter, Science and Society
in Restoration England (Cambridge: Cambridge University Press, 1983); Pamela O. Long,
"The Contribution of Architectural Writers to a 'Scientific' Outlook in the Fifteenth and Six-
teenth Centuries," Journal of Medieval and Renaissance Studies IS (1985): 265 - 298; and idem,
"Power, Patronage, and the Authorship of Ars," Isis 88 (1997): 1-41; Bruce T. Moran, "Ger-
man Prince-Practitioners: Aspects in the Development of Courtly Science, Technology, and
Procedures in the Renaissance," Technology and Culture 22 (1981): 253-274'
46. Lorraine Daston has warned against invoking a struggle to legitimate oneself by
claiming access to nature, but in the case of artisans, their claim to legitimacy or authority on
the basis of their knowledge of nature was made over and over again and justified by their
ability to produce: to effect works and tangible things rather than words. Lorraine Daston,
"The Nature of Nature in Early Modern Europe," Configurations 6 (1998): 149 - 172.
47. Among works arguing that artists were seeking to legitimate themselves and their art,
see Joseph L. Koerner, The Moment of Self-Portraiture in German Renaissance Art (Chicago:
University of Chicago Press, 1993); Victor I. Stoichita, The Self-Aware Image: An Insight into
Early Modern Meta-Painting, trans. Anne-Marie Glasheen (Cambridge: Cambridge Univer-
sity Press, 1997); Hans Belting and Christiane Kruse, Die Erfindung des Gemiildes: Das erste
Jahrhundert der niederliindischen Malerei (Munich: Hirmer Verlag, 1995). In "Augustan Reali-
ties: Nature's Representatives and Their Cultural Resources in the Early Eighteenth Cen-
tury," in Realism and Representation: Essays in the Problem of Realism in Relation to Science,

Introduction 23
Literature, and Culture, ed. George Levine (Madison: University of Wisconsin Press, 1993):
279-311, Simon Schaffer argues that marginal scientific practitioners in the eighteenth cen-
tury "worked hard to make a direct access to nature count and then make it clear that they
had such access" (296). Such an observation also applies to practitioners and artisans of all
kinds in the sixteenth and seventeenth centuries.
48. Natalie Zemon Davis, Women on the Mmgins: Three Seventeenth-Century Lives (Cam-
bridge, Mass.: Harvard University Press, 1997); Charlotte Jacob-Hanson, Maria Sibylla Mer-
ian, Artist-Naturalist (New York: Brant Publications, 2000).
49. In addition to Alpers's important book The Art of Describing, see Brian Ogilvie, The
Science of Describing: Natural History in the Sixteenth Century (forthcoming).
50. Baxandall, Painting and Experience.
5!. John Michael Montias, Artists and A,·tisans in Delft: A Socia-Economic Study of the Sev-
enteenth Century (Princeton, N.J.: Princeton University Press, 1982); idem, Vermeer and His
Milieu: A Web of Social History (Princeton, N.T.: Princeton University Press, 1989); John
Michael Montias, Gilles Aillaud, and Albert Blankert, Vermeer (Paris: Hazan, 1986). For an
overview, see Michael North, Art and Commerce in the Dutch Golden Age, trans. Catherine
Hill (New Haven, Conn.: Yale University Press, 1997; German ed. 1992).
52. Reindert Falkenburg, Jan de Jong, Dulcia Meijers, Bart Ramakers, Mariet Wester-
mann, eds., Kunst Vaal' de Markt, 1500-1700 (Zwolle: Waanders Uitgevers, 2000).
53. Many valuable contributions to this discussion will appear in The Culture of Exchange:
Real and Imagined Markets in the Low Countries, 1500 - 1800, ed. Liliane Weissberg (in prepa-
ration).
54- Max Weber, The Protestant Ethic and the Spirit of Capitalism, trans. Talcott Parsons
(New York: Harper Collins, 1930).
55. Robert K. Merton, Science, Technology and Society in Seventeenth Century England
(Bruges: St. Catherine Press, 1938), 23!. In the sections he titled "Science, Technology and
Economic Development," Merton explicitly tied the rise of science to the rise of capitalism.
56. Ibid., 238.
57. Similarly, other works of the 1930S and early 1940S that treated the relationship
between capitalism and science-such as Boris Hessen, The Social and Economic Roots of
Newton's Principia (London: Kniga, 193 I); Edgar Zilsel, Die Entstehung des Geniebegriffes; ein
Beitrag zur Ideengeschichte del' Ant1ke und des Friihkapitalismus (Tiibingen: Mohr, 1926); and
Edgar Zilsel, "The Sociological Roots ofScience"- did not lead to any significant research in
this area, at least not in anglophone scholarship. In the late 1950S and early '60S, Paolo Rossi.
published what came to be titled in the 1970 English edition Philosophy, Technology, and the
Arts in the Early Modern Era, which took up many of Zilsel's themes.
58. Michael A. Dennis, "Historiography of Science: An American Perspective," in Science
in the Twentieth Century, ed. John Krige and Dominique Pestre (Amsterdam: Harwood Aca-
demic Publishers, 1997), I -26, gives a very nuanced evaluation of the effects of the particular
circumstances of postwar United States on the growth and character of the profession of the
history of science.
59. In this light it is perhaps not surprising to find the degree to which science was sepa-
rated from material relations, which can be seen in A. R. Hall's "The Scholar and the Crafts-
man in the Scientific Revolution," in Critical Problems in the History of Science, ed. Marshall
Clagett (Madison: University of Wisconsin Press, 1959) and, somewhat more surprisingly, in
Thomas Kuhn's "Mathematical versus Experimental Traditions in the Development of the
Physical Sciences," The Essential Tension (Chicago: University of Chicago Press, 1977)' In this
work, Kuhn even naturalized the difference by claiming that it was "rooted in the nature of
the human mind" (64)'

Pamela H. Smith and Paula Findlen


60. An example of such an orientation is Antonio Perez-Ramos's excellent Francis Bacon's
Idea of Science and the Maker's Knowledge Tradition (Oxford: Clarendon Press, I988).
61. Francis Bacon, The Great Instauration, I620.
62. It should be noted here that in his forgotten chapters, Merton made the relationship
between commerce and science quite clear also, but there is a very significant difference
between his story and the new historiography of this question: the number and range of peo-
ple who are relevant to the story of the scientific revolution have been vastly extended in the
new history of science. Merton, like his contemporaries, told the well-known story of the Sci-
entific Revolution, which recounted the lives and ideas of a few scholars and dwelled long
and lovingly on England in the last half of the seventeenth century.
63. The centrality of medical practitioners in the scientific revolution is suggested by
Harold J. Cook, "The Cutting Edge of a Revolution? Medicine and Natural History Near
the Shores of the North Sea," in Renaissance and Revolution: Humanists, Scholars, Craftsmen,
and Natural Philosophers in Early Modern Europe, ed. J. V. Field and Frank A. J. L. James
(Cambridge: Cambridge University Press, I993), 45-61.
6+ See Otto Pacht, "Early Italian Nature Studies," on early herbals and tacuina sanitatis.
From the Middle Ages, the illustrations in these manuals often showed the herbalist digging
up the plants, or in some other way making clear his personal eyewitnessing or handling of
the plant, even when the herbs themselves were not rendered realistically.
65. Merton made commerce one of the prime legitimators of science in England, but
placed this phenomenon in the late seventeenth century: "The development of scientific soci-
eties was not unrelated to this interest in enlisting the scientist in the service of industry, com-
merce and army. For the rising bourgeoisie, science and technology held out a promise which
was not to be ignored; for the scientist-inventor, economic developments introduced or
emphasized problems which, if attacked and solved, promised some financial reward and
more prestige" (I59).
66. Pagden, European Encounters.
67. Albert o. Hirschman, The Passions and the Interests: Political Arguments for Capitalism
before Its Triumph (Princeton, N.J.: Princeton University Press, I977) was one of the first to
notice the prevalence of the discourse about the passions in early modern Europe. Recent
interest in the passions has yielded Susan James, Passion and Action: The Emotions in Seven-
teenth-Century Philosophy (Oxford: Clarendon Press, I997); and Stephen Gaukroger, The Soft
Underbelly of Reason: The Passions in the Seventeenth Century (New York: Routledge, I998).
68. Lorraine Daston, "Curiosity in Early Modern Science," Word and Image I I, no. 4
(I995): 39I - 404· See also Lorraine Daston and Katharine Park, Wonders and the Order of
Nature (New York: Zone, I999). For one preliminary attempt to write a history of the senses,
see J. R. R. Christie, "The Paracelsian Body," in Paracelsus: The Man and His Reputation, His
Ideas, and Their Transformation, ed. Ole Peter Grell (Leiden: Brill, I998).
69. Translated and quoted in Parshall, "Imago Contrafacta," 561.

Introduction
This page intentionally left blank
STRUGGLING WITH REALITY
Visualizing Nature and Producing Knowledge
This page intentionally left blank
Splendor In the Grass
The Powers of Nature and Art in the Age of Durer

LARRY SILVER AND PAMELA H. SMITH

Sheep and oxen, all of them,


Yea, and the beasts of the field;
The fowl of the air, and the fish of the sea;
Whatsoever passeth through the paths of the seas.
o Lord, our Lord,
How glorious is Thy name in all the earth!
-Psalm 8

O ne of the most striking new images in German art in the era of


Albrecht Diirer remains the Virgin Mary situated out of doors. Its
combination of natural description and spirituality offers a key to
understanding the special yet unarticulated powers associated with nature in
early modern Germany. Investigating the new ways in which Diirer and his
imitators began depicting nature in the period from 1450 to 1550 can provide
an entry point into artisanal attitudes to nature and natural knowledge.
Diirer's Madonna conforms to a late medieval type, associated with the
humus of the earth, known as the "Madonna of Humility.'" Late medieval
visual tradition also situated Mary in a symbolic enclosed garden, or hortus
conclusus, of the Song of Songs (4: 12), redolent of her virginity; a notable
Rhenish example is the name painting of the Master of the Enclosed Gar-
den (Frankfurt, Staedel).2 Striking about the images of the Madonna out-
doors by Diirer and his contemporaries is the absence of these traditional
protective garden walls; these figures sit in wide-open spaces, immersed in
expanse rather than confinement.
Diirer's images of the Virgin outdoors on either the ground or a grassy
bench begin with his engraving Madonna with the Mayfly, ca. 1495 (Meder
/./ Albrecht Diirer, Madonna with a M a y f y , ca. 1495. Courtesy of the
Albertina, Vienna.

42; fig. 1.1)~which features a characteristic local landscape, akin to the


watercolor settings made in the vicinity of Nuremberg, which Durer first
recorded at around the same time.3 Yet erupting into this ordinary setting,
on the central vertical axis with the Madonna and child in this print, is an
extraordinary heavenly epiphany, as God the Father and the dove of the
Holy Spirit appear in open clouds directly over Mary, to complete the Trin-
ity with the infant Christ.
In other, slightly later Durer images of the Madonna outdoors, such as the
woodcut Holy Family with Thlre Hayes (Meder 212) the same grassy bank
recurs, although in front of the extensive, receding landscape setting stands
a stone enclosure, reestablishing the tradition of the hortus conclusus. Here

30 Larry Silver and Pamela H. Smith


the Christ child reads precociously from a holy book while a pair of hover-
ing angels above the Virgin bear her heavenly crown, even as she sits
humbly on her grassy bank. The presence of the hares has never been
explained, but the inevitable association of this most fertile of animals (espe-
cially for a group of them together across the bottom of the print) with
fecundity suggests generation as an important aspect here of the holy family,
although the conception of Christ is expressly not sexual (or worldly or
basely animalistic), as the advanced age of St. Joseph assures. The other most
important element in the print is botanical: the several large weed like plants
on the favorable right (dexter) side of the Virgin. These plants have never
been identified, though both position and size suggest their positive, possibly
medicinal properties. 4 Indeed, the shorter foreground plant with fleshy
leaves resembles the frequent depiction of a mandrake, a plant whose leg-
endary magical and medicinal (chiefly narcotic or poisonous) properties
included a perceived homunculus in its forked root system, making it shriek
when pulled from the ground. s The presence of this plant beside the holy
figures lends a particular charge, related to the conception of Christ, which
can also be related to the passage in Genesis 30:14-24, where eating man-
drake is expressly connected to divine intervention in the pregnancy of
Rachel with Joseph, an Old Testament antitype for Christ: "God hearkened
to her and opened her womb. She conceived and bore a son." (30:22-23).
This combination of mandrake (and grasses, the significance of which
remains to be determined) with the fertility of the hares in the enclosed set-
ting of the Madonna suggests her miraculous virgin birth; at the same time,
the affinity between the natural world and the holy figures in this work joins
with the explicit link between heaven and earth in the Madonna with the
Mayfly to suggest a cosmic resonance between sacred and profane, celestial
and terrestrial, macrocosm and microcosm.
Most commentators on these images of the Madonna outdoors have
assumed that the scene has a foundation in narrative, specifically the story of
the Flight into Egypt, when the holy family fled for its own safety from the
dangers of Herod's Massacre of the Innocents. Yet Durer never seems to try
to represent exotic scenery as the landscape background of these scenes, in
striking contrast to his later Flight into Egypt woodcut from the Life of the
Virgin cycle (ca. 1503-04; Meder 201). In that image, the left-to-right move-
ment of the figures across the setting shows their flight, and two distinctly
exotic non-European trees, palm tree and "dragon tree," mark the location
as "foreign," even close to the lost paradise of Eden.6 Moreover, the presence
of the hovering angels with the crown of Mary in the woodcut offers a hier-
atic theophany through symbolic royal synecdoche, as if in distillation of the
actual vision manifested in the Mayfly engraving. Such scenes are not narra-
tives, but instead provide assertions about the nature of the holy figures,

Splendor in the Grass


especially the incarnate figure of Christ, poised between the world and the
divine order of the cosmos.
As if to confirm this mediating role of Christ and the symbolic presenta-
tion of theological concepts, Durer next (ca. 1497-98) produced a more
mature engraving, the Madonna with the Monkey (Meder 30). Scholars agree
in reading this animal as a symbolic image of human sin, the error of old
Adam, enchained and redeemed by the presence of Christ. 7 This fettered
monkey contrasts with the freedom of the bird in Christ's own hand; wings
associate the creature with both angels and the human soul, and the creature
is actually nourished by the hand of the child. 8 On the grassy bench to the
(favorable) right of the Virgin, a prominent foreground plant, a "star of
Bethlehem," surely signals the location and refers unobtrusively to the event
of the NativityY In this print, no structure of enclosure remains to isolate a
garden from the wider landscape; in contrast, the background landscape
derives closely from a watercolor study from the vicinity of Nuremberg, the
'Weiherhaus' (London, British Museum, W. II5).1IJ This Madonna and child
image, which completely eliminates St. Joseph and any suggestion of the
holy family, could not be misconstrued as a narrative moment.
Climaxing these suggestions that the natural world responds to the char-
acter and virtues of the holy figures, Durer produced a colored ink drawing
in 1503, Madonna with a Multitude of Animals (Vienna, Albertina; fig. 1.2).11
This drawing, a prized cornerstone of the old master collection of Emperor
Rudolf II at the end of the sixteenth century, was copied in both an engrav-
ing (in the same orientation as the original) by Aegidius Sadeler as well as a
painting by Jan Brueghel (1604; Rome, Palazzo Doria)Y This image does
contain background narratives, chiefly the Annunciation to the Shepherds
by the angel, as well as the figure of St. Joseph standing at a small distance
behind the central Madonna and child. Yet fundamentally it is neither a
Nativity nor a narrative scene but rather another image of resonance
between these holy figures and the rich diversity of life in nature, exempli-
fied by the multitude of animals that surround Mary and Christ. Again the
epiphany of heaven finds expression not only in the form of the hovering
angel but also in the form of the star of Bethlehem directly above the holy
figures (occupying the spot of God and the dove in the Mayfly engraving).13
If the size of the main figures as well as their seated or enthroned posture on
the inevitable grassy bank suggests a hierarchy (with Joseph reduced in size
and confined to the middle distance and extreme right), it is noteworthy that
plants and animals fill the immediate foreground, which is marked off as
the holiest space by another "garden" wall, albeit one only as high as the
level of the grassy bench.
Continuity with the Madonna with the Monkey can be seen in the two holy
figures, their postures and gestures, though here Christ gestures off in the

Larry Silver and Pamela H. Smith


humanity, but in effect her benign protection serves to extend grace to the
world, including the animals and plants. In this later redaction the fox at
the feet of Mary, bound with collar and leash, continues the symbolism of
the Madonna with the Monkey of restraint for humanity's animal sinfulness. ls
The parrot, the bird of intelligence and speech, was associated with Mary as
the Throne of Wisdom and traditionally equated with the Annunciation,
even credited with saying "ave."16 Surrounding the Virgin, as if emanating
from her person, are flowers (rose/peony, iris) and fruits (strawberry) that
traditionally embody her virtues. 17 Moreover, the hollyhocks beside the owl
offer a positive antidote to the evil influences of that bird. IS
Durer's imagery of the Madonna outdoors, using animals and other nat-
ural items to demonstrate the resonance between holy figures and God's
creatures, continued in the works of younger imitators. Shortly after the
turn of the sixteenth century, Lucas Cranach painted a I504 image of the
holy family in a forest, which is usually designated as a Rest on the Flight into
Egypt (Berlin) and was followed in turn by a I509 woodcut of the same sub-
ject (fig. 1.3).19 This work is not explicitly a narrative either, though there is
more suggestion of movement. Here the holy figures, including St. Joseph,
pause within a landscape setting of forest or mountain glade in order to rest
upon the ground. The forest setting serves principally as a retreat and haven
for the holy figures instead of being their visible domain (suitable for show-
ing the Madonna enthroned, even if on her grassy bank). More important,
the suggestion of narrative events is underscored by the presence of miracles,
such as the drawing of water by a cluster of small angels. In all likelihood,
the textual source for this event is an apocryphal gospel, Pseudo-Matthew
20, where the child Jesus bids a palm tree to bend low and offer its fruit at
the feet of Mary, while a spring emerges from its roots. In the Cranach
woodcut the tree is bent low by angelic agents, and its spring lies below.
Symbolic plants in the glade at the foot of Mary suggest the influential pres-
ence of her holiness: strawberries, fruits of paradise; cowslip, the plant
known in German as "heaven's key" (Himmelschliissel); columbine, tradi-
tional Marian flower of sorrow, showing the canonical seven sorrows in its
floral shape as well as the mood of melancholy in its purple color.20 Beside
them a thorn serves as contrast and as an anticipation of Christ's Passion. The
finch, associated with thorns, is carried (rather roughly) by one of the angels;
this bird usually symbolizes Christ's triumph over the Passion in the form of
the Resurrection. 21 The presence of this bird in the Cranach scenes could
refer to another apocryphal text (Gospel of Thomas), in which the child Jesus
(age five, according to the text) fashioned birds out of clay (twelve sparrows)
on the Sabbath; these new creatures took flight while he played with many
other childrenY In short, the very landscape in Cranach's Rest on the Flight
into Egypt becomes energized by their divine powers, visibly denoted by the

34 Larry Silver and Pamela H. Smith


rSDizfw,e 1.3 Lucas Cranach, Rest on the Flight into Egypt, 1509.
plethora of angels of various sizes and shapes (possibly of different kinds;
there were nine orders of angels, according to medieval theology). Moreover,
the site manifests the glory of the holy figures through the presence of mean-
ingful plants and animals, already the case with Durer's earlier images of the
Madonna outdoors.
Related to the model of Cranach's images of the holy family in a forest
retreat, a large panel by Altdorfer of Saints John the Evangelist and John the
Baptist (Regensburg, Stadtmuseum; fig. 1.4) features the two holy men in
their roles as hermits, the Baptist on the right in his role as the "voice crying
in the wilderness," the Evangelist on the left as the author of the Book of
Revelation during his sojourn on the island of PatmosY Here, too, as
Behling pointed out,24 the wilderness is crowded with portraitlike represen-
tations of actual plants, highlighted next to the Baptist at right by the large
and magnificent mullein plant, known more colloquially for its shape in
German as "King's candle" (Konigskerze; candela in Latin). This summer-
blooming plant was understood to be a powerful medicament, tied to pow-
erful celestial forces, which would accord with the visionary epiphanies of
these two isolated holy men. Moreover, the plant was a prominent decora-
tion associated with the feast day of the birth of John the Baptist, June 24,
itself a marker of solstice and a popular folk festival often celebrated with
bonfires. 25 Two other prominent plants in the center foreground of the Two
Saints John (fig. 1.4) are sage and henbane (or "deadly nightshade"), the for-
mer a healing herb and digestive, the latter a poison shown just in front of
the Lamb of God). One can recall in this context that the normal attribute of
John the Evangelist is a chalice with a snake, because of a legend that the
saint survived an attempt by a pagan priest to poison him, so the presence of
sage and henbane in front of St. John are appropriate opposites, a deadly
threat and its natural antidote. Animals, tiny but significant, also populate
the Altdorfer image beside the Evangelist: a snail and a butterfly. Tradition-
ally both refer symbolically to Christ's Resurrection (according to medieval
beliefs, snails remained buried in the earth for the three coldest months of
winter and then emerged as the weather warmed Up).26
It should be noted that the outlook associating tiny animals with great,
even cosmic significance endured through the course of the sixteenth cen-
tury. This view is particularly vivid in the presentation of insects within the
overall luxury miniatures by Georg Hoefnagel illustrating the animal world
of the Four Elements (Washington, National Gallery of Art), a work pro-
duced for Emperor Rudolf II around the beginning of the last quarter of the
sixteenth centuryY As Hendrix's fine study makes clear, insects dominate
the volume on the element Fire (Ignis), exemplifying the concept of the
microcosm, or multum in parvo; moreover, the text atop folio I of Ignis testi-
fies to the significance of connections between earth's creatures and their

Larry Silver and Pamela H. Smith


d8&pmi . 4 Albrecht Altdorfer, The Two Saints John, 1515. Courtesy of the Stadtmu-
seum, Regensburg.

heavenly creator: "Of all the miracles made by man, a greater miracle is
man. Of all visible miracles, the greatest is the world. Of those invisible,
God. If we see that the world exists, we believe God exists."28
When Altdorfer painted his 1510Rest on the Flight into Egypt (Berlin; fig.
8), he underscored the visible manifestation of holiness in the landscape
through magical tran~formations.~~ This painting doubtless had strong per-
sonal significance for the painter, since he not only signed it but also dedi-
cated it prayerfully with a dedication to the Virgin on the base of the
miraculous fountain: A[l]b[er]tus Altofleer [plictor Ratis/ponen[sis] In salutem
a[nima]e hoc tibilmunus diva maria samavit/ cordejideli: I510 (Albrecht Alt-
dorfer, painter of Regensburg, dedicated this gift to you, St. Mary, with
faithful heart in (hopes of) the health of his soul).
At first glance, the Altdorfer Rest on the Flight (fig 1.5) offers another
instance of the theme of the holy family beside a riverside landscape akin to
the Diirer engravings, although one can perhaps make an interpretive equa-
tion between the fortified gateway in the background town and the tradi-
tional metaphor of the Virgin as both the "portal of Heaven" (porta coeli) or
the "tower of David" (turris David). Again Mary is seated upon a modest,

Splendor in the Grass 37


&* i.5
Albrecht Altdorfer,
Rest on the Flight to
Egypt, 1510.
Courtesy of Staatliche
Museen Preussischer
Kulturbesitz, Gernalde-
galerie, Dahlem, Berlin.

draped "throne" before or upon a grassy bank. She takes cherries, another
fruit of heaven:' from the aged Joseph. But the striking anomaly within Alt-
dorfer's vision of this setting is an incongruous, dominant foreground foun-
tain along the left side.31This fountain surely evokes recollections-within a
worldly landscape-of paradise as well as traditional symbolism of Mary-as-
fountain, something that Altdorfer reinforced shortly afterward with a
woodcut, Holy Family by a Fountain (ca. 1512115,W. 83).32What makes the
fountain in the painting so unusual is its figure at the top: a pagan statue on a
plinth. But this is not a threatening paganism, for in the lower basin another
lively cluster of playful, cherubic angels climb, swim, and make music, while
the Christ child, unswaddled and naked, leans from Mary's lap over to the
basin to join their play. Moreover, it is at the foot of this astonishing fountain
that Altdorfer appended his prayerful Latin inscription, monogram, and
date, as if it were his own pictorial "gift" (munus) to the Virgin.

38 Lavvy Silver and Pamela H . Smith


The pagan god atop the fountain must be Apollo. Although bearded and
muscular, like standard representations of Hercules, he wears a crown of
laurel, sacred to Apollo. His quiver of arrows is an ambiguous attribute; on
the one hand, Apollo was a renowned archer, who slew Python (perhaps an
appropriate serpent analogue to the Virgin and Christ, conquerors of Satan)
as well as the children of vainglorious Niobe. Yet Hercules, too, was a noted
archer, who slew the Stymphalian Birds with arrows among his celebrated
twelve labors.33 The main figure is accompanied by a blindfolded Cupid,
who holds a bow with two arrows, one sharp and the other blunt. This
secures the Apollo identification (as well as the divine nature of the muscu-
lar bearded figure), because the two contrasting arrows of Cupid refer to the
respective shafts that were shot into Apollo (sharp) and Daphne (blunt),
after the Python incident, as recounted by Ovid (Metamorphoses I, 446-52).
Thus the emphatic presence atop the fountain of Apollo, with his pure (and
unrequited) love for Daphne, and Cupid suggest pagan analogies, or even
typologies, for the divine love principle (including the fruit of cherries) of
the Christian holy family alongside that fountain. At the same time, the
presence of the fountain itself attests to the extraordinary powers of the holy
figures in an otherwise ordinary landscape, akin to the exploits of the Christ
child as recounted in the apocryphal pseudo-Gospels.
Still more mysterious on the fountain is the item held by the Apollo figure.
It appears to be an egg with wings and ducklike feet and its own laurel crown
on top. A similar combination of egglike cup with wings and duck feet (sur-
mounted, however, by a lion rather, than a laurel) forms the subject of an
emblematic Durer drawing of 1513 (Berlin; W. 703), along with the inscribed
motto Fortes Fortuna Juvit (Fortune favors the strong) (fig. 1.6).34 This combi-
nation of lion with egg recalls the presence of lions' heads on the fountain
ornament by Altdorfer in both the Berlin painting and the woodcut, which
suggests some overlap of meaning with the Durer drawing. Lions are often
associated with Christ not only by virtue of being considered the king of
beasts, but also because bestiary lore claimed that lion cubs were born dead
but raised to life after three days when breathed upon by their sire. 35
As Eisler and Hartlaub both observe, Durer also paired lions with the fig-
ure of Apollo (shown with a laurel crown and a bow but beardless) in an early
drawing made during his first stay in Venice, ca. 1495 (Vienna, Albertina; W.
87). In that drawing, Durer also shows a turbaned "oriental" or Turkish
alchemist figure holding a skull as a book sits at his feet. On a small tripod
between the god and the savant sits a smoking globe, labeled "lutus," that is
the lutum sapientiae, the "sealing-wax of wisdom" in alchemy.36 Indeed, the
direct association of alchemy with an exotic, Levantine "magus" strongly sug-
gests that this knowledge is both ancient and occult, a frequent belief of early
modern natural philosophers. 37 Both its shape and association with Apollo

Splendor in the Grass 39


&*am i.6 Albrecht Diirer, Drawing of an Egg Cup with Wings. Courtesy o f
Kupferstichkabinett, Berlin.

suggest that this globe should be thought of as the analogue to what Altdor-
fer produced as an egg on his fountain summit.
Alchemical themes and images were broadly disseminated in manuscript
and in print by the sixteenth century, and both Diirer and Altdorfer could
well have been familiar with alchemy from Latin or vernacular sources. His-
torians have debated the esoteric and alchemical meaning of these artists'
imagery,38but whatever the artists' knowledge of the esoteric aspects of
alchemy, artisanal and alchemical practices ~ v e r l a p p e dIn
. ~ his
~ 1390manual
for painters, Cennino Cennini described certain pigments as being produced

40 Larry Silver and Pamela H . Smith


by "alchemia, "40 and Durer had perhaps learned metallurgical techniques in
the workshop of his father, a goldsmith. In any case, alchemy was more than
a set of theories or techniques. Throughout the early modern period,
alchemy seems to have formed a language for artisans who wanted to artic-
ulate their processes of working, as well as for scholars, who were trying to
understand how artisans created things from matter. The fact that artisans
worked with their hands and learned by apprenticeship separated them
socially and intellectually from most scholars, who regarded the mechanical
arts as existing across a deep divide. This began to change in the sixteenth
century as humanists, such as Petrus Ramus (1515-72) and Juan Luis Vives
(1492-1540), exhorted their fellow scholars not to "be ashamed to enter into
shops and factories, and to ask questions from craftsmen, and to get to know
about the details of their work."41 The late fifteenth and early sixteenth cen-
turies were crucial for the development of a new relationship between schol-
ars and craftspeople. 42 Because alchemy was one of a few disciplines in
which people worked both with texts and with their hands, scholars and
craftsmen alike had practiced it since the Middle Ages, and it played a cen-
tral role in articulating this new relationship.43 Furthermore, alchemy and
artisanry possessed a common essence: the ennobling of matter through
manual work. Although today we associate alchemy mainly with the trans-
mutation of base metals into gold, in the early modern period it was indis-
tinguishable from what we would today call metallurgy or chemistry, or
even organic chemistry, for alchemists sought the composition of the vital
principle through which they could effect transformations of matter of all
kinds, as well as heal disease and prolong life. It thus had much in common
with medicine as well as with artisanship. Indeed, on a cosmic scale, alchemy
was the epitome of all artisanal activity, for it redeemed matter, just as the
practice of the arts (and medicine) did. All such activities resulted from the
Fall and helped to redeem humankind after the expulsion from Eden.
I t is thus not surprising that we would find an interest in alchemy in
Durer's drawings and Altdorfer's paintings, and it is possible to elaborate
their alchemical allusions in terms of alchemical theory. For example, their use
of Apollo, the classical sun god, points to the privileged position in alchemical
thought of solar imagery, which was identified with the metal gold and with
a generative male principle, or fiery spirit, that joined with matter. 44 Sulfur,
one of the two principles of alchemical transmutation, by which base metals
were ennobled, was identified with the sun, gold, and nobility.
Moreover, eggs, placed as central objects by both Durer and Altdorfer,
held special significance in alchemical precepts as the model for the secret,
enclosed generation oflife out of inanimate matter. 45 The "philosophical egg"
of alchemical theory could symbolize a model of the cosmos as well as denote
the source of the philosophers' stone, the material by which alchemists trans-

Splendor in the Grass


muted base metals into gold and silver. In addition, a laboratory vessel hav-
ing a rounded shape and short neck, known as an "egg," or Hermetic vase,
was supposed to be particularly effective in generating the philosophers'
stone. When artists depicted the alchemical process by which the philoso-
phers' stone was produced, it was often within such egglike vessels. The
most fully developed of such depictions, Salomon Trismosin's alchemicalflo-
rilegium, Splendor Solis, produced in Nuremberg in the 1530S, possibly by
Albrecht Glockendon or his workshop, shows the progression from the
black putrefying material in such a vessel, through the multicolored "tail of
the peacock," to the final stage of the transmutation, when the mass in the
"egg" turns red, then is suddenly transformed into gold.
One of the illustrations in the Splendor Solis displays a hermaphrodite
holding an egg in one hand and, in the other, the cosmos, symbolized by the
four elements of earth, water, air, and fire. The egg signifies the beginning
of all things, out of which the entire cosmos develops.46 The figure of the
hermaphrodite represents the dual principles, sulfur and mercury, involved
in alchemical transmutation, which combine to form a unity. Alchemical
writers since at least the thirteenth century regarded sulfur and mercury as
the primary components of all metals, and they continued to do so well into
the seventeenth century.47 While sulfur and mercury had ordinary forms
that could be found in nature and handled by alchemists, in alchemical the-
ory they were understood as principles. Sulfur was viewed as the hot, fiery,
male principle, representing the qualities of fire and air and giving metals
their combustibility.48 It combined with the wet, cold, female principle of
mercury, combining the properties of earth and water. Alchemical writers
identified mercury as the ingredient in metals that gave them their "metal-
lic" qualities, because of its silvery shininess and the fact that it is liquid at
room temperature, analogous to the liquidity of metals when heated. In
addition, mercury forms amalgams, or alloys, with other metals, particularly
gold and silver, in such a way that it is difficult to distinguish the alloy from
the pure, noble metal. Mercury also possesses a solid state, a silvery, fluid
state, and becomes a volatile spirit when heated. It thus appeared to alchem-
ical writers capable of uniting the qualities of matter and spirit, or fixing the
volatile, a common goal of alchemists. 49 Because of its unique qualities,
alchemical authors believed mercury played a central part in transmutation,
and in the late thirteenth century some alchemists were led to reduce the
metallic principles to mercury alone. By the sixteenth century, however,
when the Splendor Solis was illustrated, the two-principle theory had once
again become accepted by most alchemical writers, although the two theo-
ries continued to coexist. 50 The "alchemical wedding" of mercury, the
female principle, with sulfur, the male principle, resulted in an offspring, the
mercury of the philosophers, the main ingredient in the philosophers' stone.

Larry Silver and Pamela H. Smith


a~ y . 7 "Mercury of the Philosophers" from Turba philosophorum, sixteenth
century. Ms. lat. 7171, fol. 16. Courtesy of Ribliothtque Nationale, Paris.

A sixteenth-century illustration in a manuscript version of the Turba


philosophorum shows a naked female figure crowned with a phoenix, stand-
ing on the sun and the moon, wings extending to her feet (fig. 1.7). In the
hands of this "Mercurius genetrix" are representations of the moon and a
chalice out of which shoot flames, perhaps conflating sun, Apollo, and
Christ (by reference to the chalice of the eucharist). The verse above this fig-
ure reads, "Here is born Sun and Moon's child/ the equal of which no one on

Splendor in the Grass 43


earth can find/ but in the world it's nevertheless well known! it's called Mer-
curius philosophorum."51
A seventeenth-century engraving shows the marriage of Sol (sulfur) and
Luna (mercury), from which issues the mercury of the philosophers in a
philosophical egg (fig. 1.8). The flowers springing from the neck of the
"egg" denote the generative qualities of the philosophers' mercury that led
the eleventh-century alchemical author Rhazes, to call it the "water of life."
One of the most famous of all alchemical illustrations, the green lion devour-
ing the sun, out of which runs red blood, signified a similar process. It stood
for the process of raw antimony ore drawing in the universal generative
spirit to produce a vivified mercury, the vital character of which was
denoted by the red blood. This living mercury could dissolve and then reviv-
ify gold to make it grow and multiply. Such a substance was the common
goal of alchemists from Rhazes to Newton. 52
The fountain in Altdorfer's painting also runs with the "water of life."
While the recurring motif in art of the Fountain of Life clearly had a pri-
marily spiritual dimension, ultimately deriving from the source of waters
and flowing rivers described in the Garden of Eden (Genesis 2), it could also
be understood in an entirely material way: alchemical theory posited a vivi-
fying principle concentrated in the philosophical mercury but also dissemi-
nated throughout nature. It fell with the rain and penetrated the earth with
the sun's rays, causing all things to grow and flourish. This view was no
doubt reinforced by interest in Neoplatonism in the fifteenth and sixteenth
centuries, but the concept of a universal spirit already existed in medieval
alchemy and popular pantheism. 53 Thus artists like Leonardo da Vinci
wrote of a "spirit of growth" that permeated all of nature. 54 Even artistic cre-
ativity itself could be understood as a fluid dispensed by the heavens; artist-
author Karel van Mander repeatedly characterized artistic talent as a fluid
endowed by natura generans. 55
Elements of Altdorfer's Rest on the Flight and Durer's works can thus be
read as referring to components of alchemical theory, but they also point to
a more general understanding of nature and function as a means for these
artists to make claims about their understanding of nature and nature's
processes. One such claim that linked artists and alchemists was their com-
mon assertion of "mirroring nature" in their work. 56 From the beginnings
of Flemish naturalism the painters' efforts to render precise and detailed
representations of nature or of reality were viewed by their contemporaries
as a striving after exact replication, not just of nature as it appears to the
human eye, but also of nature as the source of generation and production,
comprising a kind of double imitation of nature. In 1449, Cyriacus of
Ancona described a Flemish painting (probably by Rogier van der Wey-
den) in similar terms:

44 Larry Silver and Pamela H. Smith


f.8 Johann Daniel Mylius, "Coitus"from Anatomiae auri sive tyrocinium
medico-chymicum, 1628. Courtesy of the University of Wisconsin, Madison.
multicolored soldier's cloaks, garments prodigiously enhanced by purple
and gold, blooming meadows, flowers, trees, leafy and shady hills, ornate
halls and porticoes, gold really resembling gold, pearls, precious stones,
and everything else you would think to have been produced, not by the
artifice of human hands but by all-bearing nature herself.5 7

"All-bearing nature" was regarded as prolific and copious, a creative force


that might be imitated by art. Although the idea of art imitating nature or
Creation (ars imitatur naturam) went back to antiquity,58 it could mean more
than just producing a mirrorlike image of created nature. Imitating nature
could mean knowing not just how to mirror nature but could also mean imi-
tating (or reproducing or harnessing) the creative power of nature herself.
The Madonnas in Durer's prints and Altdorfer's paintings can be viewed
not only as alluding to the Fountain of Life but also as signifying the mir-
ror of God (as Jan van Eyck adorns the Virgin in heaven in his Ghent
Altarpiece with a passage from the Book of Wisdom, reading, "She is more
beautiful than the sun and all the order of stars ... a spotless mirror of
God.").59 The natural surroundings in which these two artists place Mary
also alluded to God, for created nature was also seen as the speculum dei.
But for Durer, Nature was also a source of certainty and creativity. In his
late treatises on perspective (I525) and human proportion (I528), he stated
that certainty lay in nature, and that this certainty was expressed through
naturalistic representation:

But life in nature manifests the truth of these things. Therefore observe it
diligently, go by it and do not depart from nature arbitrarily, imagining to
find the better by thyself, for thou wouldst be misled. For verily art [i.e.,
Kunst, or theoretical knowledge, as opposed to Brauch, or simple practice]
is imbedded in nature; he who can extract it has it. 60

Certain knowledge for Durer lay not in theory, as scholars commonly held,
but rather in nature itself, and his naturalism made clear this primacy of
nature as a source of knowledge.
Making this same point, Durer's contemporary, Paracelsus, also articu-
lated an artisanal mode of cognition and a vernacular epistemology. In his
written works, he constantly held up artisans and their knowledge of the
material with which they work as models for gaining natural knowledge
and for creating effects. Alchemy was both the framework for his entire phi-
losophy and the exemplum for all human arts. For Paracelsus, knowledge of
nature was gained not through a process of reasoning, but rather by a union
of the divine powers of mind and of the entire body with the divine spirit in
matter. This he called "experience." In explaining this concept he drew upon

Larry Silver and Pamela H. Smith


the ancient terminology of theory and experience, but he inverted the tradi-
tional understanding of these terms in a remarkable way. In the schema that
organized the pursuit of knowledge from Aristotle up through the sixteenth
century, theory (based on geometric demonstration and syllogistic logic) was
regarded as the sole source of certain knowledge, while experience was
viewed as knowledge of particulars that could never be certain. Paracelsus
inverted this relationship. He defined scientia, or certain knowledge, as the
divine power in natural things, which the physician must "overhear" and
with which he must achieve bodily union in order to gain knowledge of nat-
ural materials out of which to make medicaments. 61 Thus for Paracelsus cer-
tain knowledge was embedded in nature, while experience was the process
by which the physician/natural philosopher united with nature and learned
this science. "Scientia is inherent in a thing .... For instance, the pear tree has
scientia in itself, and we who see its works have experientia of its scientia . ...
Thus in this book, I show the way scientia enters into yoU."62 This is an extra-
ordinary inversion of the concepts of theory and practice, one that Paracel-
sus derived from artisans. In the late fifteenth and early sixteenth centuries,
artisans were experts on the processes and transformations of nature, and
individuals who wished to know (and take possession of) nature looked to
art as the medium through which to accomplish this. This was true not only
for physicians and scholars like Paracelsus but also for princes and city gov-
ernors, who came to regard artisans as holding the key to unlocking the
powers of nature.
Naturalism in artisanal work gained momentum in the North through-
out the sixteenth century, as did the image of an artist as privileged inter-
preter of nature. Durer's example, extended by the next generation of
German artists, produced an extreme form of imitating nature in the tech-
nique of actual casting from life. A master goldsmith from Nuremberg,
Wenzel Jamnitzer (I508-85), was celebrated for this technique (fig. I.9).63
One of Jamnitzer's most famous extant pieces is the Merkel Centerpiece of
I549 (fig. loIO). It features the artist's specialty: grasses and flowers, cast from
life, springing from an egglike vessel at the top, while around the base, rep-
tiles, also cast from life, creep forth from the earth. The central female fig-
ure represents Mother Earth, and the whole piece symbolizes the fertility
and generative powers of nature. It even bears an inscription, "I am the
Earth, mother of all things, beladen with the precious burden of the fruits
which are produced from myself."64 Jamnitzer clearly meant to display his
own powers of creation in this work as well as his ability to imitate nature,
not only by producing an absolute likeness of nature but by harnessing the
processes of metallurgy (alchemy) to create this very representation.
Paracelsus's work also gives insight into the way that nature had, for an
artisan, an immediacy and primacy: nature itself, not words about nature,

Splendor in the Grass 47


&~*LN 1.9 Wenzel Jamnitzer,Lizard,life cast. Courtesy of Staatliche Museen Preussis-
cher Kulturbesitz, Kunstgewerbemuseum, Berlin.

was the certain scientia, which the craftsman got to know through individ-
ual bodily struggle with matter. This awareness emerges from Paracelsus's
idea about a bodily union between the natural philosopher and his object of
as we11 as in his statement that the creative power of humans resem-
bled God's but lay in the human body rather than in the Word. What God
creates through the word, Paracelsus wrote, a human being creates "with his
body and his instruments sense^]."^^

48 Larry Silver and Pamela H . Smith


03~ /./0 Wenzel Jamnitzer, Merkel Centerpiece, l549. Courtesy of the Rijksmu-
seum, Amsterdam.
The bodily aspect of artisanal inte raction with nature emerges with par-
ticular clarity in the work of the ceramicist Bernard Palissy (ca. I5IO-90)
and his 1580 account of his discovery of white glaze. His book as a whole
discusses processes of generation, growth, and change, and is cast as a dia-
logue in which Theory attempts to pry out of Practice the secret of his
enamel-making. 67 Rather than simply giving a recipe to Theory, however,
Practice recounts Palissy's harrowing search for the white glaze, and this
account is remarkable for its Paracelsian type of experience-Palissy's body
and his home are consumed to form a unity with the materials he works. In
the end, his labor redeems him and his household. His ceramic amphibians,
fish, and reptiles show the realist products, also molded from life itself, of
that experience (fig. 1. I I).
A related phenomenon recurs in the account by Benvenuto Cellini
(I5oo-71) of the process of casting his statue Perseus Beheading Medusa: his
body is wasted by fever and illness, the roof of his house catches fire, he
attempts to bring the "corpse" of the metal back to life by throwing all his
pewter utensils into the molten mass. While his body and his house are con-
sumed, he finally falls to his knees in prayer.(,R Contemporaries, including
the painter Bronzino, saw the realistic blood spouting from the head and
torso of Medusa as lifelike, while for Cellini this form mimicked the vivify-
ing force that had sent the life coursing back through both the dead metal as
well as his own veins.
By the late sixteenth century, Palissy and Cellini were themselves respond-
ing to the evolution of the tradition of the artisan as an interpreter of nature
that had begun in the work and writings of figures like Durer and Leonardo,
but their view of the functions and processes of nature shows continuity with
the earlier period. Durer, too, had used similar body language to recount his
own process of creation. He wrote that in the process of looking, the painter
develops his own Augenmass (measure or sense of proportion) in his eyes and
that by much Abmachen (or reproducing nature from life),69 he will fill his
mind full , and thus accumulate a "secret treasure of the heart," from which
he can pour forth in his work, what he "has gathered in from the outside for
a long time." For Durer, the artist was "inwardly full of figures" that he
poured out in new inventions, still part of his "direct and faithful representa-
tion of a natural object."7o This bodily experience of the particulars of nature
resulted for the artist in knowledge of matter and its transformations, which
proved itself through his creation of "effects," or works of art. 71
As the example of Cellini suggests, the concept of bodily knowledge in
the creative process also implicates the bodily labor of the artist's workshop,
including the use of body fluids (saliva, urine, blood) along with the natural
chemicals of painting pigments, clays, stones, and metals. Artists, like other
artisans, were engaged in bodily struggle with and against matter itself, and

50 Larry Silver and Pamela H. Smith


/.// Attributed to Bernard Palissy, Oval Plate. Courtesy of The J. Paul Getty
Museum, Los Angeles.

that matter was not dead, but alive, capable of acting in idiosyncratic fash-
ions, which artisans must come to know-and master-through experience.72
One of the central components of alchemical theory was the concept that
all generation occurred through a process of putrefaction, that is, decay and
regeneration. In putrefaction, particular forms of life appeared first: "ser-
pents, toads, frogs, salamanders, spiders, bees, ants, and many worms,"
according to the formulation of Paracelsus." The reptiles cast from life in
the works of both Jamnitzer and Palissy can thus be seen in a new light, as
can Martin Schongauer's engraving Flight into Egypt, in which lizards or
salamanders creep up the trunk of the Tree of Life.74Salamanders were
believed to be capable of withstanding fire, and thus were a symbol of gen-
eration through putrefaction.75In alchemical theory, the salamander could
also denote sulfur, the hot, fiery, male principle. A return to the creeping
creatures that permeate Diirer's Madonna with a Multitude of Animals could
suggest that these insects and amphibians, too, could signal the beginning of

Splendor in the Grass 51


new life out of the putrefied matter of the old, all under the order and guid-
ance of the divine.
While Altdorfer's Rest on the Flight into Egypt does not contain insects,
lizards, or salamanders, the winged phoenix egg in Apollo's hand signifies
the very same process of regeneration through destruction and putrefaction,
for the phoenix, like the salamander, emerges resurrected out of fire. 76 Like
Jamnitzer's Merkel Centerpiece, Altdorfer's votive picture appears to have
three overlapping themes: the correspondence between spiritual and natural
powers, the transformative power of nature, and the place of art in the great
work of spiritual and material redemption. Transformation itself is indi-
cated by the arrows in Cupid's hands, those same arrows that eventually
caused Daphne to be transformed into a tree. Altdorfer points to the trans-
formational powers of nature in his allusion to the alchemical theory in
which Apollo could be seen as sulfur, Mary as mercury, and the fountain as
the water of life, the generative principle of nature.
At the same time, Mary, mirror of God and fountain of life,77 is the vehi-
cle of the macrocosmic transmutation of the world, the process of redemp-
tion set in motion by Christ's birth, life, and death. In microcosm, the
individual artisan replicates these processes of transformation and redemp-
tion. By giving over his painting to the Virgin, Altdorfer alludes to the role
that the practice of his art plays in redeeming the "health of his soul." He
employs the knowledge of nature, gained by a bodily experience of nature
itself, to imitate nature's creative processes and to transform natural materi-
als into a work of art. The powers of nature-inextricably intertwined with
divine power-are the wellspring of his own artistic creativity, and the
painting as a whole conveys the message that, as an artisan, he is both a
knower and a redeemer of matter.

1. Millard Meiss, Painting in Florence and Siena after the Black Death (New York: Harper

and Row, 1964), 132-156; H. W. van Os, Marias Demut und Verherrlichung in der sienesischen
Malerei 1300-1450 (The Hague: Ministerie Van Cultuur, Recreatie en Maat Schappelijk,
Werk, Staatsuitgeverij, 1969),77-142. While the prototype images of the "Madonna of
Humility" stem from the art of Trecento Siena, this concept spread widely throughout
Europe over the course of the century and was widely taken up, for example by Flemish
painter Robert Campin. On Campin, see Erwin Panofsky, Early Netherlandish Painting (Cam-
bridge, Mass.: Harvard University Press, 1953), 127- 128, 143.
2. Panofsky, Early Netherlandish Painting, 186; on the larger theological implications of
van Eyck's use of the hortus conclusus, Carol Purtle, The Marian Paintings of Jan van Eyck
(Princeton, N .J.: Princeton University Press, 1982), 157-167. More generally, Reindert
Falkenburg, The Fruit of Devotion. Mysticism and the Imagery of Love in Flemish Paintings of

Larry Silver and Pamela H. Smith


the Virgin and Child, 1450-1550 (Amsterdam/Philadelphia: John BenJamins, 1'1'14), esp. 8-21
and references at n. I r; Mirella Levi d' Ancona, The Garden of the Renaissance: Botanical Sym-
bolism in Italian Painting (Florence: L. Olschki, 1977). For the Master of the Frankfurt Par-
adise Garden, see Ewald Vetter, "Das Frankfurter Paradiesgartlein," Heidelbelp;er Jahrbiicher
9 (1965): I02- 14 6; the theme of the enclosed garden in general is discussed by Vetter, Maria
im Rosenhag (Dusseldorf: L. Schwann, I956).
3. For Durer watercolors, still understudied, see Walter Koschatzky, Albrecht Diira The
Landscape Water-Colours (New York: St. Martin's Press, 1973). The banks of open water do
not correspond exactly to extant watercolors but do show elements of the south German sub-
alpine lake region in the distant horizon of the print, while the buildings resemble steeply
pitched wooden structures around Nuremberg's river, the Pegnitz, as shown in later water-
colors, notably the large Mills on a River Bank (Koschatzky, no. 30; Paris, Bibliotheque
Nationale; W. 113). A slightly earlier Durer drawing (Berlin, W. 30) anticipates an image of
the holy family on a grassy bench before a background landscape, albeit a constructed land-
scape of receding trees and distant peaks without open water. Among the elements in favor
of an early date are the artist's undeveloped monogram, perhaps used here for the first time;
in addition, the burin technique and modeling remain comparatively sketchy and irregular.
See Diirer in America: His Graphic Work, exh. cat. (Washington, D.C.: National Gallery of Art,
1971), II2-II3, no. 2, where the date of 1495-96 is suggested, in part because the landscape
elements make it likely that the work was done after the artist's first trip to Italy. For the
identification of the tiny insect as a mayfly and for a general discussion of this print and other
Durer images of the Madonna with animals, see Colin Eisler, Diirer's Animals (Washington,
D.C.: Smithsonian Institution Press, 1991),34-36, who argues that the short life of the mayfly
underscores the brevity of Christ's time on earth. Certainly this particular identification of
species is more plausible than most of the other inscct names (dragonfly, cricket, butterfly,
locust) that have been put forward by previous scholars, despite their differing body types.
4. For plants in general in German art of this period, Lottlise Behling, Die Pjl.anze In dey
mittelalterlichen Tafelmalerei, 2nd ed. (CologneiGraz: Bohlau, 1967); see also Behling, "Betra-
chtungen zu einigen Durer-Pflanzen," Pantheon 23 (1965): 279-291. A rigorous analysis of
Durer watercolors of plants (and imitations thereof) is givcn by Fritz Koreny, Albrecht Durer
und die Tier- und Pjl.anzenstudien der Renaissance (Munich: Prestel-Verlag, 1985), 176-253.
There a close comparison to the several plants in the woodcut print is provided by no. 63, p.
182 (German, beginning of 16th century; Potsdam, Sanssouci, inv. 536b), where the combina-
tion of plants, identified by Koreny as goose-kress, common Ruchgras, and speedwell (Veron-
ica). Of course, Durer painted a celebrated watercolor in 1503, usually known as the Great
Clump of Twf(Vienna, Albertina), discussed by Koreny as no. 61, I76-179, along with some
other botanical representations by the artist. Koreny follows botb Anzelewsky and Gombrich
in suggesting a possible Neoplatonic goal of Durer's immersion in such minutiae of nature,
which might also be implicit in the use of the short-lived mayfly in his earlier engraving, and
in his watercolor studies of plants and animals in general (see below). For a later illuminated
botanical herbal, painted in Holland in the late sixteenth century (today Cracow, Jagiellonian
University Library) for the pharmacist Theodorus Clutius, see Claudia Swan, The Clutius
Botanical Watercolors (New York: Harry N. Abrams, 19'18).
5. Behling, Pjl.anze, 132-136, figs. 19-20,39; a similar plant occupies the center fore-
ground of Pieter Bruegel's Beekeepers drawing (ca. 1568 Berlin), for which see Ethan Matt
Kaveler, Pieter Bruegel. Parables of Order and Enterprise (Cambridge: Cambridge University
Press, 1999), 233-254, but the plant is not discussed; also, Pietcr Bruegel d. ii.'. als Zeichner:
Herkunft Und Nachfolge, exh. cat. (Berlin: Staatliche Museen Preussischer Kulturbesitz,
1975),86-87, no. 100, where the plant is identified as mandrake (German Alraune), "3 rare

Splendor in the Grass 53


plant, which according to superstition grows under the gallows and consequently is called
'Gallow youth' (Galgenjunge) or 'Gallow dwarf' (Galgenmiinnlein)." See also Richard Kieck-
hefer, Magic in the Middle Ages (Cambridge: Cambridge University Press, 1990), 13- 14,66.
6. Gotz Pochat, Der Exotismus wdhrend des A1ittelaltas und der Renaissance (Stockholm:
Almqvist & Wiksell, 1970), II8-136; Robert Koch, "Martin Schongauer's Dragon Tree,"
Print Review 5 (1976) Tribute to Wolfgang Stechow, ed. Walter Strauss, 114-II9· The 1493
Nuremberg Chronicle woodcut of the Garden of Eden shows three trees: a date palm behind
Adam, an apple tree with :make between the first parents, and a dragon tree behind Eve and
next to the fountain oflife, suggesting that it is to be equated with the scriptural Tree of Life.
Bosch also puts the same three trees in his Paradise wing from the triptych of the Garden of
Earthly Delight., (Madrid, Prado); as Pochat recognized, the source t,)r this imagery of animals
is the pilgrimage guidebook Peregrinatio in ten'am sanctam by Bernhard von Breydenbach,
published in Mainz in 1486.
7. For example, Durer in America, 121, no. '4, citing H. W. Janson, Apes and Ape Lore in
the Middle Ages and Renaissance (London: Warburg Institution, University of London, 1952),
15I. Eisler, Durer's Animals, 260-262, discusses monkeys at greater length.
8. Eisler, Dlfrer's AmmalJ, 37, calls the bird a thrush and more explicitly associates its !light
with the Resurrection; see also 56-91 for extended discussion on Durer's birds of various species.
9. This same plant appears in a Leonardo da Vinci drawing (Windsor Castle, no. 12424)
and in his two versions (Paris and rA)ndon) of the Madonna of the Rocks. See Kenneth Clark,
Leonardo da Vinci, rev. ed. (Harmandsworth: Penguin, 1988), 17Y, fig. 72; Martin Kemp,
Leonardo da Vinci (Cambridge, Mass.: Harvard University Press, 1981),273, fig. 74. To my
knowledge this plant has not been identified in the Durer Madonna with the Monkey.
10. Koschatzky, no. 25. This pond setting also resembles the background of the Mayfly

engravmg.
II. Eisler, Diirer's Animals, 31-55, with discussion of two other drawing versions, an ear-
lier one in Berlin (W. 29'1)' the other in Paris (1503; Louvre, W. 297)' Koreny, Albrecht Dlfrer,
, 14- I I 8, with full references.
12. For background on the print and the Durer revival at the court of Rudolf II, see
Dorothy Limouze, "Aegiclius Sadeler," Bulletin, Philaddphlil Museum of Art 85 (spring l(89),
esp. 20 n. 22; for the Jan Brueghel, Breughel-Brueghel, cxh. cat. (Essen, 1997), 169-171, no. 38.
13. For a later manifestation in the art of Albrecht Altdorfer of starbursts, coronas, and
comets in the heavens above religious scenes of the early sixteenth century, see Larry Silver,
"!'\ature and Nature's God: Landscape and Cosmos of Albrecht Altdorfer," Art Bulletin 91
(1999): 194-214- This earlier article suggests some of the arguments about macrocosm/micro-
cosm resonances that are pursued further here.
'4. Michael Baxandall, The Limewood Sculptors of Renaissance Germany (New Haven,
Conn.: Yale University Press, 1980), 165-172.
IS. Eisler, Durer's Animals, 37-39, discusses the fox in this watercolor drawing as repre-
senting "the devil, to be caught by Christ's sacrifIce." The twa nearby owls also arc often asso-
ciated, as birds of night, with evil. On owls in Durer, see Eisler, Durer's Animals, 83-85. On
the owl in the Netherlands tradition, see Paul Vandenbroeck, "Bubo significans 1. Die Eule
als Sinnbild der Schlechtigkeit und Torheit, vor allem in der niederlandischen unci deutschen
Bilddarstellung und bei Hjeronimus Bosch," Jaarboek Koninklljk Museum voor Schone Kun-
.'ten Antwelpen, 1985, 19-136.
T6. L. Nafi:ulin comments on the presenct: of the parrot in van Eyck'sMadol1na with Canon
van del' Paele (Bruges, Grocninge Museum; 1434-36) in his "A Note on the Iconography of
the van der Paele Madonna," Oud Holland 36 (197'): 7; see also Purtle, Marian Paintings, 92.
Discussing the overall significance of the animals in Durer's Madonna with the lvfultitude of
Animals, Eisler, Durer's Animals, 33, cites Francis of Retz's Defensorum virginitatis Mariae, a

54 Larry Silver and Pamela H. Smith


book from the prior generation that used bestiary lore on animals to argue that their behav-
ior heralded both the Immaculate Conception and the miraculous birth of Christ.
17. On flower symbolism in relation to the Virgin, by which her qualities are expressed
through "similitudes" of the flowers to her beauty, purity, sorrow and the like, there is con-
siderable literature. See, for example, Falkenburg, Fruit of Devotion, ro-II; Robert Koch,
"Flower Symbolism in the Portinari Altarpiece," Art Bulletin 46 (1964): 70- 77. Also Eisler,
Durer's Animals, 42-43.
18. Eisler, Durer's Animals, 42-43, notes that "the hollyhock was believed to heal the bite
of poisonous snakes, scorpions, and spiders, and thus, like Christ, to save humankind from
the serpent-Satan."
19. Lucas Cranach, exh. cat. (Berlin, 1973), 13-14, no. I; Hans Mahle, Lucas Cranach d. A.
Ruhe auf der Flucht nach Agypten (Stuttgart: Reclam, 1966); for the plants of the Berlin paint-
ing, see Behling, Pjlanze, 120- 121.
20. Panofsky, Early Netherlandish Painting, 146 n . 6,333; Koch, "Flower Symbolism." On
strawberries as the fruit of paradise, see Behling, Pjlanze, 19; also James Mundy, "Gerard
David's Rest on the Flight into Egypt: Further Additions to Grape Symbolism," Simiolus 12
(1981-82): 213-215; more generally, see Falkenburg, Fruit of Devotion.
21. The association of the finch with Christ's Passion derives in part from the blood-red
mark on the bird's throat; see Herbert Friedmann, The Symbolic Goldfinch (New York: Pan-
theon Books, 1946).
22. M. R. James, The Apocryphal New Testament (Oxford: Clarendon Press, 1975),49; for
the text of Pseudo-Matthew and the miraculous fruit and spring, see 75.
23. Larry Silver, "Forest Primeval: Albrecht Altdorfer and the German Wilderness Land-
scape," Simiolus 13 (1983): 4-43, esp. 30-36; Franz Winzinger, Albrecht Altdorfer. Die Gemalde
(Munich: R. Piper, 1975),84-86, no. 27, dated there ca. 1513-15 but often situated somewhat
earlier. This large work was once, probably originally, located in the monastery of St.
Emmeram in Regensburg, where it is documented in the same century as its creation.
2+ Behling, Pjlanze, 125- 129.
25. On "St. John's fire," in the context ofPieter Bruegel's 1560 Children's Games (Vienna,
Kunsthistorisches Museum), see Sandra Hindman, "Pieter Bruegel's Children's Games,
Folly, and Chance," Art Bulletin 63 (1981): 453 - 454, esp. n. 37. In commemorating a birth-
day and falling almost exactly six months away from Christmas itself, this holiday held spe-
cial calendrical significance. This is the midsummer night alluded to as a lover's holiday by
Shakespeare.
26. See Herbert Friedmann, A Bestiary for St. Jerome (Washington, D.C.: Smithsonian
Institution Press, 1980), 291 -293, for the snail; Eisler, Durer's Animals, 119 (snail as a symbol
of virginity), 127 (butterfly as image of the soul or Resurrection), 133-134 (snail's virginity
tied to belief that air alone impregnated the animal).
27. Marjorie Lee Hendrix, "Joris Hoefnagel and the 'Four Elements': A Study in Six-
teenth-Century Nature Painting," Ph.D. dissertation (Princeton University, 1984); Thomas
DaCosta Kaufmann, Drawings from the Holy Roman Empire [540-[680, exh. cat. (Art
Museum, Princeton University, 1982), 154-157, no. 56.
28. Hendrix,foris Hoefnagel, 215-262, esp. 219 n. 9. She points to the significance of beetles
in the illustrations as examples of nature's own artifice and of the succeeding butterflies, both
inscribed with biblical quotations such as Psalm 145:5 at folio 6: "They shall speak of the mag-
nificence of the glory of thy holiness; and shall tell thy wondrous works." The last folio of Ignis
(fol. 80) quotes Ecclesiasticus 43: "There are many things that are hidden from us that are
greater than these: for we have seen but a few of his works." This is nature as both presentation
of God's manifold glory as well as nature as enigma, shielding divine secrets. Hendrix points
out 257, n. 16, that biblical quotations dominate the epigrams of the Four Elements illustrations:

Splendor in the Grass 55


thirty from Psalms, eight from Ecclesiasticus, eight from Isaiah, six from Proverbs, five from
Job, and three from Ecclesiastes. See also her appendix of inscriptions, 263-332.
29. Winzinger, Albrecht Altdoifer, 7S-77, no. 7·
30. Falkenburg, Fruit of Devotion, esp. 91 -92, citing cherries in particular as "love fruit"
in the convergence of erotic imagery of the Song of Songs with the Marian interpretation of
divine love for humanity.
31. A fountain, of course, was also an important feature of the Garden of Eden, hence for
enclosed gardens evocative of Paradise regained, such as Jan van Eyck's Madonna ofthe Foun-
tain (1439; Antwerp); see Purtle, Marian Paintings, 163-166. Falkenburg, Fruit of Devotion,
10, refers to the Song of Songs' "sealed fountain and/or garden fountain, a well of living

water" (4:12, IS), to be understood as a reference to the Virgin and her chastity. See also
Friedrich Muthmann, Mutter und QueUe: Studien zur ouellenverehrung im Altertum und im
Mittelalter (Basel: Archaologischer Verlag, 1975), esp. 414.
32. Hans Mielke,Albrecht Altdoifer, exh. cat. (Berlin, 1988), 152, no. 73. The rich Renaissance
ornament of this fountain further suggests its wonder, in contrast to the ornate Gothic vocabu-
lary of actual contemporary fountains and the surrounding chapel space of the print. For the
pictorial sources of the fountain and its relation to other Altdorfer works, see Franz Winzinger,
"Der Altdorfer-Brunnen," Jahrbuch del' Berliner Museen 13 (1963): 27-}2. Here as well as in the
Berlin painting the presence of accom panying angels and a pagan deity atop the fountain sug-
gest the supernatural forces at work in its manifestation. Mielke points out a logical fact: that
during the lifetime of the holy figures, only pagan deities could have been publicly presented;
however, as will be discussed below, it is possible to interpret these deities as linked to planetary
gods and thus to the natural forces inherent to both alchemy and astrology in this period.
33. DUrer painted the scene of Hercules Slaying the Stymphalian Birds (Nuremberg, Ger-
manisches Nationalmuseum; on canvas) for the court of Frederick the Wise in Saxony; see
Fedja Anzelewsky, Albrecht Durer. Das malerische Werk (Berlin: Deutscher Verlag fUr Kunst-
wisssenschaft, 1971), 168- 170. For another German painting of a bearded Apollo with bow
and arrow, see the 1530 Lucas Cranach work, Apollo and Diana (Berlin; exh. cat., 1973,21 -22,
no. 10). The nude Altdorfer figure also lacks the most characteristic marks of Hercules: his
club and his lions kin mantle.
34. Rarely analyzed, this drawing does receive attention from Eisler, Durer's Animals,
IS6-IS7, plate 22, who follows G. F. Hartlaub ("Albrecht DUrers 'Aberglaube,'" ZeitschriJi des
deutschen Vereinsfur Kunstwissenschaft 7 [19401, 183-184 n. 23), in asserting that the egg might
well be "an alchemical image of the 'philosopher's egg,' that egg-shaped vessel for thought and
matter." Hartlaub claims that the Altdorfer fountain in the Berlin painting represents the
"bath of Mary" (balneum Mariae), which corresponds to alchemical writings that discuss dis-
tillation as an act of purification, akin to the rite of baptism. There is also an association in
alchemy between virginity and the prima materia. A related drawing, also dated ISI3, Eisler's
plate 36 (Berlin; W. 703) shows a tall bird, a crane (or a heron) with talonlike feet, standing
tall. This bird is associated with vigilance (Eisler, 72, citing Ambrose as well as DUrer's own
marginal illustrations around the same time for the "translated" Hieroglyphica of Horus
Apollo), so it also has a virtuous character, appropriately paired with the strengths of the lion.
35. Eisler, Durer's Animals, 139- 161, discusses lions in general along with their varied
inclusions in DUrer's oeuvre. For the bestiary lore, based on a translation of a twelfth-century
"book of beasts," see T. H. White, The Bestiary: A Book of Beasts (New York: Putnam's, 1960),
7- 1 1, which makes explicit the identification with Christ by analogizing and allegorizing the
qualities of the animal with the mission and nature of the Savior.
36. "Lutum sapientiae" could also refer to the mixture of clay, hair, straw, and horse dung
that was used to seal and protect glass vessels from breakage when they were placed in the
flames of the distilling furnace.

Larry Silver and Pamela H. Smith


37. Richard Kieckhefer, on "Arabic Learning and the Occult Sciences," Magic in the Mid-
dle Ages, II6-150, esp. 144- 150 on the Renaissance magus and the intensified interest in
Hebrew Kabbalah as an additional source for occult knowledge and potential magical incan-
tation. Wayne Shumaker, The Occult Sciences in the Renaissance: A Study of Intellectual Pat-
terns (Berkeley and Los Angeles: University of California Press, 1972), esp. 201-251 on
Hermes Trismegistus and associations of occult learning with ancient Egypt. See also D. P.
Walker, Spiritual and Demonic Magic from Ficino to Campanella (London: Warburg Institute,
1958), esp. II7- 156; Frances Yates, Giordano Bruno and the Hermetic Tradition (Chicago: Uni-
versity of Chicago Press, 1964), esp. 75-96.
38. G. F. Hartlaub, "Albrecht Durer's 'Aberglaube,'" and "Arcana artis. Spuren
alchemistischer Symbolik in der Kunst des 16. Jahrhunderts," Zeitschrifi fiir Kunstgeschichte 6
(1937): 289-324, interpreted Durer's alchemical imagery as the manifestation of the long
enduring workings of the human subconscious. More recently, historians have claimed that
Durer's Melencolia I concerns the first stage of the alchemical process: Maurizio Calvesi, La
melencolnia di Albrecht Diirer (Turin: G. Einaudi, 1993). James Elkins and Didier Kahn have
debated the value of such research: James Elkins, "On the Unimportance of Alchemy in
Western Painting," Konsthistorisk tidskrifi I, no. 1-2 (1992): 21 - 26, and "Reply to Didier
Kahn: What is Alchemical History?" Konsthistorisk tidskrifi 64, no. I (1995): 51-53; Didier
Kahn, "A propos de I'article de James Elkins: On the Unimportance of Alchemy in Western
Painting," Konsthistorisk tidskrifi 64, no. I (1995): 47-51.
39. Cennino D'Andrea Cennini, The Craftsman's Handbook (ll libro dell'Arte), trans.
Daniel V. Thompson Jr. (New York: Dover, 1960), lists the colors made by "alchemy" as ver-
milion, red lead, orpiment yellow, arzica yellow, verdigris, and white lead (24, 25, 28, 30, 33,
34). Other colors fell into the categories of "natural" or "artificial" (but not manufactured by
alchemy). See also A. Wallert, "Alchemy and Medieval Art Technology," in Alchemy Revis-
ited, ed. Z. R. W. M. von Martels (Leiden: Brill, 1990).
40. Jacques van Lennep, Alchemie. Bijdrage tot de geschiedenis van de alchemistische kunst
(Brussels: Gemeentekrediet, 1984) argues that artists' distillation procedures for producing
varnishes were regarded as alchemy. Similarly, Laurinda S. Dixon, "Bosch's Garden of
Delights Triptych: Remnants of a 'Fossil' Science," Art Bulletin 63 (1981): 96- 113; Alchemical
Imagery in Bosch's Garden of Delights (Ann Arbor: UMI Research Press, 198r); and "Bosch's
'St. Anthony Triptych'-An Apothecary's Apotheosis," Art Journal (summer 1984), maintains
that Bosch's familial ties to apothecaries and their procedures of alchemical distillation
account for much of the imagery in Bosch's paintings. While both these authors make their
arguments in general and expansive terms, they do point out an important convergence
between alchemical and artisanal activities, in which material transformation of any sort
could be regarded as alchemical. For an interesting recent meditation on the overlap between
alchemy and artists' practices, see James Elkins, What Painting Is: How to Think about Oil
Painting Using the Language of Alchemy (New York: Routledge, 1999).
41. Juan Luis Vives, De tradendis disciplinis, 1531, trans. Foster Watson (Totowa, N.J.:
Rowman and Littlefield, 1971), 209.
42. For the relationship between scholars and artisans, see Elsbeth Whitney, Paradise
Restored: The Mechanical Arts from Antiquity through the Thirteenth Century, Transactions of
the American Philosophical Society, vol. 80 (Philadelphia, 1990); Paolo Rossi, Philosophy,
Technology, and the Arts in the Early Modern Era, trans. Salvator Attanasio (New York: Harper
& Row, 1970); William Eamon, Science and the Secrets of Nature: Books of Secrets in Medieval
and Early Modern Culture (Princeton, N.J.: Princeton University Press, 1994); Pamela O.
Long, "Power, Patronage, and the Authorship of Ars," Isis 88 (1997),1-41; and R. Hooykaas,
"The Rise of Modern Science: When and Why?" British Journal for the History of Science 20
(19 87): 453-473'

Splendor in the Grass 57


43. On the connections between artisanal epistemology and the later New Philosophy of
the sixteenth and seventeenth centuries, which helped to shape attitudes toward nature and
the pursuit of knowledge about nature in the new science, see Pamela H. Smith, "Artists as
Scientists: Nature and Realism in Early Modern Europe," Endeavour 24 (2000): 13-21; idem,
"Science and Taste: Painting, the Passions, and the New Philosophy in Seventeenth-Century
Leiden," Isis 90 (1999): 420-461.
44. Christ was likewise frequently associated with Apollo or with the "true sun," Sol Justi-
tiae. See also images from the Early Christian period, esp. the Jewish mosaics at Beth Alpha
and Hamat synagogues in modern Israel, which feature Apollo in his solar chariot at the cen-
ter of the zodiac within an alien religious setting (Bezalel Narkiss, "The Jewish Realm," in
Age of Spirituality, ed. Kurt Weitzmann, exh. cat., New York, 1979,374-375, no. 342). For
Apollo and Christ in Early Christian art, see the mosaic in the Mausoleum of the Julii under
St. Peter's basilica in Rome, late third century, where the nimbus around the head of the char-
ioteer Heliosl Apollo transfers divine imaging to the figure of Christ, ascending into heaven
on the chariot and representing the "victorious sun," Sol invictus, for which see Marilyn Stok-
stad, Medieval Art (New York: Harper and Row, 1986), 15-17, fig. 14. Another important
connection between Christ and the image of the sun, specifically by Durer in his engraving
Sol Justitiae (ca. 1500; B. 79), where the judging God rides upon a lion. This almost neglected
print, so redolent of ancient solar iconography transposed into a Christian context, was dis-
cussed by Erwin Panofsky, "Albrecht Durer and Classical Antiquity," Meaning in the Vzsual
Arts (New York: Doubleday, 1955),256-265, also discussing the late antique spirituality tied
to "astral mysticism," which even led to Augustine's warning against carrying the identifica-
tion of Christ with Sol so far as to relapse into paganism (260). Altdorfer's muscular Apollo
on the Berlin fountain also resembles the figure type (again without beard) and pose used by
Durer in his drawings of Apollo (W. 261-64), nude and holding an orb in his left hand, for
example, the famous Apollo with Diana (London, British Museum, W. 261; Panofsky, fig.
76). While Altdorfer's access to such drawings remains unknown, even doubtful, a more pub-
lic, visible touchstone of this figure concept by Durer would have been the engraving Apollo
and Diana (ca. 1502; Meder 64), where the muscular god (beardless) carries a bow and quiver
of arrows and wears a laurel wreath. In this context, it is worth noting the importance of Alt-
dorfer's equation of the Resurrection (1518; Vienna) with the rising sun; see Silver, "Nature
and Nature's God," 199-201. For Durer's equation of the resurrected Christ with an Apol-
lonian sun god, see Panofsky, 260-261 n. 77, fig. 81, citing the Latin poem by Benedictus Che-
lidonius on the reverse of the Small Woodcut Passion Resurrection woodcut, translated as
follows: "This is the day on which the Creator began to make the world, dedicated, accord-
ing to the perennial belief, to the Lord of Heaven and Phoebus. On this day the all-seeing
Sun, affixed to the cross, hidden and dying when the sun set in darkness, splendidly reap-
peared when it rose." Moreover, Durer wrote: "Then, at the same time, as they [the artists of
antiquity] attributed the most beautiful human figure to their idol, Apollo, we would praise
Christ the Lord, who is the most beautiful in all the world. And as they praised Venus as the
most beautiful woman we would ourselves set forth the gracious figure of the most pure
maiden Mary the Mother of god. And for Hercules we would substitute Samson, and we
would do the same with all the other [gods]." Hans Rupprich, Durer's schriftlicher Nachlass, 3
vols. (Berlin: Deutscher Verein fur Kunstwissenschatt, 1956-69),2: 103ff.
45. Eggs also could be specifically associated with the Incarnation of Christ. In particular,
the ostrich egg included by Piero della Francesca above the central holy figures in his
Madonna and Child with Saints (Milan, Brera) has been interpreted as a symbol of supernat-
ural birth. See Millard Meiss, "Ovum Struthionis: Symbol and Allusion in Piero della
Francesca's Montefeltro Altarpiece," The Painter's Choice. Problems in the Interpretation of
Renaissance Art (New York: Harper and Row, 1976), 105- 129. This supernatural role of an

Larry Stlver and Pamela H. Smith


egg's miraculous powers is especially true for an ostrich egg, hatched only by the rays of the
sun according to medieval bestiaries. White, The Bestiary, 121 - 122, speaks of the incubation
in warm sand. This link between the sun and the eggs corresponds to Altdorfer's association
of Apollo and the hatching egg within his hand.
46. H. J. Sheppard, "Egg Symbolism in Alchemy," Ambix 6 (1958): 140-148, 144.
47. See the work of Allen G. Debus, B. J. T. Dobbs, and, briefly, E. J. Holmyard,Alchemy
(New York: Dover, 1990; reprint of 1957 ed.) and John M. Stillman, The Story ofAlchemy and
Early Chemistry (New York: Dover, 1960; reprint of 1924 ed.). In the sixteenth century,
Theophrastus von Hohenheim, called Paracelsus (1493-1541), added a third principle-
salt-to sulfur and mercury, probably reflecting the importance of the production of salts
beginning in the fourteenth century. Gunpowder, saltpeter (from which fertilizer was pro-
duced), and the preservative properties of common salt came to have increasing importance
in the fourteenth and fifteenth centuries.
48. Sulfur appears in a pure form in nature. On heating, it turns a dark red color, which
when rapidly cooled forms a glassy red substance. Similarly mercury is found in a pure form
in nature.
49. For an excellent discussion of the significance of mercury in alchemical theory, see
Karin Figala, s.v. "Quecksilber," Alchemie. Lexikon einer hermetischen Wissenschaft, ed. Claus
Priesner and Karin Figala (Munich: Beck, 1998),295-300; see B. J. T. Dobbs, The Foundations
of Newton's Alchemy (Cambridge: Cambridge University Press, 1975),35-37, for the mat-
ter/spirit problem in alchemy.
50. Figala, "Quecksilber," 298.
5f. Hie ist geboren Solis und Lune kindt! Desgleichen nyemant auf Erden findt! Und in
die weldt doch gern erkhennt! Mercurius philosophorum ist Er gennent.
52. Dobbs, Foundations of Newton's Alchemy, 184-185.
53. In alchemical theory and practice, the conception of nature as actively creating, and of
ars as knowing how to reproduce the natural processes of creation and generation, had long
predated the arrival of the Hermetic corpus. See William Newman, "Technology and
Alchemical Debate in the Late Middle Ages," Isis 80 (1989): 423-445; idem, "Art, Nature, and
Experiment in Alchemy," in Texts and Contexts in Ancient and Medieval Science, ed . Edith
Scylla and Michael McVaugh (Leiden: Brill, 1997), 304 - 317; idem, The Summa peifectionis of
Pseudo-Geber (Leiden: Brill, 1991).
54- Quoted by Jan Bialostocki, "The Renaissance Concept of Nature and Antiquity," The
Renaissance and Mannerism 2 (Princeton, N .J.: Princeton University Press, 1963), 19-30, esp.
25, quoted as an atypical statement of Leonardo, taken from Neoplatonic views that the
world was animated. Although this idea was certainly expressed, particularly by Marsilio
Ficino, much older sources were available to artists and artisans. One author who regards
Leonardo's comments on the vital forces of nature as extremely significant, though he does
not precisely know how to categorize them (he calls them "myth"), is James Ackerman, "Sci-
ence and Art in the Work of Leonardo," in Leonardo's Legacy, ed. C. D. O'Malley (Berkeley
and Los Angeles: University of California Press, 1969), 222ff. See also Martin Kemp,
Leonardo da Vinci. The Marvellous Works of Nature and Man (Cambridge: Harvard University
Press, 1981), esp. 263, 277 on the earth as an organism with geological flux; 312-319 on the
relationship between this geology and the system of the four elements.
55. Karel van Mander, The Lives of the Illustrious Netherlandish and German Painters from
the First Edition of the Schilder-boeck (16°3-°4), trans. and ed. Hessel Miedema (Doornspijk:
Davaco, 1994); see esp. the lives of Lucas van Leyden (I: 104-105, with commentary 3: 2),
Spranger (I : 330- 331; commentary 5: 87), and Aert Mijtens (I: 313; commentary 4: 24)·
56. This goal of mirroring nature was the fundamental ambition of Netherlandish paint-
ing, for which, see Hans Belting and Christiane Kruse, Die Erfindung des Gemiildes. Das erste

Splendor in the Grass 59


Jahrhundert del' niederliindischen Malaei (Munich: Hirmer, 1995), with references; see also the
classic study by Erwin Panofsky, Early Netherlandish Painting (Cambridge: Harvard Univer-
sity Press, 1953), esp. the discussion "Reality and Symbol in Early Flemish Painting,"
13I - 148, and the discussion of the epitome of this ambition, Jan van Eyck: "Thus Jan van
Eyck's style may be said to symbolize that structure of the universe which had emerged, at his
time, from the prolonged discussion of the 'two infinites'; he builds his world out of his pig-
ments as nature builds hers out of primary matter. The paint that renders skin, or fur, or even
the stubble on an imperfectly shaved face seems to assume the very character of what it
depicts; and when he paints those landscapes which, to quote Fazio once more, 'seemed to
extend over fifty miles,' even the most distant objects, however much diminished in size and
subdued in color, retain the same degree of solidity and the same fullness of articulation as do
the very nearest. Jan van Eyck's eye operates as a microscope and as a telescope at the same
time" (181-182). Here it is worth recalling that Jan van Eyck was (falsely) credited as being
the inventor of oil paint technique, surely a form of chemistry/alchemy, for which he was
praised in the following century in the Lives of the Artists by Vasari (1566), who states in the
life of Antonello da Messina that one Jan of Bruges, "a painter very much esteemed for the
practice which he had acquired in becoming a master ... began to try [provare 1different sorts
of colors, especially those which pleased him from alchemy, to make an oil to use as a var-
nish and other things, according to the notions of the learned men of that age." Cited from
the edition of Rosanna Bettarini (Florence: Sansoni, 1971), Vo!' 3,302.
57. Quoted by Panofsky, Early Netherlandish Painting, 2, with full citation.
58. Bialostocki, "Renaissance Concept of Nature and Antiquity"; E. H. Gombrich, "The
Style All' Antica: Imitation and Assimilation," The Renaissance and Mannerism 2: 31 -40;
Ernst Kris, "Georg Hoefnagel und der wissenschaftliche Naturalismus," Festschrift fur Julius
Schlosser zum 60. Geburtstag, ed. Arpad Weixlgartner and Leo Planscig (Zurich: Amalthea,
1927); Otto Pacht, "Early Italian Nature Studies and the Early Calendar Landscape," Journal
of the Warburg and Courtauld Institutes 13 (1950): 13-46.
59. Elisabeth Dhanens, Van Eyck: The Ghent Altalpiece (New York: Viking, 1973), 80. Wis-
dom T 29, 26.
60. From book 3 of his work on human proportions, Vier Bucher von menschlicher Propor-
tion, 1528, trans. and quoted by Erwin Panofsky, The Life and Art of Albrecht Durer (Prince-
ton: Princeton University Press, 1955), 279-280.
61. Walter Pagel, Paracelsus. An Introduction to Philosophical Medicine in the Era of the
Renaissance, 2nd ed. (Basel: Karger, 1982),51.
62. Paracelsus, Das Buch Labyrinthus medicorum genant (1538, 1st pub!. ed. 1553), in Vo!'
I I of Siimtliche Werke, ed. Karl Sud hoff (Munich/Berlin: Oldenbourg, 1928), 163-221, chap.

6, 192. Paracelsus also stated that knowledge (erkantnus) is not in the physician but in nature,
as are the disease and the cure. Das Buch Paragranum (1530), in Vo!' 8 of Siimtliche Werke, ed.
Sudhoff (Munich: Barth, 1924), 140. Pagel, Paracelsus, 51, 218-223, notes the Ncoplatonic
overtones of this passage, but in this and other essays, he also points out the multiple sources
for Paracelsus's work, including popular pantheism, Salomo ibn Gebirol's (I021 - 70) notion
of "prime matter" (229-230), and medieval alchemical treatises (259ff.). See also Walter
Pagel, "Paracelsus and the Neoplatonic and Gnostic Tradition," Ambix 8 (1960): 125- 166.
63. Ernst Kris, "Der Stil ' Rustigue': Die Verwendung des Naturabgusses bei Wenzel
Jamnitzer und Bernard Palissy," Jahrbuch der Kunsthlstonschen Sammlungen in Wien N.F. I
(1928), 137-207, esp. 147; see also Klaus Pechstein, "Der Goldschmied Wenzel Jamnitzer,"
Wenzel Jamnitzer und die Nurnberger Goldschmiedekunst 1500-1700, exh. cat. (Nuremberg:
Germanisches Nationalmuseum, 1985),60,219-223,342-343, nos. 15-16,299, esp. the
Merkel Centerpiece, dated 1549.

60 Larry Silver and Pamela H . Smith


64. Quoted in Wenzel, Jamnitzer, 220, no. 15: "Sum terra Mater Omnium/ Omusta caro
ponderelNascentium ex me fructuum."
65. Although this aspect of the thought of Paracelsus has usually been attributed to Neo-
platonic influences, two recent works have suggested the vernacular nature of his ideas. For
example, Stefan Rhein, '''Mein bart hat mer erfaren dan aile euer hohe schulen'. Ein Zwis-
chenruf zur Quellenfrage bei Paracelus," 99 - 104, and Gundolf Keil, "Mittelalterliche
Konzepte in der Medizin des Paracelsus," 173-193 in Paracelsus. Das Werk- die Rezeption, ed.
Volker Zimmermann (Stuttgart: Steiner, 1995)'
66. Paracelsus, Astronomia magna: oder die gantze Philosophia sagax der grossen und kleinen
Welt/ des von Gott hocherleuchten/ eifahrnen/ und bewerten teutschen Philosophi und Medici (fin-
ished 1537-38; 1st pub!. ed. 1571); in Vo!' 12 of Siimtliche Werke, ed. Karl Sudhoff
(Munich/Berlin: Olden bourg, 1929), 1-444, esp. 56.
67. Bernard Palissy, The Admirable Discourses, 1580, trans. Aurele la Rocque (Urbana:
University of Illinois Press, 1957), 188-203; for an illuminating discussion of this point, see
Neil Kamil, "War, Natural Philosophy and the Metaphysical Foundations of Artisanal
Thought in an American Mid-Atlantic Colony: La Rochelle, New York City, and the South-
western Huguenot Paradigm, 1517-1730," Ph.D. dissertation (Johns Hopkins University,
1988). The most important recent works on Palissy are: Bernard Palissy: My the et Realite
(Niort, Saintes: Musees d' Agen, 1990); Bernard Palissy, Recepte veritable (Geneva: Droz,
1988), with introduction by Keith Cameron; Frank Lestringant, ed ., Barnard Palissy 1510-
1590. [;E;crivain, Le Riforme, Le Ceramiste (Coedition Association Internationale des Amis
d' Agrippa d'Aubigne, 1992); Leonard Amico, Bernard Palissy: In Search of Earthly Paradise
(ParislNew York: Flammarion, 1996); and the recent exhibition catalog, Un 01fivre de Terre:
Bernard Palissy et la Chamique de Saint-Porchaire (Chateau D'Ecouen, 1997)' Also, with an
emphasis on Palissy's Calvinist minority status and architectural and garden designs, Cather-
ine Randall, Building Codes. The Aesthetics of Calvinism in Early Modern Europe (Philadelphia:
University of Pennsylvania Press, 1999),44 - 77. See also Kris, "Stil rustique."
68. Benvenuto Cellini, Autobiography, trans. George Bull (Harmondsworth: Penguin,
1956),343-358, discussed (partly in terms of alchemy) by Michael Cole, "Cellini's Blood," Art
Bulletin 81 (1999),215-235. Here the crucial element of the image is precisely its metallic sub-
stance and its forging in the crucible of a furnace with the addition of "spirits" to resuscitate
the dead metal into flowing, corporeal vitality.
69 While today "Abmachen" means to undo, Panofsky translated Di.irer's use here as
"reproducing nature from life." (Erwin Panofsky, Idea. A Concept in Art Theory, trans. Joseph
J. S. Peake [Columbia: University of South Carolina Press, 1968, 121 - 126.) In Di.irer's usage,
it was probably closer to the modern "abschreiben," or copying verbatim. Di.irer also used the
term "abkunterfet" to mean portraying after life or with verisimilitude. Similarly Paracelsus
used "ablernen" to denote the process oflearning the "scientia" from the plant itself. Paracel-
sus, Labyrinthus, 191.
70. Panofsky, Art and Life of Albrecht Durer, 243 - 244, discussing the artist's "individual
struggle with reality"; Panofsky, Idea, 121 - 126, esp. 123, characterized here as a unique com-
bination in Di.irer of contemporary Italian art theory, influenced by N eoplatonism, and the
artist's own, "personal," here characterized as artisanal, vision of his creative process. This
combination of study from nature and evocation from the m emory and spirit corresponds
fairly closely to what Dutch art theory at the turn of the seventeenth century, exemplified by
Karel van Mander would discuss in the contrasting terms "after life" (naer het leven) and
"from th e spirit/memory" (uyt den geest). See Walter Melion, Shaping the Netherlandish
Canon. Karel van Mander's Schilder-Boeck (Chicago: University of Chicago Press, I991), esp.
63 - 66,243 n. 10.

Splendor in the Grass 6I


71 "Effects" were synonymous with the products of (al)chemical experiments and with
works of art. For example, Lorenzo Ghiberti's called his naturalistic relief panels on the doors
of the Florence baptistery (1425 - 52) the "effeti" of the biblical stories illustrated in each one.
Julius von Schlosser, Lorenzo Ghiberti's Denkwurdigkeiten (I Commentarii), 2 vols. (Berlin:
Julius Bard, 1912), Vol. 1: 49.
72. Paracelsus believed that an artisan must perfect the art of reading a thing from its
external character through experience, thus understanding the virtue and power of the mate-
rial, what he calls "chiromancy": "People who work wood, carpenters, joiners and such, have
to understand their wood by the chiromancy of it, what it is apt and good for." In the same
way a miner will know the chiromancy of the mine. Paracelsus, Liber de imaginibus, vol. 11 of
Siimtliche Werke, ed. Karl Sudhoff (Munich/Berlin: Oldenbourg, 1931), 361 - 386, esp.
375-376. This translation is taken from Baxandall, The Limewood Sculptors, 160, who finds
that Paracelsus "throws a more penetrating light on the characteristic forms of sculpture than
anything else available."
73. Paracelsus, Die neun Buecher De Natura rerum (1537), book 1 in Vol. II of Siimtliche
Werke, ed. Sud hoff, 314.
74. Hartmut Krohm and Jan Nicolaisen, Martin Schongauer. Druckgraphik (Berlin:
Staatliche Museen, 1991),83 - 84, no. 3c (Lehrs 7). For the identification of the Tree of Life as
a derivation of the exotic "dragon tree," see Robert Koch, "Martin Schongauer's Dragon
Tree"; also Gbtz Pochat, Der Exotismus, II8-136.
75. Robert Koch, "The Salamander in Van der Goes' Garden of Eden," Journal of the War-
burg and Courtauld Institutes 28 (1965): 323-326; also White, Bestiary, 182-184; Edward
Topsell, The Elizabethan Zoo (Boston: Nonpareil, 1979), lO8 - II5. Benvenuto Cellini's first
childhood memory was of seeing a salamander in the fire (Autobiography, 20).
76. White, Bestiary, 125- 128; Topsell, Elizabethan Zoo, 22-23; Herbert K essler, "The Soli-
tary Bird in Van der Goes' Garden of Eden," Journal of the Warburg and Courtauld Institutes
28 (1965), 326- 329.
77. See the equation of Mary with the fountain in Jan van Eyck'sMadonna of the Fountain
(1439; Antwerp, Koninklijk Museum), as noted above, n. 31. Purtle, Marian Paintings,
157- 167; Larry Silver, "Fountain and Source: A Rediscovered Eyckian Icon," Pantheon 41
(1983): 95 - 104; Paul Vandenbroeck, Catalogus Schllderkunst 14e-15e eeuw (Antwerp: Konin-
klijk Museum, 1985), 174-178. The fountain as a Marian symbol derives from passages in the
Song of Songs, 4:15, where the beloved is described as "a garden fountain, a well of living
water, and flowing streams from Lebanon."

62 Larry Silver and Pamela H. Smith


Objects of Art/Objects of Nature
Visual Representation and the Investigation of Nature

PAMELA O. LONG

T his essay explores some of the ways in which visual representation


came to legitimate knowledge claims in the period between about
1490 and the 1540s. The use of images to further knowledge about
the natural world represents a significant cultural development. For both
philosophical and historical reasons, ancient and medieval thought tended
to separate things made by human artifice - paintings and drawings, as
well as constructed objects such as machines-from the natural world. In
part, this reluctance to use visual images to demonstrate claims about the
world came out of the Aristotelian view that art and nature were opposed
to one another. 1
Alan Gabbey frames the issue as it concerns mechanics. He points to the
traditional separation between the mechanical arts and Aristotelian physics.
For Aristotle, machines and devices rearrange things against nature and for
human ends; they belong to mechanics as a practical art. The theoretical
discipline of physics did not include an interest in mechanical devices or
their specific effects in the world. Physics "did not share the concerns of
mechanics, since physics did not deal with artificial things qua artificial."
Traditional Aristotelian definitions of physica, philosophia naturalis, or phys-
iologia concern "the science [scientia] of natural bodies in so far as they are
natural." Gabbey investigates the ways in which mechanical arts became
mechanics, a discipline in which the ancient distinctions between the
mechanical arts and physica or natural philosophy came to be seriously
eroded or eliminated. Thereby could seventeenth-century experimental
philosophers use machines and devices to demonstrate claims about the nat-
ural world. 2
Yet the transformation involving mechanical arts and mechanics is just
one episode in a larger cultural development in which the artifactual and
the natural moved into closer proximity. The use of visual images (created
by human artifice) to represent observed natural objects had to overcome
serious impediments, including a focus on the capacity of drawings and
paintings to create optical illusions. In ancient times, as the story of Zeuxis
attests, the most skillful painter is the one whose painting is radically decep-
tive. The story as told by Pliny recounts the rivalry of Zeuxis and Parrhasius.
Zeuxis, attemping to demonstrate his superiority as a painter, produced a
picture of grapes so expertly that birds flew up to it. Then Parrhasius created
a picture of a curtain so realistic that Zeuxis, "proud of the verdict of the
birds, requested that the curtain should now be drawn and the picture dis-
played." When he discovered that the curtain was actually a painting by Par-
rhasius he honorably yielded the prize to him. 3 The story emphasizes the
remarkable skill and ingenuity of the painters, but it hardly offers visual
representation as a modality for demonstrating truths about the natural
world. It underscores instead the capacity of humans both for creating opti-
cal illusions and for being deceived by them.
The inherent capacity of visual images to deceive made them suspect as
aids to discovering truths about nature. This suspicion was augmented by
the sharp medieval social divisions between those who studied natural phi-
losophy at the universities and those who produced images in workshops.
Scholastic natural philosophy was not fundamentally experimental, observa-
tional, or visual, but was logical and textual involving extensive commen-
taries, lectures, and formal disputations based on the study of authoritative
writings. Artisanal practitioners did not attend universities, but were trained
by apprenticeship in workshops. The mechanical arts and natural philoso-
phy were separated physically and socially, as well as conceptually.4
Yet gradually during the fifteenth and sixteenth centuries in settings out-
side of the universities in cities and courts, humanists and others combined
an interest in learning with a preoccupation with the constructive arts. The
basic context for such a development involves medieval commercial capital-
ism and urbanism in which the production and exchange of all kinds of
goods was aided by the expansion of both local and distant markets. Worldly
goods proliferated just as the cultural value of material objects increased.
Elite rulers undertook massive building programs to reorganize the urban
spaces that they controlled and to legitimate their own power. From the late-
fifteen century, they increasingly practiced conspicuous consumption. Finely
wrought objects, including paintings, sculpture, and other forms of visual
representation gained cultural value.'
At the same time, learned culture expanded beyond the universities as
humanists took positions as secretaries in cities and courts, or found patron-
age from oligarchs and princes who also employed workshop-trained
painters, architects, and engineers. Across Europe, in many courts and cities,
learned humanists, wealthy patrons, and artisan-trained painters, sculptors,

Pamela O. Long
and architects lived in closer proximity-both physically and in terms of
their mutual interests - than they had in previous centuries. Artisan-
trained individuals sometimes wrote treatises about their crafts. Learned
humanists such as Leon Battista Alberti also wrote treatises and commen-
taries on painting, architecture, and other constructive arts. 6 These broad
cultural developments took place in the context of a thriving commercial
culture involving the production and exchange of everyday and luxury
goods within both local and European-wide markets. One result was that
objects fabricated by humans, including visual images, came to be used for
the study of the natural world.
This essay examines a group of diverse sources in an investigation of the
ways in which art and nature came to be closely associated or even inter-
changeable between the 1490S and the 1540s. Sources include a lengthy
architectural romance, the Hypnerotomachia Poliphili [Struggle of Poliphilus
in a dream], a literary work published in 1499; Leonardo da Vinci's Madrid
Codex I, an extensively illustrated manuscript treatise on machines and
mechanics created in the 1490S; some of the paintings and writings of
Albrecht Durer; the treatise on the architectural orders by Sebastiano Serlio,
published in 1537; and finally, Andreas Vesalius's famous anatomical trea-
tise, the De Humani Corporis Fabrica, published in 1543. 7
These sources are often placed in separate disciplinary categories. The
Hypnerotomachia Poliphili is a literary work of interest to architectural and
literary historians. Madrid Codex I falls into the category of mechanics and is
of interest to historians of science. Durer's paintings and prints are studied
primarily by art historians. Serlio is studied by architectural historians,
whereas Vesalius draws the attention of historians of anatomy, medicine, and
science. Yet these modern disciplinary categories are at least in part anachro-
nistic. They tend to separate these works one from another in ways that are
inappropriate for the sixteenth century, obscuring their common culture.
The sources considered here share several characteristics. In diverse ways,
for example, they display close relationships, and even interchangeability
between artificial and natural objects. This interchangeability is particularly
evident in the Hypnerotomachia Poliphili, and it is explicit or implicit in the
other sources as well. Analogies between natural and artificial objects, and
the use of one to demonstrate a truth about the other, are evident in most of
them. Leonardo, for example, uses the motions of actual and visually
depicted machines to explore principles of motion. Durer creates the first
free-standing images of landscapes and creates other "slices" of the natural
world by means of art. Vesalius in his anatomical treatise uses methods strik-
ingly similar to those that Serlio used in his earlier treatise on architecture.
He also compares anatomical structures such as bones to artificial structures
such as the beams of houses. While the differences among these works are

Objects of Art/Objects of Nature


apparent, they share a tendency to closely associate objects of art and objects
of nature.

ART AND NATURE IN THE HYPNEROTOMACHIA POLIPHILI

The Hypnerotomachia PoIiphiIi, published in 1499, was written possibly by


one Francesco Colonna, who probably can be identified as a Dominican
monk of Venice of the same name. The authorship of the work is contro-
versial. Whoever Francesco Colonna was, his complex romance encom-
passes a dream within a dream in which Poliphilo walks through a varied
terrain filled with pyramids, ruins, gardens, forests, meadows, and streams,
as he searches for his lost love, Polia. s
For my purposes here, Hypnerotomachia Poliphili is interesting because of
the author's persistent interest in itemizing objects of nature and objects of
art, including architectural fragments, in profuse detail, and also because of
his habit of interchanging constructed artifacts and natural things. For exam-
ple, at one point in his dream within a dream, Poliphilo comes to an enclosed
valley where he encounters an enormous half-ruined pyramid carved out of
the surrounding mountains and topped by an obelisk that is dedicated to the
sun. The author provides a detailed description of the monument itself
including structures, ornamental details, fragments of sculpture and architec-
ture, inscriptions, and the names of many varieties of herbs and grasses grow-
ing among the ruins. Poliphilo hears frightening groans and discovers that
they are emitted by the colossal statue of a man. The groans are caused by
wind blowing through the open mouth of the colossus. Using the hairs of the
chest and beard of the statue, Poliphilo pulls himself into the open mouth and
then climbs through the viscera. Each part of the body, "intestines, nerves,
bones, veins, muscles, and flesh" is present, and each is labeled in Chaldean,
Greek, and Latin. The inscribed body parts describe what sickness is gener-
ated in each part, the cause, the required care, and the remedy.9
Poliphilo emerges from this polyglot body, and looks at other ruins such
as the colossal statue of a horse that "seemed almost to tremble in its flesh,
and more alive than fabricated." He enters the pyramid through the great
portal, encounters a dragon and other amazing things, emerges from the
bowels of the great structure into a lovely meadow. He describes the
meadow by providing an encyclopedic enumeration of the specific varieties
of plants and trees. As he continues on his journey he describes an ancient
bridge over a stream with overgrown banks populated with many varieties
of birds, and then a plain fill ed with creatures, flowers, and fruit trees, each
of which he specifies by name. He meets the nymphs of the five senses,
whose clothing he describes minutely. He finally arrives at the palace of the

66 Pamela O. Long
Queen Eleuterilyda (free thought), and describes the approach to the palace
where the nymphs take him first to the baths. As he draws near to the palace
and enters, he describes the courtyard, the exterior, the ornamentation of the
entrance, the ornate tapestries of the successive rooms, the decorations, the
throne of the queen, her clothing, and jewelry, and each item of the lavish,
bejeweled services used at the sumptuous banquet. lO
After the banquet, the nymphs lead Poliphilo to the lower courtyard where
they walk through an orange grove and a series of extraordinary gardens, one
made entirely of glass and gold, a second that is an aquatic labyrinth and a
third made of silk and decorated with pearls and vines of gold. The glass gar-
den, for example, is filled with pots of artificial plants. In place of real green-
ery, "every plant was of very pure glass, excellently [made] beyond what one
could imagine or believe, into pia ted boxes with the roots and stems of gold."
Here is just one of many examples in which the natural and the artifactual are
interchanged to the delight and stupefaction of the wandering Poliphilo. "
The Hypnerotomachia Poliphili is a literary work by an author who is
enthralled by architecture, fabricated objects, and myriad specific natural
species. It contains numerous detailed lists and encyclopedic descriptions of
both natural and fabricated objects. Frequently, Poliphilo expresses aston-
ishment at both kinds of objects. The examples mentioned here-the pyra-
mid that is actually a carved-out mountain, the colossal statue of a man that
groans, the fabricated horse that trembles as if alive, and the gardens made
of beautifully fashioned glass, gold, silk, and pearls-each obscure or trans-
gress the boundaries between the artifactual and the natural. The Hypnero-
tomachia Po liph iIi exhibits an interchangeability between natural and
constructed things as it also displays a kind of descriptive exuberance in
which plants, birds, and trees are described in exorbitant detail, as are archi-
tectural ruins, structures, inscriptions, and crafted objects of all kinds.

MACHINES AND MOTION IN MADRID CODEX I

Leonardo created the Madrid Codex I, a treatise on mechanics and the


mechanical arts, between 1492 and 1497 shortly before the publication of the
Hypnerotomachia Poliphili. It is far different from the architectural romance,
but also provides evidence of the view that the natural and the artificial are
entities that are closely associated with each other. The folios of Madrid
Codex I are filled with beautiful drawings of machines and devices sur-
rounded by detailed textual explanations concerning how they work and
how they demonstrate the ways in which the natural world works. The trea-
tise as a whole focuses on practical mechanics and the sciences of weights and
motion. It treats the motions of toothed wheels, springs, clock escapement

Objects of Art/Objects of Nature


mechanisms (what he calls tempi), continuous and discontinuous motions of
various sorts, counterweights, chain gears, screws, pinions and wheels, endless
screws, lifting devices, mills, keys and locks, and crossbows, among others. In
this work and in his other writings as well, Leonardo constructs analogies of
various kinds between mechanical devices and natural phenomena. 12
He compares the mechanical and the natural early in the treatise when he
equates the earth to a cannonball: "If it were possible," he writes, "to make a
bombard having the earth for its ball, and because a bombard throws a ball of
one braccio to a distance of 3 miglia, we can measure the distance 9,000 brac-
cia which equals 9000 balls. We could therefore assume that such a bombard
would throw our earth to a distance equal to that of 9000 times its diameter.
Assuming this to be 7000 miglia, it would come to 63 thousand miglia."13
What is interesting about this passage is not that Leonardo makes a multipli-
cation mistake, but that he makes a cannonball and the earth itself inter-
changeable. He imagines the earth behaving in the same way as a ball shot by
a bombard, and attempts to calculate the resulting distance mathematically.
Madrid Codex I contains many illustrations of machines and devices that
in part are the result of Leonardo's studies of motion, but are also grounded
in his practice as an engineer, painter, and sculptor. In the 1490S Leonardo's
relationship with the Sforza court involved him in a system of court patron-
age, which he acquired primarily on the basis of his abilities as an engineer
and productive artist. I suggest a model of patronage here in which the sys-
tem of court patronage is mixed with a system of production and commerce.
The Sforza created a court that included learned humanists and famous
artisan practitioners such as Leonardo. They also strenuously involved
themselves in the armaments industry and in canal building undertaken to
improve transportation and agriculture primarily for commercial purposes.
Court patronage, material production, and commerce were not contradic-
tory activities, but functioned together in reciprocal support. 14
This is the context within which Leonardo wrote Madrid Codex I. The
folios of the codex are filled with drawings of mechanical devices and machines
that illustrate not just themselves, but serve as devices for conceptualizing prin-
ciples in the natural world and for thinking about how things both natural and
artificial work. Leonardo is interested in how devices work in practice. He
often describes their specific motions, as well as factors such as friction that
cause problems. Yet he is also interested in the principles of motion.
For example, in a section that begins with the heading "Questions," he
asks: "Why does the flame rushing to its [natural] place take the shape of a
pyramid while water becomes round in its descent, that is, at the end of the
drop?" After posing this question involving the differing shapes of two dif-
ferent elements as they seek their natural place, Leonardo launches into a
discussion of what I would call the shape of motion of a device consisting of

68 Pamela O. Long
sr';rr an-
.B,.C .

&@OM 2 . y Leonardo da Vinci, Madrid Codex I, fol. rr. Cranks


with Connecting Rods. Courtesy o f Biblioteca National, Madrid.

two cranks joined by a connecting rod (Fig. 2.1). H e illustrates the same
device twice within a chronological time sequence, first on the left, and then
in the middle after the crank has been moved. (The depiction on the far right
shows a variation.) H e invites his reader to consider the position of these two
cranks. If you move the lower crank to the left, the upper one will go right,
he says, but the length of the sensale or connecting rod will not permit it to go
all the way around because "it could not overcome the perpendicular line that

Objects of ArtlObjects of Nature 69


unites the centers of their axles. Therefore, the crank will have more ability
to turn back than to make a complete revolution."15 And indeed, in a model
of this device that I constructed, the top crank usually goes halfway around
and then reverses itself. Occasionally it makes a full circle.
As this example shows, Leonardo's drawings are not superfluous illustra-
tions of what he explains in the text. Rather the text concerning the motion
of these crank devices would not really be comprehensible without the
drawings. Text and drawings are necessary to one another. The drawings
represent devices that Leonardo may have constructed and used for obser-
vation and experiment. The pages show the devices themselves, how they
are made and how they work. They illustrate the particular motions of spe-
cific devices. Leonardo frequently reiterates the point that slight deviations
in construction will cause major problems. Craft skill is necessary in order to
construct instruments that will properly demonstrate principles of motion.
Leonardo offers hundreds of close-up studies of actual local motions made
by various devices in a great variety of different circumstances. The under-
standing of motion in the natural world, the observed motions of the devices
themselves, and the visual images of those devices are closely linked.
As this page demonstrates, Leonardo's devices function on some level as
observational tools. He constructs various devices and then observes and
describes how they move, in what direction and how fast. It is probable that
Leonardo often actually constructed, or had constructed by others, many of
the devices that he draws. We know that he frequently hired German arti-
sans to make machines and devices for him. He also mentions from time to
time his own experimentation. Ii> Some of the effects that he describes proba-
bly could be learned only by making the device and actually observing its
motions.
For Leonardo, the study of motion entails construction, observation, and
explanation using visual images. One can hypothesize that his procedure
usually consisted of constructing a particular device, drawing it, moving it,
and then describing and explaining its motion. He often adopts Aristotelian
assumptions about motion (e.g., his remark about the flame and the water
going to their natural places). Yet his mechanics are not fundamentally Aris-
totelian because they require constructing and then observing the motions of
constructed things, thereby ignoring the Aristotelian distinction between art
and nature. Neither are his mechanics Galilean, and they should not be read
anachronistically in the context of Galileo's mechanics. Unlike Galileo,
Leonardo usually does not analyze motion mathematically, nor does he cre-
ate idealized situations that eliminate such local conditions as air resistance
and friction. Leonardo creates machines and devices, and then he watches
them move and describes that motion; or perhaps in some cases he does not
construct the devices but draws them and imagines how they might move. It

Pamela O. Long
is not uncommon for him go back and forth between the motion of a device
and motions in the natural world.
For Leonardo, local motion of machines and the motion of wind, air,
water, and the earth itself can be reciprocally explanatory. He is in part
indebted to Archimedes and to the pseudo-Aristotelian mechanics as well as
to some scholastic writings on impetus. Yet his mechanics are uniquely his
own. They might be described as an observational and experimental
mechanics dependent on visual representation to explain and demonstrate
particular motions in a material world in which friction, air resistance, and
small glitches of construction are perpetual concerns.

ALBRECHT DURER AND THE COMMERCE OF ART AND NATURE

The reciprocity of natural objects and images occurs in a very different sort
of way in the work of Albrecht Durer. Durer's city of Nuremberg was an
important center of trade in the empire, and he was influenced by the com-
mercial environment in which he lived. He made two trips to Italy by
accompanying the Nuremberg merchant convoy that regularly traveled to
Venice. These trips were important to him both for artistic and social rea-
sons, and he may have met Leonardo or some of his acquaintances (such as
Luca Paccioli) on one of these trips. During his lifetime, Durer used the
printing press to gain a degree of economic autonomy and control over his
artistic production by managing the marketing of his prints. Rather than fol-
lowing the customary practice of waiting for orders, he created a stock of
woodcut prints and copper plate engravings. He kept his worked plates and
blocks in his possession and used them whenever he needed a new supply of
prints for market. At times he hired salesmen to go from city to city to sell
his prints. He also sold them himself in Nuremberg. His wife, Agnes Frey,
throughout her husband's productive work life functioned as his business
manager and handled many of his print sales, both by selling in Nuremberg
and by traveling to markets in nearby cities such as Frankfurt. l ?
Durer's work in general was far more portable than the work of the aver-
age artist of his time. Most of his production consisted of small objects that
could be carried, and bought and sold, or offered as gifts. Yet this portability
was not limited to prints, as is attested by his watercolor landscapes. In 1494
before he left Nuremberg for Venice, Durer painted two watercolors oflocal
subjects, the St. Johannis Kirche and the Wire Drawing Mill. As Peter Strieder
describes it, they are "the first known autonomous representations of identi-
fiable localities and served to raise watercolour to the status of an indepen-
dent medium, the full potential of which Durer had grasped." Shortly
thereafter, on his return from his first trip to Venice in 1495, Durer painted

Objects ofArt/Objects of Nature


u 2.2 Albrecht Diirer, View of the Val d'Arco, in the South Tyrol, 1495. Water-
color. Louvre, Paris. GiraudodArt Resource, N e w York.

watercolor views of places he passed through. An example is his painting of


Arco (fig. 2.2) which combines a masterful composition with beautiful and
diverse details of rocks, flora, and buildings. One could almost call it a
portable view that he carried back to Nuremberg. Yet as Hermann Leber
analyses in detail, it is not only a record of a view, but a carefully constructed
composition. This is just one of many examples whereby Durer created
portable pieces of "nature" by means of his art; his genius for creating natu-
ralistic representations made such paintings as his renowned hare and The
Great Piece of Turf(fig. 2.3), the latter a wonderfully detailed painting of a
grassy piece of turf, famous in his own lifetime."
For Durer objects of art and objects of nature became almost interchange-
able especially as he traveled and admired and collected a great variety of

72 Pamela 0.Long
CAyd£'J'e 2 .3 Albrecht Durer, The Great Piece of Turf, 1503.
Watercolor on paper. Vienna, Graphische Sammlung Albertina.
Courtesy of Albertina, Wien.

objects, natural and constructed, including paintings. This interchangeability


is especially evident in his diary of a trip to the Netherlands in 1420-21, as has
been pointed out by Dagmar Eichberger. For example, in the Nassau house
of Margaret, he reports that he saw "the good picture that Master Hugo
painted," two fine large halls, treasures everywhere in the house, "also the
great bed wherein 50 men can lie." He saw further "the great stone which
the storm cast down in the field near the Lord of Nassau." And finally, from
the high-standing house, "a most beautiful view, at which one cannot but won-
der; and I do not believe that in all the German lands the like of it exists."19 Var-
ious treasures, a painting by Hugo van der Goes, marvels such as a huge bed
and a stone, and an extraordinarily beautiful view are all remarked in the
same paragraph and seem to occupy the same conceptual realm.

Objects ofArt/Objects of N ature 73


Durer's delight in both natural and fabricated objects, and his careful
itemization of both, resembles in some ways the detailing of specific nat-
ural and artificial objects in the Hypnerotomachia Poliphili. Durer's won-
derfullandscapes, plants, animals, and "slices" of nature such as the Great
Tuif strikingly celebrate the ability of human artifice to create elements of
the natural world. In so doing, the artist/creator closely imitates the divine
Creator, just as artificial things and natural things come to resemble each
other.

SERLIO AND VESALIUS

The proximity of the artificial and the natural increased through the inter-
action of artisanal and learned cultures in the cities and courts of Europe.
Thus the influence of Serlio's treatise on the architectural orders, published
in 1537, upon Vesalius's Fabrica on human anatomy, published in 1543,
resulted in part from the proximity of the two men themselves and their
shared culture. Serlio trained in a workshop as a painter, lived in the Veneto
in the 1530S and associated with the painter Titian and his circle, which
included both painters and learned humanists such as Pietro Aretino.
Vesalius lived in Padua and Venice in the late 1530s; he was teaching
anatomy at the University of Padua when he wrote his treatise. Vesalius
had the illustrations for the Fabrica created in Titian's workshop, clearly
under his own close supervision (as the level of anatomical detail makes cer-
tain). He was probably familiar with Serlio's treatise on architecture. Each
man used visual representation in ways that were innovative for their
res pecti ve disci plines. 20
Serlio's book, which would eventually become part of a more extensive
treatise, focuses upon five architectural styles or orders. Serlio wrote his
didactic handbook for designers of buildings and others who were interested
in Vitruvius and in classical building styles. Pictures appear on virtually
every page and depict various parts of buildings designed in a variety of
ways. One example shows pictorially and also describes three different gates.
The first, called rustic work, is suitable for a country villa; the second, suit-
able to the Tuscan style, was seen in Trajan's Forum for a long time before it
fell into ruin; the third door with "a segmental arch which is the sixth part
of a circle, is a work of great strength." Serlio in this way depicts various
kinds of architectural styles pertinent to particular elements of the building,
many if not all from observation. He shows them to the architect who can
then pick and choose various elements for his own designs. As Serlio says
about his drawing of the Trajan Forum: "The two niches on either side are
out of place, but I have put them here in order to demonstrate the different

74 Pamela O. Long
GF 2.4 Sebastiano Serlio, De architectura libri quinque (Venice: Franciscus de
Franciscis and Joannes Chriegher, 1569),235. Elements of architecture. Courtesy of
the Burndy Library, Dibner Institute for the History of Science and Technology, MIT.

types of niches which would suit such work, so that the judicious architect
can make use of them and put them in the right place."21
Serlio often seems to work with Vitruvius's D e architectura in one hand,
and ancient architectural fragments and ruins in the other. For him, the dis-
cipline of architecture requires detailed study of the ancient text, careful
observation of ancient structures and ruins, and depiction of those observa-
tions in the form of copious illustrations. He writes, "Because I find a great
discrepancy between the buildings in Rome and other places in Italy and the
writings of Vitruvius, I wanted to show some elements which, to the great
pleasure of architects, can still be seen on buildings." He provides pictorial
representations of various parts of different buildings that he has drawn
from observation (fig. 2.4). He labels the parts of buildings with letters and
identifies them. R was found outside Rome on a bridge over the Tiber. V is
above a triumphal arch in Verona. T is in Rome on a Doric temple in the
Tullian Prison. "The capital P was found in Pesaro with many other praise-

Objects of ArtlObjects of Nature 75


worthy things: its proj ection, although it may be large, is nevertheless very
pleasing to viewers." The podium, base, and capital A are in the Forum
Borarium in Rome. 22This is a collection of architectural fragments, carefully
drawn from actual examples, for the use of practicing architects and stu-
dents of Vitruvius and of ancient architecture.
It is probable that Vesalius saw Serlio's book while he was working on the
Fabrica and took some ideas from it. Serlio's architectural representations
with their softly shaded renderings look very similar to some of Vesalius's
carefully depicted body parts. Vesalius uses illustrations as one way of nar-
rowing the wide gap between university medical learning and the artisanal
practice of surgery. He wants to integrate the artisanal practices of the sur-
geon and the apothecary into the practice of medicine as a whole. He
believes that he is restoring anatomy to its ancient splendor. In his view,
medicine was destroyed when its various components such as surgery were
broken off from it and relegated "to laymen and people with no knowledge
of the disciplines that go to serve the healing art." Similarly, the art of drugs
and medicines was handed over to apothecaries. Trained physicians only
prescribed medicines and regimes for hidden or internal ailments. As a
result, "they shamefully cast aside the foremost and most ancient limb of
medicine, the limb that above all is founded ... on the study of nature."23
Vesalius emphasizes that the order of books in the Fabrica is the same as
the order that he has followed during his own dissections in the company of
the eminent men of the city. "This means," he notes, "that those who were
present at my dissections will have notes of what I demonstrated and will be
able with greater ease to demonstrate anatomy to others." Nevertheless,
Vesalius believes that his books will be "particularly useful also for those
who cannot see the real thing." These individuals will be able, from Vesal-
ius's treatise by itself, to study each part of the body, "its position, shape, size,
substance, connection with other parts, use, function and many similar mat-
ters." In sum, from his illustrated treatise, they will be able to learn all the
things they could study during a dissection. In a remarkable defense of vir-
tual witnessing, Vesalius tells his readers, "pictures of all the parts are incor-
porated into the text of the discourse, so as virtually to set a dissected body
before the eyes of students of the work of Nature."24 Just as the architect can
study the elements of the building by perusing Serlio's treatise, so the student
of anatomy can study human anatomical parts by scrutinizing the illustra-
tions and text of the Fabrica.
Vesalius is aware of the opinions of some "who strongly deny that even
the most exquisite delineations of plants and of parts of the human body
should be set before students of the natural world; they take the view that
these things should be learned not from pictures, but from careful dissection
and examination of actual objects." He concedes that he would never urge

Pamela O. Long
students to use the pictures alone without dissecting cadavers: "Rather, I
would, as Galen did, urge students of medicine by every means at my com-
mand to undertake dissections with their own hands."?';
Nevertheless, he defends visual representation as a way of learning
anatomy. "In fact," he says, "illustrations greatly assist the understanding,
for they place more clearly before the eyes what the text no matter how
explicitly, describes." In addition, Vesalius insists that his "pictures of the
parts of the body" will give particular pleasure to those who do not have the
opportunity to dissect real bodies or are too squeamish to do SO.26
As he gets into the subject of his first book, namely bones and cartilage,
Vesalius compares the function of bones to the function of certain elements
in constructed things. Bone, he writes is "the hardest, the driest, the earthi-
est, and the coldest" of all the constituents of the human body. "God the
great Creator of all things" formed its substance to be this way for good rea-
son, "intending it to be like a foundation for the whole body; for in the fab-
ric of the human body bones perform the same function as do walls and
beams in houses, poles in tents, and keel and ribs in boats." Vesalius later
describes the cartilages that form the larynx as resembling "the beams
which form the framework of a country cottage before the thatch, the fac-
ings and the mud are applied. In fact, when the human bones and cartilages
are stripped of their flesh and then assembled together there is no better
analogy to describe them than that of the framework of a hut which has
been raised but not yet finished off with branches or earth." One gathers
from Vesalius's account of how to acquire a corpse and prepare a skeleton,
that the usual procedure involved constructing the skeleton from pieces
that had fallen apart during boiling. Even if an entire corpse was available
at the outset, which was often not the case, it would fall apart during the
usually necessary boilingY
Vesalius's many illustrations of particular bones are labeled with numbers
and letters that tie them tightly to his explanatory text. Within a discussion
of the substance of bone, for example, he shows an illustration (fig. 2.5) of a
humerus bone split lengthwise. He uses letters and visual indicators to show
the nature of the bony substance: the little holes like pumice in the capula are
marked A; the scale over these holes are marked B; C shows the outer sur-
face of the bone, D the large hollow space along the length surrounded by
the hardest bone E and F. Underneath the illustration of the humerus, Vesal-
ius shows the navicular bone which is cut through the middle to show
pumicelike bony substance. Finally, at the bottom, N is a tiny bone at the end
of the toe which is cut through the middle to show it has no pumicelike holes
at all. This illustration shows Vesalius's extensive cross-referencing as well.
The navicular bone (a small bone of the wrist) can be seen in its entirety, he
says, in figure I I of chapter 33. 28

Objects of Art/Objects of Nature 77


&quampsalslmllentur, ut eaquae 1ugu1~11.r uvcua.--- - - - .
.
: m w c \ a * c d z : &quzenfis effigiem ~ f i r n d a n t ~ u t ~ ~ ~ ~ ~ ~ ! "
zvr&Cm~ngu~mdiciin>po~tunl 05: & q ~ z r a d l uO1atl* ~
I taunmr,6gon1 erpnmant,ut CUbitios,mdij nonlinf:jOnatUm'
bo tdemi; p n d w comparcnius, ut Pedis OG cublforma no:
,ens: &quaPc mbae feufcapharrnod~cauaedm4utPedls
a g o n o m e d e ~ t : & q u r F,6;,,,~FIIw~j++&+.
F
o,& patell fimiliadicm + ~ 1 - ~ + 6 ~ ~ r n . ~ t f " ~ ~ ~ ~ ~ ~ $
FU articu o p q o G t k
z totius l t d i z circunkri 6r*ordwun
.ndir, irqpmzm-l

--
mfi& 1,yA''-
$odimodo proponmt, 4 , ,rtwm ~ p u ~ t m f~ .
qu~fibulamre~rafen +j", ,T~,.7dfi+

&,.
,rnLI.~,<d+
ruiusin dbiaosfibulaap- "&@"-.w-A~-
& ~ U CZO C C Y ~ ~ ~S ,U C U - bm,-,ommn~Fd
"&w f+-m
~ k comparentur,ut
o ti- umrbrdAt7"rqanc
xppofiturn os,quod coc- ,,,an. c a u ; , ~ ~ m ~ p

ipatur:,dein qugincudis, ti,


+,hs ~ ~ t ~ ~ f i ~ t ~ ~ t t t a , ~
r#"t,<u, q l h k
1 s denhs efigiem leuiter /v.tamm~/irr~
tent, ut minus or'ganiau- 1
- "-PR ~ntfit~
&.< ,#"jL&pm' ".Jii
culum: & quaemdleum, +#,"* E , E~ d J r d f . ,.-
us eiufdem organi oGicu- nn~t",, p w ~ ahuk .
I."gd&cm ,>pn*,lP.
terea quae A uerticuli for- &jaw u n d , ~ ~& .
nfomantur,ut dorfi uer- Dnp,ru,erinwdrH
mft"r,#fi,w-b*&
quedauo correfjSdeant, Ah ~ f i = ~ - t t r r
&quaemediam maiuku a,-~ii 6-

-
*,.WUO.
m m nq d m nfmmm,rr
uacerisfom~ampmfefe dm8~jfpr.rr&krn
uo pedis ofiicula, prim0 q =+awc.r;.dum, hk
p d m&fih.
ternodio fuppofita: eibs- e r r a <rnf~/;du-&fi- .
iisofiium,qriarformain-
iit,permulta,noniraobi
,cniofiadhucincognita '* .
p.n~ <I- 9,dm
c m F n
fi<f<G'"I

*mfp"m,gu . +
IigerEtur. Atq;hacetia~n bm*o+ UIUJ~~J?PNN-
uae in o&bus roceffud,
zm, captun>, &Perdlio- ~~zrAF>1z~2~z2 ~rp,"wJ~firn.W rC

m,&tub~rculorumOC- * p f i ~ ~ * t q ~ . ~j r+ e~ rr n . ~ , ~ h ~ ~ , ~ , , , - , , ,
"=p' eSjn pmfenti neu- L ~ i ~ m ~ Fh I ) ~ Z * ~ + U ~ + ~ ~ W W , Q ~ -
CC endzuenmnt: auern-
hptd &fitum,i" < w ~ h p w p % wm,,,,

I aliqua q u z hiCCerurnipofdt o f urn differentia,qunrn quid


endix, eiusq; ordinis reliquafvlt, haEtenus non exprefferim,i&
~ ~ c ~ a p i t e p . m & - s ~ c P r o i n d e e t i a md ~ c r ~ t iabi eofsium
conlmiffu-

2.5 Vesalius, De hurnuni corporisfabrica libri septem (Basel: I o a n n u s (


inus, 1555)' 2. T h e h u m e r u s bone, split longitudinally. Courtesy of the Burndy I
Dibner Institute for the History of Science and Technology, MIT.

T h e careful, beautiful drawings of the Fabrica and Vesalius's assiduous


cataloging of the bones of each illustration by lists of descriptive identifica-
tions marked by letters and numbers make this a treatise in which visual
representation and textual description are integrated in a remarkably close
fashion. This integration is furthered by the fact that each time a bone or
bone part is mentioned in the text, an italic indication is given in the margin
showing where the illustration of the bone or part mentioned can be
viewed -sometimes chapters away.

78 Pamela 0.Long
Serlio and Vitruvius each created new uses for visual representation in
their respective disciplines. Both men understood the value of observing the
elements of real buildings or real bodies, and both provided detailed illus-
trations that would allow such observation apart from actual objects by
means of detailed drawings. They lived in a world in which artisanal and
learned cultures were growing increasingly proximate. Each left the Veneto
immediately after the publication of his respective book, attracted by courtly
patronage. In 1537 Serlio moved to the court of Francis I in France, and in
1543 Vesalius moved to the court of Charles V in Spain. Each used his
authorship for patronage, but each also exploited the commercial book mar-
ket. Both certainly recognized that their illustrated treatises would find
readers and admirers well beyond practitioners of architecture and anatomy,
respectively, especially because of the attractive illustrations. As they used
their illustrated treatises to acquire courtly patronage, they also had their eye
on the commercial book market. In regard to that market, both complained
bitterly about the theft of their writings. 29

The sources treated here provide evidence for the pleasure taken in objects,
both natural and artificial, whether it be gardens made of glass and pearls,
crank machines, a landscape, a piece of turf, architectural elements, or bones.
They also illustrate the increased significance of visual representation of those
objects for knowledge about the world. The urban commercial economies of
the sixteenth century promoted the cultural value of objects whereas hierar-
chical social structures increasingly were delineated by means of construc-
tion, fabrication of various sorts, and conspicuous consumption. At the same
time, interactions among certain workshop-trained individuals and those
from more learned and elite backgrounds became increasingly common
within certain settings such as cities and courts. One aspect oflearned culture,
namely natural philosophy, came to be influenced by the rising cultural status
of objects and their visual depiction. What is commonly understood as the
increased importance of observation within the natural sciences is the result
of a complex set of cultural developments having to do with the status of
objects, both natural and artificial, and their visual representation.
What was deemed to be knowledge was not separate from the material!
economic/social world that treasured objects. From another point of view,
the cultural importance of objects was reflected in the the expansive number
of visual representations of natural and mechanical objects in the sixteenth
century. Those visual representations in turn functioned to explicate and
validate statements about the world whether it be the motion of machine
elements, architecture, or anatomy. In most of the examples presented in this
essay, textual and visual material belong together and are intricately joined.
Both are used to challenge traditions in at least two ways. Works such as

Objects of Art/Objects of Nature 79


Madrid Codex I showed that machines were not just objects to be constructed
to perform certain tasks within the mechanical arts, but objects to be visually
depicted and thereby able to produce knowledge about local motion. Trea-
tises such as Vesalius's Fabrica challenged the textual emphasis of the previ-
ous anatomical tradition by showing anatomical parts visually, associating
them with hands-on dissections, and suggesting that visual representations
could substitute for such dissections. Objects of art and objects of nature,
having lost their ancient distance one from another, became more or less
interchangeable instruments in the construction of knowledge.

I. For a wide-ranging study of some of the relevant issues, see David Summers, The Judg-

ment of Sense: Renaissance Naturalism and the Rise of Aesthetics (Cambridge: Cambridge Uni-
versity Press, 1987).
2. Alan Gabbey, "Between Ars and Philosophia Naturalis: Reflections on the Historiogra-

phy of Early Modern Mechanics," in Renaissance and Revolution: Humanists, Scholars, Crafts-
men and Natural Philosophers in Early Modern Europe, ed. J. V. Field and Frank A. J. L. James
(Cambridge: Cambridge University Press, 1993), 133- 145, citation on 134; and for accounts of
the conflict between Aristotelian and experimental approaches, see especially Peter Dear, Dis-
cipline and Experience: The Mathematical Way in the Scientific Revolution (Chicago: University
of Chicago Press, 1995); and Steven Shapin and Simon Schaffer, Leviathan and the Air-Pump:
Hobbes, Boyle, and the Experimental Life (Princeton, N.J.: Princeton University Press, 1985).
3. Pliny, Natural History, trans. H. Rackham, Loeb, (Cambridge, Mass.: Harvard Univer-
sity Press, Loeb, 1952) 9: xxxv. xxxvi. 65-66 (3°8 - 31 I); and see J. J. Pollitt, The Ancient View
of Greek Art: Criticism, History, and Terminology (New Haven: Yale University Press, 1974),
63- 64.
4. For an account of traditional natural philosophy and its transformations, see William A.
Wallace, ''Traditional Natural Philosophy," in The Cambridge History of Renaissance Philoso-
phy, ed. Charles B. Schmitt et al. (Cambridge: Cambridge University Press, 1988),201 - 235.
5. For an introduction to this large subject, see especially, Richard A. Goldthwaite, Wealth
and the Demand for Art in Italy, 1300-1600 (Baltimore: Johns Hopkins University Press, 1993);
Lisa Jardine, Worldly Goods: A New History of the Renaissance (New York: Doubleday, 1996);
Pamela O. Long, "Power, Patronage, and the Authorship of An: From Mechanical Know-
How to Mechanical Knowledge in the Last Scribal Age," Isis 88 (March 1997): 1 -41; and
Randolph Starn and Loren Partridge, Arts of Power: Three Halls of State in Italy, 1300-1600
(Berkeley and Los Angeles: University of California Press, 1992).
6. Long, "Power, Patronage, and the Authorship of Ars"; and Pamela O. Long, Openness,
Secrecy, Authorship: Technical Arts and the Culture of Knowledge from Antiquity to the Renais-
sance (Baltimore: Johns Hopkins University Press, 2001).
7. Francesco Colonna, Hypnerotomachia Poliphili, ed. Giovanni Pozzi and Lucia A. Ciap-
poni, 2 vols. (Padua: Editrice Antenore, 1984 [reprint, 1964]); Leonardo da Vinci, The Madrid
Codices,5 vols., Vol. I: Facsimile Edition of Codex Madrid I, and Vol. 4: Transcription and Trans -
lation of Codex Madrid I, trans. Ladislao Reti (New York: McGraw-Hill, 1974); Sebastiano
Serlio, On Architecture, Vol. I: Books I- V of 'Tufte L'opere d'architettura et prospetiva, trans. with

80 Pamela O. Long
introduction and commentary by Vaughan Hart and Peter Hicks (New Haven, Conn.: Yale
University Press, 1996); and Andreas Vesalius, On the Fabric of the Human Body: A Translation
of 'De Humani Corporis Fabrica Libri Septem, Book I: The Bones and Cartilages, trans. William
Frank Richardson and John Burd Carman (San Francisco, Calif.: Norman Publishing, 1998).
8. A lengthy romance, written in a highly Latinized Italian, it was the first illustrated
book and the first vernacular book to be published by the Aldine press. Colonna, Hypnero-
tomachia Poliphili, ed. Pozzi and Ciapponi. See also Maria T. Casella and Giovanni Pozzi,
Francesco Colonna: biografia e opere, 2 vols. (Padua: Editrice Antenore, 1959), a biography of
the Dominican monk of the Venetian monastery of SS. Giovanni e Paolo, Francesco Colonna
(1433/34-1527) who is the probable author of the work. The recent claim that the author was
Alberti does not seem convincing to me. See Liane Lefaivre, Leon Battista Alberti's Hypnero-
tomachia Poliphili: Re-Cognizing the Architectural Body in the Early Italian Renaissance (Cam-
bridge: MIT Press, 1997).
9. Colonna, Hypnerotomachia, 19-28. ("intestini, nervi et ossa, venej musculi et pulpa-
mento.")
ro. Ibid., 34, ("vedevasi quasi il tremulare degli sui pulpamenti, et pili vivo che fincto")
and 53-95.
II. Ibid., II2-I23, citation on II6, ("in loco di virentia, omni pianta era di purgatissimo
vitro, egregiamente oltra quello che se pole imaginare et credere, intopiati buxi cum gli stirpi
d'oro").
12. Leonardo, Madrid Codices, Vol. I, f. 9r., vol. 4, 24-25 (springs, clock escapements, con-
tinuous and discontinuous motions); vol. I, f. ro r., Vol. 4, 27-28 (chain gears); Vol. I, f. IS r.,
Vol. 4, 39-40 (screws); vol. I, f. 19 r., Vol. 4, 47-48 (endless screws); and Vol. I, f. 46 v., and
Vol. 4,89-90 (mills).
13. Ibid., Vol. I, f. 0 r., Vol. 4, 2, ("Se possibile fussi fare una bonbarda, che 'I mondo fussi
sua ballotta, e che sicome una bonbarda gitta una balotta d'un braccio 3 miglia, che si po mis-
urare il tal corso 9000 braccia, cioe 9 mila ballotte. Noi possiamo ad un que dire, che tal bon-
barda gitterebbe il nostro mondo novemila volte Ie grandeza del diamitro d' esso mondo
distante da sse. [Sare]bono a settemila miglia per monda [sic], sarebono 63 migliara di
miglia.")
14. For the Sforza court, see Franco Catalano, Francesco Sforza (Milan: dall'Oglio Editore,
1983); Giorgio Chittolini, ed. Gli Sforza, la chiesa lombardia, la corte di Roma: Strutture e
pratiche beneficiarie nel ducato di Milano (1450-1535) (Naples: Liguori, 1989); Gary Ianziti,
Humanistic Historiography under the Sforzas: Politics and Propaganda in Fifteenth-Century
Milan (Oxford: Clarendon Press, 1988); Evelyn S. Welch, Art and Authority in Renaissance
Milan (New Haven, Conn.: Yale University Press, 1996); and Gregory Lubkin,A Renaissance
Court: Milan under Galeazzo Maria Sforza (Berkeley and Los Angeles: University of Califor-
nia Press, 1994). For Milanese canal building, see William Barclay Parsons, Engineers and
Engineering in the Renaissance (Cambridge: MIT Press, 1968),367-419'
IS. Leonardo, Madrid Codices, Vol. I, f. I r, Vol. 4,3-4.
16. Ibid., Vol. 3, 40-41 for Leonardo's references to the various German artisans that he
hired; and for explicit references to experiments, Vol. I, ff. 77 r., 78 r., 122 v., and Vol. 4,
179-180,183-184,325-366.
17. For recent Durer studies and discussion of the importance of Nuremberg for his work,
see especially Jane Campbell Hutchison, Albrecht Durer: A Biography (Princeton, N.J.: Prince-
ton University Press, 1990), 14,23-26,57- 66, and 78-83 (for his travels and the sale of his
prints); Dagmar Eichberger and Charles Zika, ed., Durer and his Culture (Cambridge: Cam-
bridge University Press, 1998); and Peter Streider, "Durer. (I) Albrecht Durer," The Dictio-
nary of Art, ed. Jane Turner (New York: Grove Dictionaries, 1996), Vol. 9, 427-445.

Objects of Art/Objects of Nature 81


18. Strieder, "Durer," 429; and Hermann Leber, Albrecht Darers Landschaftsaqarelle:
Topographie und Genese (Hildesheim: Georg Olms Verlag, 1988).
19. Dagmar Eichberger, "Naturalia and Artefacta: Durer's Nature Drawings and Early
Collecting," in Da"er and his Culture, ed. Dagmar Eichberger and Charles Zika, 13-37; and
Albrecht Durer, Schriftlicher Nachlass, ed. Hans Rupprich, 3 vols. (Berlin: Deutscher Verein
fur Kunstwissenschaft, 1956- 69), Vol. I, 146-202. I cite the translation in Albrecht Durer,
Diary of His Journey to the Netherlands, 1520-1521, trans. Frank Conway (Greenwich, Conn.:
New York Graphic Society, 1971),64-65,
20. Serlio, On Architecture, xix; and Vesalius, On the Fabric. My comparison of the two
works takes up a suggestion by Vaughan Hart and Peter Hicks that Serlio's treatise bears
notable similarities to Vesalius's Fabrica.
21. Serlio, On Architecture, 266-269.
22. Ibid., 286-287.
23. Vesalius, On the Fabric, "To King Charles V," xlvii- xlix.
24. Ibid., liv-lv.
25. Ibid., lvi. The reference is to Galen, Procedures 2:1.
26. Vesalius, On the Fabric, lvi.
27. Ibid., and 8 and (for finding and preparing bodies) 370-384.
28. Ibid., 4.
29. Serlio, On Architecture, xi-xxxi; and Vesalius, On the Fabric, xviii- xxi.

Pamela O. Long
Mirroring the World
Sea Charts, Navigation, and Territorial Claims
in Sixteenth-Century Spain

ALISON SANDMAN

I n sixteenth-century Spain, navigation and cartography were critical


industries. Commerce with the New World formed the basis of many
fortunes and the hopes for many more. But it was also crucial for the
survival of the fledgling Spanish colonies and the interests of the govern-
ment, which carefully regulated the pilots who guided the ships across the
ocean and the charts they used to navigate. Not least among their functions,
the charts used by the pilots served as evidence in the ongoing disputes
between Spain and Portugal about ownership of territories in the Indies.
The tension between the two roles of charts-as navigational tools for the
pilots of merchant ships, and as political tools of diplomats in territorial dis-
putes-catalyzed several important changes in early modern Spain. As
experts debated the uses of charts in the first half of the sixteenth century,
their arguments crystallized ideas about the proper ways of representing the
world in a sea chart and the value of systematic theoretical knowledge in
navigation.
This chapter uses debates about sea charts as the lens through which to
view changing ideas about representations of the world and their links to
controversies about the utility of theoretical knowledge. While the practical
culture of navigation was shaped by the commercial requirements of the
merchants who provided cargo for the ships, and proved adequate to their
needs, the territorial implications of sea charts took the debate out of the
realm of commerce and brought it into that of imperial politics. This polit-
ical importance brought new people into the field, cosmographers with
closer ties to the royal court than to the pilots or the merchants. Expert in
astronomy, geography, and hydrography, these cosmographers tried to
reform navigation, focusing on general and systematic knowledge in place
of local craft knowledge. l They used the newfound official attention to sea
charts as a lever to implement their reforms.
Unfortunately, between losses at sea and the routine destruction of out-
dated charts, no examples remain of charts used at sea. We are left with a
few of the pilots' own notes and sketched charts, and with presentation
copies of charts that probably match, more or less, those sold to the pilots. 2
Fortunately, a large variety of sixteenth-century texts survive that chronicle
debates about sea charts both in Seville and at the royal court. 3 These texts
contain most of the available information about the use of charts in the six-
teenth century, and so are an invaluable source for ideas about the proper
representation of nature.
Since Seville was the center for travel to the Indies, the debates there
focused on the use of charts in navigation, and, more narrowly, on what charts
should be sold to pilots. These debates took place within the Casa de la Con-
trataci6n (House of Trade), which had been founded in 1503 to regulate trade
with the New World, including the actions of the pilots. 4 In 1523 the growing
importance of the new territories led the crown to create a new council, based
at court, to take over policy decisions relating to the Indies. As soon as debates
about charts were referred to this council, however, it became clear that more
was at issue than which charts were most useful at sea. Since 1494, when the
Treaty of Tordesillas set the boundary between Spanish and Portuguese over-
seas territories as a line oflongitude, east-west distance had been crucial to ter-
ritorial claims and disputes. As longitude could not be reliably measured at
sea, however, it was not particularly important to the pilots when navigating.
Pilots customarily did their best to estimate distance traveled, using their
knowledge of currents and winds and the characteristics of the ship. They
then set an especially careful watch when they thought they were approaching
land. Their craft did not need an accurate representation of longitude on sea
charts, which mattered more to diplomats than to navigators. Indeed the
Council of the Indies was attuned to precisely such diplomatic concerns.
Abandoning local disputes in Seville, a handful of reforming cosmogra-
phers, whom I call theory proponents to distinguish them from an even
smaller group of more traditional cosmographers, took their case directly to
the council and the king. They used the evidentiary potential of charts to
win support for their view that the key to navigation was the ability to accu-
rately locate both the ship itself and the ports to be reached in terms of a
coordinate grid of latitude and longitude. Since the only methods to do this
involved astronomical observations, they thought navigation necessarily an
application of general rules, and as such a science. 5 Though the councillors
never ruled on how pilots should navigate, in supporting the charts of the
theory proponents they also supported the latter's efforts at reform, under-
cutting the craft culture of the pilots.
The arguments about the construction and use of the official pattern
chart illustrate the key issues in the case: the representation of nature, the

Alison Sandman
disputed uses of that representation, and the ways in which these problems
were connected to debates about the nature of navigation and the impor-
tance of general knowledge. 6 During the sixteenth century the theory pro-
ponents made a concerted effort to gain control of both the institutions of
navigation and the practices of the pilots while at sea. The creation of a new
pattern chart was central to their endeavors. The royal attention that sur-
rounded the creation of a new pattern chart not only enticed new cosmogra-
phers into navigation, increasing the influence of the theory proponents, but
also provided them with an audience that was more receptive to their claims
than were pilots and merchants, and one that possessed sufficient authority
to demand changes.

CONSTRUCTING THE PADRON REAL: WHO SHOULD MAKE A CHART

All charts used at sea were supposed to match a central exemplar, known as
the padron real, or royal pattern chart. 7 In 1526, after failed boundary talks
with Portugal had underscored the inadequacy of the available charts, the
Council of the Indies decided there was need for a new pattern chart. In the
absence of the pilot major, the person at the Casa de la Contratacion in
charge of charts and instruments, they entrusted the revision to Hernando
Colon, the younger son of Christopher Columbus, ordering him to make
both a sea chart and a world map.8 As a cosmographer and bibliophile with
some diplomatic experience, Colon had long been involved in the conflicts
surrounding Spain's territorial claims in the Indies. 9 Colon chose as an assis-
tant Alonso de Chaves, arranging for him to be appointed as one of the royal
cosmographers at the Casa.1O Though little is known of Chaves's early life or
training, by 1526 he was working as a cosmographer in Seville, and he
remained a steadfast supporter of all attempts to increase the pilots' level of
education, especially their knowledge of astronomy. I I With help from
Chaves, Colon began to compile a book filled with information from return-
ing pilots, who were ordered to keep daily records while at sea. 12 Though the
book eventually contained statements from more than 150 pilots, the attempt
to make a new pattern chart seems to have been abandoned unfinished. 13
Almost ten years later, in 1535, the Council of the Indies sent a letter to
Colon in Seville, asking him about the status of the chart and ordering him to
finish it as soon as possible.1 4 The same letter also empowered the Casa de la
Contratacion officials to order all of the pilots and cosmographers in Seville to
help him finish it. At this point, however, Colon was no longer active in navi-
gation, and without his support Chaves's authority had been much diminished,
especially after the return of the pilot major, Sebastian Cabot. Cabot took over
most of the cosmographical work in Seville, helped by his friend and ally Diego

Mirroring the World


Gutierrez, who made most of the charts used at sea. l ) Both men were allies of
the pilots, and resolutely opposed the reforms of the theory proponents.
Cabot had been pilot major since 1518, shortly after Charles I (later
Emperor Charles V) came to power, but had spent the years between 1526
and 1530 on a voyage of exploration. 16 He had long identified himself with
the pilots and emphasized the importance of practical experience. In fact, he
was first hired away from England on the basis of his claims to experience in
the northern route to the New World. l ? He himself had little or no cosmo-
graphical training and did not place particular importance on the education
of the pilots, though as pilot major he was supposed to be in charge of their
training. In 1526, Cabot was sent to reinforce the Spanish presence in the
Spice Islands, and his absence gave the theory proponents their opportunity
to impose reforms. Cabot's voyage was striking for its lack of success, and on
his return he was sentenced to fines and several years of exile in North
Africa. The sentence seems never to have been carried out, and soon he was
again serving as pilot major and struggling to regain his former authority.ls
Over the next several years Cabot worked to counter the influence of
Chaves and Colon, appealing changes in the rules on licensing exams that
had weakened the office of pilot major during his absence, and supporting
the appointment of Gutierrez, who shared his emphasis on experience. 19 As
with Chaves, little is known about Gutierrez's early life. He first appeared in
the Casa records with an application to be appointed cosmographer, sup-
ported by a petition signed by Cabot and a long list of pilots. 20 Though he
won the appointment in 1534, his salary was negligible, and the bulk of his
income came from selling charts and instruments to the pilots and tutoring
them for the licensing exam. He earned enough from such sources to set up
his son Sancho as a shipowner as well as a cosmographer. 21
By 1535, when Colon was ordered to finish the padron, Cabot and
Gutierrez were firmly in charge. This changed, however, when three new
cosmographers responded to the crown's call to help revise the pattern chart,
as they were predisposed by training and social status to ally themselves with
Chaves. The first of the new cosmographers, Francisco Falero (or Faleiro),
had come to Spain from Portugal in 1518 to help organize Magellan's voy-
age. 22 He did not go on the voyage, but remained in Spain, helping as called
upon with various projects. He confirmed his credentials as a learned cos-
mographer in 1535 with the publication of a commentary on Sacrobosco's
Sphere and its application to navigationY Though he was certainly not new
either to cosmography or to Seville, Falero seems to have stayed out of local
politics whenever possible.
The others who helped revise the pattern chart were new to the profes-
sion. Alonso de Santa Cruz was just beginning to make a reputation for
himself as a cosmographer. He brought to the project an abiding mistrust of

86 Alison Sandman
Cabot, gained while sailing under him on his ill-fated voyage as an agent of
the investors. Santa Cruz had become interested in cosmography during the
voyage, and spent much of the rest of his life writing books on the subject
and devising instruments and methods for finding longitude at sea. 24 Pedro
Mexia, a local dignitary with some reputation as a humanist and an
astrologer, also joined the revision, and in 1537 the council rewarded him
with a post as cosmographer. 25 The success of his literary endeavors soon
propelled him to more prestigious positions, but even at this point in his
career he added weight to the idea of the cosmographer as educated expert.
The addition of the new cosmographers to the padron real project shifted
the balance toward people with a more scholarly and theoretical orientation;
though Santa Cruz and Falero both had experience at sea, they presented
themselves as knowledgeable cosmographers rather than as experienced
pilots. The new cosmographers also had closer ties to the royal court than
did Chaves or Gutierrez: Falero had been employed at various tasks since
1519, Santa Cruz held an appointment in the royal entourage to supplement
his income in the Casa, and Mexia was later appointed as royal chronicler.26
Though the three by no means presented a united front, all tended to ally
themselves with Chaves rather than with Cabot and Gutierrez.
With such a large and diverse group, constructing the padron was far from
easy, especially as the committee members could not agree on how to recon-
cile the pilots' reports. The six men spent more than a year in contentious
meetings in the house allocated to Juan Suarez de Carbajal, the visiting
inspector from the Council of the Indies who was overseeing the project.
Finally Suarez de Carbajal ordered them to vote on each point at issue and
abide by the majority decision. When they objected, he argued that panels of
judges routinely decided legal cases by voteY Why should mapmaking be
different? This ruling pleased no one, but it did allow them to finish. 28
Santa Cruz's objections to the proceedings were so strong that he aban-
doned the project. Before the chart was even finished he left Seville to com-
plain to the Council of the Indies. He later claimed that because of his report
the councillors held the chart in little esteem, though this is hard to reconcile
with their subsequent approval of the chart. 29 In any case, his travels were
not in vain, for Santa Cruz returned from court with increased authority
over cosmographical matters in Seville and with permission to gather the
information he thought necessary to revise the padron real. 30 Cabot fought
this increase in Santa Cruz's authority, and the issue of their relative prece-
dence was never satisfactorily resolved, though Santa Cruz began to spend
more time at court and traveling instead of in Seville. 3 ! By the end of the
1540S he had returned to Seville but was shut up in his house, writing fever-
ishly but participating little in the cosmographical work of the Casa. 32
Despite all protests, the chart approved by Suarez de Carbajal became the

Mirroring the World


official pattern chart and served as the basis both for making sea charts for
pilots and for subsequent revisions.

OBTAINING RELIABLE INFORMATION

Subsequent events, however, revealed more problems in the process, as


Cabot, Gutierrez, and Falero all joined Santa Cruz in disavowing the chart.
The ensuing discussions and justifications show the deep disagreements not
only about who should make the chart, but also about how to get reliable
information and how to decide which information to trust. The pilots' sup-
porters were outvoted, while the theory proponents argued among them-
selves about whose expertise counted and whether the problems were
sufficient to justify junking the chart.
Everyone involved agreed that the pilots were indispensable, for they
were often the only available source of information. Access to reports from
pilots was considered crucial for anyone interested in correcting charts, so
crucial that Santa Cruz, on his return from court, fought the Cas a officials
for the right to hear reports from returning pilots in a timely fashion. 33
There was also general agreement that oral testimony from experienced
pilots could usefully supplement the statements collected over a period of
time, so that it made sense to call the pilots to testify at meetings, as indeed
happened. l4 Clearly, no one thought the pilots could be ignored.
At the same time, however, most of the cosmographers did not trust the
reports of the pilots. As one pointed out in an anonymous pamphlet, when
three pilots on the same ship could calculate their position variously as 100
leagues from land, 45 leagues from land, and sailing on dry land, there was
clearly some problem in their observations, and thus some role for cosmog-
raphers. 35 The author was arguing that since the astronomy behind the
methods was unimpeachable, the problems must stem either from the
instruments or from the pilots' use of them. Though the example was surely
apocryphal, its use illustrates the contempt that many cosmographers felt for
the reports of the pilots.
Indeed, Falero gave a masterful summary of the problems of relying on
such reports. 36 Most pilots, he said, located positions using compass bearings
rather than latitudes. These compass bearings were frequently inaccurate,
because many failed to account for magnetic declination, the amount by
which the north shown on the compass was offset from geographic north.
Those who did correct their compasses for declination often made the
wrong adjustment, compounding the error. Furthermore, the few pilots
who did determine location by latitude were rarely accurate enough for
their observations to be useful. Falero concluded that since the charts made

88 Alison Sandman
from pilots' reports were based on shaky foundations, it would be best to
start anew.
Falero's proposal for reforming the charts involved nothing less than a
new survey of the West Indies. He suggested that the survey be done by
skilled pilots, all taking frequent latitude observations, using new and accu-
rate instruments, and adjusting their compasses for declination at frequent
intervals. He also wanted the pilots to keep a daily written record of their
observations, remarking that it would be impossible to give a full report
from memory. This plan essentially involved making the pilots behave as
cosmographers, and in fact he went on to suggest that if possible a cosmog-
rapher should be sent to oversee the survey. Falero's proposal shows how
deeply he distrusted the reports and methods of the pilots.
Santa Cruz, on the other hand, felt that the available reports could be
made to serve if the pilots could only work without interference from biased
cosmographers. He complained bitterly, however, about the qualifications of
the other cosmographers involved in the project. He objected that his vote
was held equal with that of Pedro Mexia, "who at that time had never in his
entire life seen a sea chart nor understood that language," and of Francisco
Falero, "who only knew a little about the sphere and judicial astrology," but
each "presumed to give his opinion, just like one who understood better,"
while despite his position Cabot could not be relied on. According to Santa
Cruz, he himself was often overruled to the great detriment of the chart, and
in sum the whole procedure had made laughingstocks of all the cosmogra-
phers involved. As an alternative, he suggested that the pilots be taken one
by one into a room with no cosmographers present and asked to point out
any flaws in the chart. If each pilot was carefully questioned under oath
about where he was licensed to go as pilot, where hehad been, and what was
good and bad about the chart as it was, they could soon have a chart much
better than any that had yet been produced. Santa Cruz never said, however,
what should be done if the pilots disagreed. Instead he seemed to assume
that the combination of oaths and detailed questioning would yield accurate
reports, and hence remove all disagreements. 37
Chaves and Mexia were working from a similar assumption, but thought
that this degree of consensus had already been obtained for most areas of the
chart. They argued that the flaws in the pattern chart had been greatly exag-
gerated, and that most of the problems were in newly discovered areas and
so would disappear when there were more reports. 38 They did not, however,
solve the problem of disputed testimony. Chaves said that one reliable state-
ment for each area would be enough, provided it was from a man "who
knew and had seen what he said," but it is not clear whether he thought any
of the pilots reached this standard. Pedro de Medina, who had arrived in
Seville only after the chart was completed, took his critique further. The

Mirroring the World


problem with relying on reports from pilots, he said, was that they risked
introducing errors rather than fixing them, which was a "very great damage
and danger to the seafarers, from the statement of just one person to lose the
truth and the statements of many."39 But he gave no indication of how to tell
which statements were reliable.
Cabot and Gutierrez blamed not the reports of the pilots, but the lack of
eyewitness experience among the members of the committee. Gutierrez did
not object to the idea of voting, but thought the wrong people were being
polled. "None of those who were there," he said, "had been in the Indies, nor
was a sailor, nor had seen the coasts, islands and bays, except only the pilot
major who was a sailor and understood the art of navigation." Though this
comment discounts Santa Cruz's time at sea, Santa Cruz had shipped as an
agent for investors in the voyage rather than as a pilot or sailor. The pilots
themselves, Gutierrez complained, who "knew by the sight of their eyes
how the coasts ran and in what latitudes they lay," were completely disen-
franchised by the procedures. 4o Cabot used this same argument as a reason to
repudiate the finished chart, claiming that he signed only under pressure, a
claim the theory proponents indignantly denied. 41
In fact a group of pilots later rejected the pattern chart on the grounds
that experience at sea was crucial to making good charts. A statement signed
by more than fifty pilots complained that "it is a harsh thing that we have to
navigate by charts made according to the padron real, which was made by
people who have not sailed, nor understand the art of navigation, nor have
experience in it, nor have seen the lands or the coasts, bays and islands."42
But protest as they might, the pilots did not have a vote.
This struggle over whose testimony should be considered reveals a fun-
damental divide between the theory proponents and their opponents. The
pilots thought the most important criterion was having seen the lands in
question, and so having a knowledge base to reject inaccurate reports. The
theory proponents, however, distrusted the ability of pilots to be adequate
witnesses. Though all agreed that the pilots needed to bring back informa-
tion on the exact locations of places, the theory proponents thought this
should be based on astronomical observations, and so distrusted reports
based on other methods. The pilots, however, determined locations not from
such cosmographical methods, but instead from the compass bearings and
distances between places. Since these were deduced from estimates of the
course the ship made good and the average speed, the pilots' reports were
closely dependent on their knowledge of seamanship and local conditions,
knowledge that was impossible without eyewitness experience.
Thus, though the theory proponents did not like having to rely on the
pilots' statements about position, the issues in this instance were methods
and type of evidence rather than the pilots' credibility per se. 43 Since, how-

Alison Sandman
ever, the pilots' fallibility as witnesses to the positions oflands was central to
the theory proponents' claims for authority, they had little incentive to com-
promise. Furthermore, as is clear from Cabot's having rejected the padron
real on the grounds that the pilots were not consulted, the issue of whose
testimony was important in making the chart was in practice inextricable
from the question of who got to judge if the resulting chart was adequate,
and on this issue the theory proponents certainly did not want to yield to
the pilots.

TWO EQUATORS AND FOUR POLES: INTERPRETING CHARTS


AND CONTROLLING NAVIGATION

The disagreement between the theory proponents and the pilots, Cabot, and
Gutierrez went well beyond a struggle over different methods of determin-
ing position and the relative importance of eyewitness testimony. More fun-
damentally, they disagreed about the relationship between a chart and the
world it represented, and what made a chart useful or true. A few years after
the construction of the padron, a new cosmographer's bid to enter the mar-
ket reopened the issue of which charts the pilots should use. Though by
royal decree all charts sold to the pilots were supposed to be exact copies of
the pattern chart, Cabot and Gutierrez ignored the padron and instead sold
charts that were more popular with the pilots.
This practice was challenged by the new cosmographer, Pedro de Med-
ina, who arrived in Seville in 1538 and tried to establish himself selling
charts and tutoring pilots. 44 Though armed with a royal license to make and
sell charts and instruments which he had obtained during a stay in court,
Medina met with consistent obstruction from Cabot, and had trouble even
gaining access to the padron real. 45 Medina's problems may have been
increased by the speculation that his presence was intended to act as a check
on Cabot's power, since during the same period the sons of Medina's chief
rival, Diego Gutierrez, opened workshops with Cabot's support and without
any apparent need to be licensed. 46
Cabot's support, however, would not have been sufficient to guarantee a
good living selling charts, for the charts also had to be acceptable to the
pilots. Cabot and Gutierrez maintained their effective monopoly through
the early 1540s, despite Medina's best efforts, in large part by being
extremely responsive to the desires of the pilots. Medina and the other the-
ory proponents, on the other hand, thought that they knew which charts
were best, and wanted to force that judgment on the pilots. Since the inten-
sive regulation of charts in effect made the court (as regulator) another cru-
cial client to be satisfied, the theory proponents seem to have hoped that

Mirroring the World


court approval would overcome the reluctance of the pilots, and so create a
market for charts matching the pattern chart.
Thus when Cabot and Gutierrez tried to block Medina's entry into the
market by denying him access to the padron real, Medina sued for access both
in Seville and at court. In the meantime, he began to study the charts sold to
the pilots by other makers. When Medina got access to the pattern chart, he
discovered that the charts in common use were a novel design by Gutierrez,
rather than being based on the padron as the law required. He then redou-
bled his efforts, beginning legal proceedings that would drag on for several
years. 47 With the support of other theory proponents in Seville, Medina
argued that the charts preferred by the pilots were unsafe and unusable
because they were false, and, being false, didn't allow the pilots to use proper
(i.e., astronomical) methods of navigation. He insisted that the only solution
was for all pilots to use charts based on the padron real.
Gutierrez's charts used two incompatible latitude scales in an attempt to
reconcile the observed latitudes of specific places with the compass bearings
between them (fig. 3.1). The problem was magnetic declination. A pilot sail-
ing westward using an uncorrected compass would travel farther south than
he thought, by an amount that increased as he moved farther west in the
Atlantic. By the time he reached the New World he would be about 3
degrees further south than expected. The solution adopted by Gutierrez was
to add a second latitude scale, offset from the first by these same 3 degrees
(fig. 3.2). In this way, the chart could still show the customary compass bear-
ings between places, the ones that the pilots were accustomed to using. But
by charting latitudes in the Old World on one scale, and latitudes in the
New World on the other, he could put places in their correct latitudes with-
out changing the compass bearings at all. On his chart a ship could sail due
west by the compass and yet wind up in a different latitude, precisely as the
pilots did. Though his charts didn't really reconcile the two sorts of infor-
mation-latitude and compass bearing-Gutierrez did find a way to rep-
resent both systems at the same time. 48
The theory proponents objected to this makeshift solution. They claimed
that to give a chart two incompatible latitude scales also implied two North
Poles, two equators, and other impossibilities, and thus "destroyed and falsi-
fied the arts and sciences, principally astronomy, geometry and cosmogra-
phy."49 In his unpublished textbook on navigation, written at about this
time, Chaves compared a sea chart to "a mirror, in which is represented the
image of the world in its absence," showing the "true description and true
location and forms of all its particularities."50 A mirror of the world could
not portray poles or equators that didn't exist.
The theory proponents also objected in no uncertain terms to any attempt
to modify the representation of the world, even to make the charts more use-

Alison Sandman
aF 3.y Diego GutiCrrez, Atlantic chart, 1550. Note especially the double
equators and tropics, more visible in detail. Courtesy of Bibliothkque National, Paris.
&%*-3.2 Detail from Diego GutiCrrez's 1550 Atlantic chart, showing the double
latitude scales. The river to the left is the Amazon. Note the two different lines repre-
senting the equator, one for each lattitude scale. The discrepancy between the two
equators is most clearly seen at the lower end of the latitude scale on the right.
Courtesy of Biblioth6que National, Paris.

ful. Chaves argued that it was better to teach the pilots to navigate by lati-
tude than to distort the charts, adding that "it is more appropriate that they
search for the path on the earth, than to put the earth in the path that they
use when it isn't there."51Medina compared GutiCrrez's solution to a doctor
who treats a man with one arm dislocated by dislocating the other.52All the
theory proponents agreed that there was no excuse for putting anything on
the chart that had no counterpart in nature.
The pilots completely rejected this interpretation of the charts. For them
the double scale was a convenience that allowed them to find the latitudes of
places, not a claim about the existence of two equators. Several pilots
affirmed that they understood the two scales to be one and the same, but
drawn as two to put the lands in the right relation to each other.53In taking
the two equators literally, they claimed, the cosmographers missed the point.
The theory proponents, however, did not trust the pilots' opinion of the sea
charts, saying that they spoke with the "common voice" and were insuffi-
ciently educated to be able to judge.54
This was not simply a fight about the propriety of different types of rep-
resentation, for underlying the debate about which charts should be used
was a long-standing disagreement about how pilots should navigate. The
theory proponents first became involved in navigation with the appointment

94 Alison Sandman
of Colon to revise the padr6n in I526. From that time on, they had tried to
change the way in which the pilots navigated by emphasizing the importance
of astronomical methods of navigation. Soon after Colon was asked to revise
the padr6n in I526, he was appointed as acting pilot major during Cabot's
absence. 55 Under Colon's aegis the pilots' licensing exams had become more
rigorous; he had even ordered some previously licensed pilots to retake their
exams. 56 Chaves, too, was a part of this new regime; one of his first actions
after being appointed cosmographer in I528 was to request permission to give
daily lessons to the pilots on the use of charts and instruments. 57
The composition of Chaves's proposed classes indicates the ways in which
the theory-proponents wanted to reform navigation. The authorization letter
from the court noted Chaves's desire to give a daily lecture in his house to
pilots and seamen on "the use of the astrolabe and quadrant and sea chart,
with the treatise of the Sphere." Although the ability to use a chart, an astro-
labe and a quadrant was already supposed to be part of the licensing exam for
pilots, the extent to which pilots found these skills useful at sea was still in
dispute twenty years later, as was the extent to which the pilots were learning
enough in the classes to use the instruments properly.58 Since the main use of
an astrolabe or quadrant at sea was to find latitude, this emphasis on instru-
ments implied a reliance on latitude measurements, and so on astronomy.
Furthermore, Chaves did not plan to limit his classes to the use of instru-
ments. He also proposed to teach the pilots from the standard introductory
astronomy text, Sacrobosco's Sphere. 59 This would probably have involved
discussions of the sphericity of the earth, the zodiac, the moon and tides, and
the equator, tropics, and poles. This curriculum assumed that pilots needed to
understand the reference system of elementary astronomy if they were to find
latitude at sea. The available methods required them not only to observe the
altitude of the sun or north star, but also to know which tables to use, and
how to apply the necessary corrections. Most sixteenth-century navigation
textbooks included discussions of these issues and offered detailed rules to
follow, in addition to tables and explanatory diagrams. 60 In fact, the need for
such tables and calculations was one of the reasons cited later for requiring
pilots to be able to read and write, another project of the theory proponents. 61
The methods in the textbooks could all be learned by rote, but Chaves, with
the strong support of Colon, wanted the pilots to understand the astronomy
underlying them, on the conviction that this would help them navigate.
When Gutierrez gained control of the education of the pilots in the early
I530s, he presided over a distinct decrease in the theoretical content of the
classes. In discussing the classes, he made no mention of the Sphere or astron-
omy, but quickly passed on to his expertise in making nautical instruments.
Other cosmographers later derided his lack of learning, complaining that a
man who didn't know the Sphere or understand Latin could hardly be con-

Mirroring the World 95


sidered an expert. 6e A midcentury investigation into practices at the Cas a de
la Contrataci6n branded his classes as superficial and corrupt. He was
accused not only of accepting bribes to pass unfit candidates (a common
complaint in the sixteenth century), but of providing a list of questions and
answers for would-be pilots to memorize. 63 Pedro de Medina, Gutierrez's
chief com petitor as a teacher during the 1540s, further testified that if the
candidate could not read this list (and thus did not qualify as even margin-
ally educated), Gutierrez would teach it to him orally.64 The thrust of the
complaint was that the candidates could then recite the answers without
actually understanding any of the cosmography underlying them. Certainly
if Gutierrez did not understand the astronomical underpinnings of celestial
navigation, he could not be expected to teach it to the pilots.
Though the argument over the extent to which pilots needed to master
astronomy seems far removed from the concerns involved in making a sea
chart, in practice the two were closely linked. Pilots trained in astronomy
and the use of instruments could use latitude measurements to navigate, and
thus would be more aware of the exact location of lands and ships at sea.
Pilots not so trained relied instead on the compass bearings between places,
which left them, according to one of the theory proponents, "ignorant and
deprived of the general and well-founded science and art [ciencia y arte] by
which they needed to govern themselves."(,)
The ability to navigate by compass bearings was precisely what Gutier-
rez's charts facilitated. According to the pilots, the reason Gutierrez's charts
were more useful than those matching the padr6n real was because its double
scale allowed them to use compass bearings to navigate instead of having to
rely on latitude measurements. They considered the compass to be their pri-
mary tool; the pilots' guild later compared a ship without a compass to a
man without eyes. 66 Though all of the pilots had to know how to take lati-
tude measurements from the sun and North Star, they consistently com-
plained that these were unreliable, since the weather might not permit
observations for several weeks at a time, and when it did the results were
often inaccurate. 67 Using the two-scale charts they could travel by compass
bearing, perhaps using occasional latitude observations as a check, and with-
out having to adjust for magnetic declination. For those sailing by compass,
the pilots agreed, the charts were extremely useful.
The charts matching the padr6n real, which the pilots called latitude charts
(cartas de altura), could not be used in this way because in making sure that
ports were put in the correct latitudes, the chart makers changed the bearings
between them. In fact many of the pilots said they didn't know how the lati-
tude charts could be used. 68 The theory proponents explained that the charts
were intended to be used with frequent latitude observations, which then
corrected courses set by a compass regularly adjusted for magnetic declina-

Alison Sandman
tion. 69 This method, however, was too complicated and cumbersome for most
of the pilots. To them, the practical utility of the two-scale charts was both
obvious and all-important. Indeed, it was at the request of the pilots that
Cabot and Gutierrez (or so they claimed) had begun to sell two-scale charts
in the first place. 70
In essence, then, everyone agreed that available charts limited the possible
methods of navigation. The theory proponents wanted to use this fact to
force the pilots to navigate by latitude, by controlling the charts they used.
The pilots, with the help of Cabot and Gutierrez, successfully resisted this
move, instead retaining (for a time) the charts that allowed them to navigate
by compass bearing. The incentive of a sufficiently large and unified market
(the pilots), as well as an ideological commitment that harmonized with the
pilots' demands, convinced Cabot and Gutierrez to defy the rules by making
and selling two-scale charts instead of ones matching the pattern chart. They
even partially succeeded in convincing the officials in Seville to ignore the
regulations they were sworn to uphold. 71 These officials listened to the
warnings of the pilots and agreed that forcing them to use charts they did
not understand could only cause shipwrecks and loss of life and property;
they counseled that any changes be slow and cautious. 72 But the Council of
the Indies, more isolated from the pilots by its position at court, found these
arguments less compelling, and banned the two-scale charts?' By 1545, ten
years after they had constructed the chart, and after twenty years of fighting
for control of navigation, the theory proponents had won.

SEA CHARTS AS TERRITORIAL CLAIMS

So far this has been a local argument between two factions of cosmographers
in Seville. The theory proponents, due to both their ideas of navigation
(which privileged latitude over compass bearings) and their ideas of repre-
sentation (which placed a high value on literal representation) favored the
charts matching the padron real, which indeed they had helped construct in
the first place. They also placed great importance on the pilots' ability to
report back about where they had been so the charts could be improved.
This would necessitate their being literate (so they could keep good records)
and making frequent latitude observations (so they had data to report) .
Cabot and Gutierrez supported instead the arguments of the pilots, that any
sea chart that allowed the pilot to guide a ship safely was a good chart. As a
group of pilots explained, there were many types of charts for different pur-
poses, but the pilots' own charts served as "nothing more than to go from
here to the Indies." They suggested that for other concerns, such as territory
and boundaries, the Cas a officials instead use charts "of all the universe,"

Mirroring the World 97


which were clearly distinct from the sea charts used by the pilots. 74 The theory
proponents, however, rejected this distinction, insisting that one type of
chart be fit for all uses.
Once the debate left Seville, it ceased to center on the practical needs of
the pilots, and focused instead on other uses of charts. The pilots demanded
only that the charts be useful, that is, that they provide an algorithm for
crossing the ocean. While the cosmographers did want charts that were use-
ful at sea, they also asked that the charts be accurate reflections of the world.
As such, the cosmographers' charts could be used to support Spanish territo-
rial claims as well as to cross the Atlantic, while those the pilots preferred
could not. While by no means central to the cosmographers' disagreements
with the pilots, these diplomatic implications did give the cosmographers
additional leverage with the Council of the Indies. Once sea charts began to
take on a larger political significance, the choice of which chart to use
became more than a technical issue, and became too important for pilots and
cosmographers to settle among themselves.
In sixteenth-century Spain, the exact location of territories was a politi-
cally sensitive issue, as all the cosmographers were well aware. To under-
stand this we have to go back to 1494, when the treaty of Tordesillas set the
boundary between Spanish and Portuguese territories. Though the treaty
specified a line of longitude 370 leagues west of the Cape Verde Islands, it
did not specify which island in the group. 7) It also failed to indicate any land-
mark to locate the line, for the simple reason that no one knew what was
there. These details were supposed to be worked out by a bilateral commis-
sion, but the commission never met. Despite some diplomatic skirmishes,
both countries were too involved in other parts of the world to worry over-
much about the exact boundary line. 76 Though the line of demarcation was
officially a line of longitude, and thus something to be determined by cos-
mographers, at first it was not very important in practice.
All this changed with Magellan. He left Portugal for Spain promising to
prove that the Spice Islands fell into Spanish territory (fig. 3'3).77 In 1522,
the successful return of one of his three ships inflamed Spanish hopes of
riches in the East, and caused both sides to reopen the question of demarca-
tion. Both agreed that the obvious solution would be to extend the line of
demarcation to complete the great circle, thus creating a line in the east.
Locating this line, however, was problematic, even in theory. First the par-
ties had to agree on the location of the line in the Atlantic, which involved
agreeing on a starting place and on the size of the earth (and so the conver-
sion between leagues and degrees). Then they had to draw a line exactly
halfway around the world from the first, and (more important) agree
where that line was in reference to known places, since neither side had any
reliable longitude measurements.

Alison Sandman
6 ;.
S
0
3
;= aJ
rfi 5
P m
L
4 2

8S
:24
ECM
S
.-
9
Y

23
w
Ea
2
0
:.>
.C c
2 s
e 0

& 3 &
-0 -0

B aJ .s
-0 8"a
5 a .g
4-r
2L o zm
u
g ;"U
5- 5 4
22 g
z:z
z-2 E
g X

2%
W

&
'SZk
c 6 'G
g "U a
c .2 ':
24
&b
-c
g
" "2
z3
.aU 24BC 2
W
r 5-0
2 y .4
5 3-e
m g .s
I m C
0 2 .z
E,2 q
9z
4 $q
-
w -
2 &- $
In 1524 Spanish and Portuguese delegates met to try to decide the loca-
tions of the demarcation lines. 78 Hernando Colon served as one of the dele-
gates for Spain. Sebastian Cabot was on hand as an advisor, along with a
variety of other pilots and cosmographers. These talks hinged on cosmo-
graphical issues-the length of a degree, the locations of certain groups of
islands, and the ability to measure longitude at sea. The talks were incon-
clusive, and the issue remained unsettled until 1529 when the Spanish sold
all rights in the Spice Islands to Portugal.
The discussions did, however, serve to underscore both the unreliability of
the pilots' reports (since none of them could agree on locations of the islands)
and the utility of each side's charts as evidence of the locations of places. In
fact, much of the diplomatic maneuvering centered on charts. The Por-
tuguese withdrew from discussion several of the charts and globes they had
prepared once it became apparent that they could be used to support Spanish
claims. Whether this was a dispute over projections or pure obstructionism,
the result of the political maneuvering was to paralyze the discussions. 79 The
Spanish, for their part, advocated the use of older charts, since only recently
had the Portuguese had any motive for falsifying the distances. 8o
Neither side seems to have expected the talks to be decisive. Perhaps this
was due to the considerable technical difficulties in deciding the longitude of
the islands, for the uncertainties were huge. The Spanish team pointed out
that between their account of travel to the west, and the Portuguese account
of travel to the east, they failed to account for almost 50 degrees of longitude,
a substantial percentage of the earth. Hernando Colon considered the loca-
tion of the line of demarcation too uncertain to provide convincing proof of
Spanish ownership of the Moluccas. He advised that the Spanish instead
assert ownership by right of discovery, and leave to the Portuguese the prob-
lem of proving any treaty rights. sl Perhaps, also, the discussions were hin-
dered by the improbability of either country being willing to give up valuable
property to comply with a disputed expert ruling. Only a unanimous decision
by all the cosmographers could be expected to exert the necessary moral
force, and the technical difficulties were too great to expect such unanimity.
It appears that the Portuguese believed the Moluccas were probably on the
Spanish side of the line, and so determined to obstruct the talks. 82
Thus when the committee of Spanish cosmographers gathered in Seville
in 1535 to revise the padron real, they were all keenly aware of the political
implications of the exact locations of lands. Cabot had been present at the
negotiations with Portugal, and one of the aims of his voyage in 1526 had
been to reinforce the Spanish presence in the East Indies. Falero, who was
born in Portugal, h ad first come to Spain with Magellan and helped him to
convince the Spanish that the Spice Islands would fall within their demarca-
tion. Though the other cosmographers had less direct ties with the territor-

roo Alison Sandman


ial dispute, they could not have been unaware of it. Furthermore, when the
padron was investigated in the 1540s, one of the issues discussed was whether
either of the two competing charts placed territories claimed by Spain on the
Portuguese side of the line.
The legal testimony directly reflected concerns about demarcation, espe-
cially in the questions asked of each witness. Sancho Gutierrez made the
diplomatic importance of the charts explicit. Discussing the charts made by
his father, he argued that "although in the locations of the land the charts do
not differ" one type was better to use than the other, because it minimized
the risk of foreigners misunderstanding the charts, and so coming to the false
opinion that certain areas were outside of Spanish territory.83 A representa-
tion needed not only to be accurate, but also transparent to even a hostile
observer. In rebuttal, Diego Gutierrez asked the defense witnesses whether
the chart misplaced the line of demarcation. 84 Falero, Mexia, and Chaves
condemned the two-scale chart on the issue, warning that since the chart
misplaced latitudes in the New World, it decreased distances, and so moved
the line of demarcation in Portugal's favor. 85 This claim still involved an
interpretation of the chart, but an interpretation that they judged valid. This
was not an argument the Council of the Indies could ignore.
These cosmographers were not suggesting that anyone falsify charts to
support Spanish territorial claims, although perhaps only because they did
not think falsification necessary in this case. In their view, however, all charts
made claims about the locations of overseas territories, regardless of their
intended use. They rejected the pilots' claim, that different types of charts
with different uses should be judged only according to that use, on both
intellectual and practical grounds: intellectual, because they did not believe
that a chart with an inaccurate picture of the world could possibly be useful,
and practical, because they did not trust the potential users of charts to
respect the intention of the makers. Only a faithful and accurate representa-
tion of the world could be useful simultaneously to navigators (provided
they used the proper astronomical methods) and to diplomats. All charts
should mirror the world.

CONCLUSIONS

The theory proponents and the pilots disagreed about the role that eyewit-
ness experience should play in the construction and evaluation of sea charts,
about whether a general knowledge of astronomy was more important to
navigation than local knowledge of winds and currents, and about the link
between charts as tools of navigation and charts as pictures of the world.
Most of these issues, however, were ignored by the Council of the Indies.

Mirroring the World !OJ


The theory proponents were able to gain the support of the council in large
part because of the growing political sensitivity of charts. Once the council
approved the charts of the theory proponents, however, the rest of their
reforms followed. Once the placement of lands on a sea chart was no longer
simply a technical matter, but part of a larger constellation of diplomatic
claims, the pilots ceased to be the ultimate arbiters of the quality of any given
chart, and the ability to describe locations accurately became more central to
navigation. Though it did not hurt that the theory proponents were more
likely to have friends at court than were their opponents, this contact also made
them more sensitive to political concerns. Their ability to speak to concerns
about both demarcation and the accuracy of charts brought them the approval
of the Council of the Indies, and so support in their attempts to reform naviga-
tion. Just a few years later the council chose Chaves to be pilot major and cre-
ated a new position of professor of cosmography which went to his son. These
changes consolidated the control of navigation by cosmographers.
Making sure that ships could safely cross the ocean was important to the
Spanish state because they carried the trade that increasingly supported both
the Spanish economy and the overseas colonies. The technical side of naviga-
tion, including the charts used, was usually relegated to a small team of
experts. After Magellan's voyage increased Spanish interest in the East Indies,
however, the political importance of the exact locations of territories focused
increasing attention on representations of the world, including sea charts.
This provided an opportunity for a small group of reformers to argue for the
importance of astronomical methods of navigation, on the grounds that they
allowed pilots to report accurately on the positions of the ports they visited.
The spread of astronomical methods and the subsequent improvements in
techniques of navigation thus owed far more to the territorial preoccupations
of the Spanish crown than to any practical problems in guiding ships across
the ocean. The politicization of cosmography was instrumental in changing
the approach to navigation from pragmatic to theoretical, and so in spreading
the idea that theoretical knowledge was indeed useful in practice.

This article forms a part of my dissertation, Cosmographers vs. Pilots: Navigation, Cosmography,
and the State in Early Modern Spain. I am grateful for research support from the University of
Wisconsin, the Fulbright organization, the National Science Foundation, and the Spencer
Foundation.
1. Cosmography in Spain has been the subject of much recent research. See, for example,

Mariano Esteban Pineiro, "Los Oficios Matematicos en la Espana del Siglo XVI," in II
trobades d'historia de fa ciencia i de fa tecnica, ed. Vicente L. Salavert, Victor Navarro Brot6ns,

102 Alison Sandman


Mavi Corell (Barcelona: Institut d'Estudis Catalans, 1994),239-251; Victor Navarro Brotons,
"La Cosmografia en la Epoca de los Descubrimientos," in Las relaeiones entre Portugal y
Castilla en la epoea de los deseubrimientos y la expansion colonial, ed. Ana Maria Carabias Tor-
res, Acta Salmanticensia Estudios Historicos y Geograficos, 92 (Salamanca: Ediciones Uni-
versidad de Salamanca, 1994), 195-2°5; M. I. Vicente Maroto and M. Esteban Pineiro,
Aspectos de la Cienda Aplieada en la Espana del Siglo de Oro (Madrid: Junta de Castilla y Leon,
1991). For a review of literature in the field, see Mariano Esteban Pineiro, "Cosmografia y
Matematicas en la Espana de 1530 a 1630," Hispania 51, no. 177 (1991): 321 -337.
2. Ricardo Cerezo Martinez, La Cartografia Nautiea Espanola en los Siglos XlV, Xv, y XVI
(Madrid: Museo Naval, 1994), 14I. In his appendix he lists all known sea charts from the
period.
3. The majority of these texts are now in the Archivo General de Indias (AGI) in Seville,
Spain, primarily in the legal records (section Justieia).
4. Jose Cervera Pery, La Casa de Contrataeion y el Consejo de Indias (Las Razones de un
Superministerio) (Madrid: Ministerio de Defensa, Secretaria General Tecnica, 1997), 13-18;
Ernesto Schafer, El Consejo Real y Supremo de las Indias; Su Historia, Organizaeion y Labor
Administrativa Hasta la Terminaeion de la Casa de Austria (Seville: Universidad de Sevilla, Pub-
licaciones del Centro de Estudios de Historia de America, 1935), especially 82ff.
5. Art (arte) and science (eieneia) were generally treated as synonyms, contrasted with the
office (ofieio) of a pilot, which involved the traditional methods learned at sea. The cosmog-
raphers did not wholly discount these traditional methods, but wanted the pilots to learn
astronomical methods in addition. For a discussion of this distinction, see Maria Isabel
Vicente Maroto, "El Arte de Navegar," in Felipe II, la Cieneia y la Teeniea, ed. Enrique
Martinez Ruiz (Madrid: Aetas Editorial, 1999),343-344, On navigation as part of cosmogra-
phy, see AGI, Justicia, II46, N. 3, R. 2, block 3, image 102,3 September 1544 statement from
Pedro de Medina; on the necessity of astronomical observations and general rules, see images
336 and 339-340,10 April 1545 statement from Alonso de Chaves.
6. On the chart's construction, see Cerezo Martinez, Cartografia Nautiea Espanola, 201 -204.
Most information on the chart comes from a 1544-1545 dispute (AGI, Justicia, II46, N. 3, R.
2), which prompted discussion of its construction, and from Alonso de Santa Cruz's 6 Septem-
ber 1549 letter to Hernan Perez de la Fuente (AGI, Justicia, 945, ff. r68r-17rr).
7. On the idea of a padron, see Cerezo Martinez, Cartografia Nautiea Espanola, 137- 140.
On inspections intended to ensure the conformity of the charts, see AGI, Indiferente, 1207, N.
61 (1546 discussion of procedures) and AGI, Indiferente, 2005 (1565 order on the disposition
of instruments failing inspection).
8. Cerezo Martinez, Cartografia Nautiea Espanola, 190. The relevant orders are dated 6
October 1526 and 16 March 1527, AGI, Indiferente, 421, L. II f. 234rv and L. 12, f. 4orv.
9. For a biography of Colon, see Tomas Marin Martinez, "Estudio Introductorio," in
Catalogo Coneordado de la Biblioteea de Hernando Colon, ed. Tomas Marin Martinez, Jose
Manuel Ruiz Asencio and Klaus Wagner (Cabildo de la Catedral de Sevilla, Fundacion
MAPFRE-America, and Editorial MAPFRE, 1993), 27-35I.
10. AGI, Indiferente 421, L. 13, f. 82r, 4 April 1528 letters to Chaves and officials of the
Casa de la Contratacion.
1I. His work as a cosmographer is inferred from his possessions listed at the time of his
marriage that year. Paulino Castaneda Delgado, Mariano Cuesta Domingo, and Pilar
Hernandez Aparicio, Transeripeion, Estudio y Notas del Espejo de Navegantes de Alonso de
Chaves (Madrid: Instituto de Historia y Cultura Naval, 1983), 15; Jose Pulido Rubio, El Piloto
Mayor: Pilotos Mayores, Catedratieos de Cosmografia y Cosmografos de la Casa de la Contrataeion
de Sevilla (Seville: Escuela de Estudios Hispano-Americanos, 1950), 609.
12. AGI, Indiferente, 421, L. 12, f. 4orv, 16 March 1527'

Mirroring the World 103


13. On the book, see AGI, Justicia, 1146, N. 3, R. 2, image 97, Pedro de Medina, 3 Sep-
tember 1544. Ricardo Cerezo Martinez argued (Cartografia Ndutica Espanola, 191-192,
201 -203) that a chart was indeed finished at this time, but that later administrators were
unaware of it.
14. AGI, Indiferente, 1961, L. 3, f. 276rv, 20 May 1535.
15. AGI, Justicia, 836, N. 6, images 580-625, testimony from various pilots. On Gutierrez,
see Robert W. Karrow, Jr., Mapmakers of the Sixteenth Century and Their Maps: Bio- Bibliogra-
phies of the Cartographers of Abraham Ortel/us, 1570 (Chicago: Speculum Orbis Press, 1993),
285-287. See also the discussions in Ursula Lamb, "Science by Litigation: A Cosmographic
Feud," Terrae Incognitae I (1969): 40-57, and Ursula Lamb, "The Sevillian Lodestone: Sci-
ence and Circumstance," Terrae Incognitae 19 (1987): 29-39.
16. On Cabot's appointment as pilot major, see Jose Toribio Medina, El Veneciano
Sebastian Caboto al Servicio de Espana, y Especialmente de su Proyectado Viaje a las Molucas por
el Estrecho de Magallanes y al Reconocimiento de la Costa del Continente Hasta la Gobernaci6n de
Pedrarias Davilla, 2 vols. (Santiago de Chile: Imprenta y Encuadernaci6n Universitaria, 1908),
vol. 1,25-33' While this book remains the best source for Cabot's time in Spain, and espe-
cially for his voyage, the details of Cabot's early life are contentious. For a brief overview of
the historiography, see Karrow, Mapmakers, 103- I 12.
17. Medina, El Veneciano, I: I -6. The actual extent of Cabot's experience is debated, but
there is no doubt that he convinced the Spanish officials interviewing him.
18. Ibid., 303-313, 332. Vol. 2 contains many of the relevant documents.
19. AGI, Indiferente, 2005, undated petition from Cabot, reproduced in Medina, El Vene-
ciano, I: 507.
20. AGI, Indiferente, 1204, N. 21, undated petition from Gutierrez, supported by pilots'
petition with twenty-nine signatures from September 1533.
21. AGI, Contrataci6n, 5784, L. I, f. 58v, 21 May 1534 cedula giving Gutierrez a salary of
6,000 maravedis (16 ducados) per year, though most cosmographers were paid 30,000 mrs or
more. On prices, see AGI, Justicia, 836, N. 6; in 1551 Gutierrez charged 4 ducados for a sea chart
and 2 to 6 for lessons. On Sancho's inheritance, see Lamb, "The Sevillian Lodestone," 204-206.
22. Ricardo Arroyo Ruiz-Zorrilla, "Estudio," in Tratado del Esphera y del Arte del Marear
con el Regimiento de las Alturas, by Francisco Falero (Madrid: Ministerio de Defensa; Ministe-
rio de Agricultura Pesca y Alimentaci6n, 1989),9- 10.
23. Francisco Falero, Tratado del Esphera y del Arte del Marear con el Regimiento de las
Alturas, con Algunas Reglas Nuevamente Escritas Muy Necessarias, 1535, facsimile edition with
introductory study and transcription by Ricardo Arroyo Ruiz-Zorrilla (Madrid: Ministerio de
Defensa; Ministerio de Agricultura Pesca y Alimentaci6n, 1989).
24. Mariano Cuesta Domingo, Alonso de Santa Cruz y Su Obra Cosmografica, 2 vols.
(Madrid: CSIC, 1983), I: 35-61.
25. Antonio Castro Diaz, Los "Coloquios » de Pedro Mexia: Un Genero, una Obra y un
Humanista Sevilla no del Slglo XVI (Seville: Excma. Diputaci6n Provincial de Sevilla, 1977),
75- 84.
26. Arroyo Ruiz-Zorrilla, "Estudio"; Cuesta Domingo, Alonso de Santa Cruz, I: 81; Castro
Diaz, Los "Coloquios» de Pedro Mexia, 81-82.
27. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 84, Gutierrez's account of Suarez de Car-
bajal's statement.
28. Ursula Lamb ("Science by Litigation," 56) has argued that Suarez de Carbajal was try-
ing to determine scientific truth by majority rule. I would argue instead that he was trying to
find a way to reach a decision in the absence of conclusive data.
29· AGI, Justicia, 945, f. 168v, 6 September 1549 letter from Santa Cruz to Hernan Perez
de la Fuente.

Alison Sandman
30. This .involved giving him an appointment as cosmographer. On gathering informa-
tion, see AGI, Indiferente, 1962, L. 5, ff. 4IV-42V, 20 and 21 November 1536. On his author-
ity over Cahot, see AGI, Indiferente, 1962, L. 5, f. 4IV, 20 Noember. 1536.
31. For their running battle over authority, see the petitions in AGI, Indiferente, 2005. For
Santa Cruz's movements, see Alonso de Santa Cruz, Cronica de los Reyes Catolicos, edition and
study by Juan de Mata Carriazo (Seville: Publicaciones de la Escuela de Estudios
Hispano-Americanos de Sevilla, 1951), v.
32. For Santa Cruz's own account of his activities at the time, see AGI, Justicia, 945, f.
168v, 6 September 1549 letter to Hernan Perez. For Perez's interpretation, see his letter of 22
September 1549, AGI, Indiferente, 1093, N. 98. On 20 November 1549 Santa Cruz was
charged with not attending licensing exams; see AGI, Justicia, 945, ff. 209V-2IIr.
33. AGI, Indiferente, 2005, various 1538 petitions.
34. AGI, Justicia, T 146, N. 3, R. 2, block 3, image 93, Pedro Mexia, 2 September 1544;
block 3, image 97, Pedro de Medina, 3 September 1544; block 2, images 37-38, Alonso de
Chaves, 9 September 1544·
35. "Coloquio Sobre las Dos Graduaciones Diferentes Que las Canas de Indias Tienen,"
in Disqul~,iciones Nautieas, ed. Cesareo Fernandez Duro, Vol. 6: Area de Noe (Madrid: Minis-
terio de Defensa, 1996 [1881 D, 513. Though Fernandez Duro identified the anonymous
author as Hernando Colon, Ursula Lamb ("Science by Litigation," 50) has convincingly
argued that the coloquio was instead written by Pedro de Medina.
36. AGI, Justicia, 1146, N. 3, R. 2, block I, images 15-17,5 May 1545 statement by Fran-
cisco Falero.
37. AGI, Justicia, 945, ff. 168v and 169rv, 6 September 1549 Santa Cruz to Hernan Perez.
38. AGI, Justicia, II46, N. 3, R. 2, block 3, image 94, Pedro Mexia, 2 September 1544, and
image 339, Alonso de Chaves, 10 April 1545.
39. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 99, Pedro de Medina, 3 September 1544.
40. AGI, Justicia, Il46, N. 3, R. 2, block 3, images 107-108 (both quotes), Diego Gutier-
rez,9 September 1544·
41. On Cabot and Gutierrez, see AGI, Justicia, II46, N. 3, R. 2, block 2, f. qv and block
3, ff. 12r-I:;r. For denials, see AGI, Justicia, II46, N. 3, R. 2, block 2, images 43 (Chaves) and
378 (Mexia). Falero (image 3(9) testified that there was a lot of controversy but everyone
signed.
42. AGI, Justicia, 1I46, N. 3, R. 2, block 3, image 371, undated 1545 statement.
43. In this it differs from the English context as described in Steven Shapin, A Social His-
tory of 'l1'uth: Civility and Science in Seventeenth-Century England (Chicago: University of
Chicago Press, 1994). For a discussion of the basis for credibility critiquing Shapin, see Bar-
bara J. Shapiro, A Culture of Fact: England, [550-[720 (Ithaca, N.Y.: Cornell University Press,
2000), especially 12-26, 70-76, and 120- I27. Since the role of juries is central to her analysis
of England, her findings are not directly applicable to Spain, but many similar factors apply.
44. For a biographical sketch of Medina see Pedro de Medina, A Navigator's Universe: The
Libro de Cosmographfa of 1538 by Pedro de Medina, ed. Ursula Lamb (Chicago: University of
Chicago Press for the Newberry Library, 1972),9-18.
45. See AGI, Indifercnte, 1962, L. 6, f. 156r, granting the right to make and sell instru-
ments; f. I64; ordering that he be given access to the padron; AGI, Indiferente, 1963, L. 7, f.
19; ordering that the other cosmographers inspect Medina's charts and instruments; and f.
84 v, ordering that there be no monopoly on making charts.
46. For speculation about Medina's intended role, see AGI, Indiferente, 2673, undated
1551 letter by Alonso Zapata, pilot, decrying rampant corruption among cosmographers in
Seville. On the Gutierrez family, see Chaves's later complaint in AGI, Justicia, II46, N. 3, R.
2, block 2, f. lOr.

Mirroring the World 105


47. The ensuing litigation has been much discussed from a variety of points of view, most
recently in Pablo Emilio Perez-Mallaina Bueno, Spain's Men of the Sea: Daily Life on the Indies
Fleets in the Sixteenth Century, trans. by Carla Rahn Phillips (Baltimore: Johns Hopkins Uni-
versity Press, 1998),233-237. The fullest treatment is Lamb, "Science by Litigation." Many of
the relevant documents are reproduced in Jose Pulido Rubio, El Piloto Mayor de la Casa de la
Contrataci6n de Sevilla: Pi/otos Mayores del Siglo XVI (Datos Biogrrificos) (Seville: Tip.
Zarzuela, Teniente Borges 7,1923),77-119, and Pulido, El Piloto Mayor (1950),482-534.
The bulk of the original documents are found in AGI, Justicia, 1146, N. 3, R. 2.
48. Many charts of the period tried to deal with this problem by introducing multiple lat-
itude scales; see D. Gernez, "Les Cartes avec Echelle de Latitudes Auxiliaire Pour la Region
de Terre-Neuve," Mededelingen: Academie Van Marine Van Belgie, book 6 (1952), 93-II7. On
Gutierrez's charts in particular, see Cerezo Martinez, Cartografia Nriutica Espanola, 205 - 208.
49. 10 April 1545 statement by Alonso de Chaves, AGI, Justicia, 1146, N. 3, R. 2, block 3,
image 335; see also similar statements by Pedro Mexia (image 343, undated 1545) and Pedro
de Medina (images 100- lOI, 3 September 1544)·
50. Chaves, Espejo de Navegantes, IIO.
51. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 339, Alonso de Chaves, 10 April 1545·
52. AGI, Justicia, 1146, N. 3, R. 2, block 3, images IOO-IOI, Pedro de Medina, 3 Septem-
ber 1544.
53. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 224 (Francisco del Barrio); 149 (Alonso
Martin), 165 (Cristobal Cerezo de Padilla), 170 (Alonso Perez), 181 (Francisco Guzado), and
many others.
54. AGI, Justicia, 1146, N. 3, R. 2, block I, image 8, 24 April 1545 petition from Pedro de
Medina; block 3, image 338, Alonso de Chaves, IO April 1545.
55. AGI, Patronato, 251, R. 22, 2 August 1527 cedula appointing Col6n as acting pilot
major, to be assisted by Alonso de Chaves and Diego Ribero.
56. See AGI, Indiferente, 2005, 1534 investigation of licensing exams, especially the testi-
mony of Francisco Vanegas and Diego Ramirez, transcribed in Medina, El Veneciano I:
50 7-5 12.
57. AGI, Indiferente, 421, L. 13, f. 295V, 21 August 1528 cedula authorizing the classes.
58. For the rules on licensing exams, see AGI, Patronato, 251 R. 22; the entire cedula has
been reproduced in Pulido, El Piloto Mayor (195°),140-143. For pilots' testimony in 1544
about problems measuring latitude at sea, see below, note 67. On licensing pilots who could not
use instruments at sea, see AGI, Justicia, 945, ff. IIOV- II2r (Pedro de Medina, 17 July 1549).
59. On the Sphere itself, see The Sphere of Sacrobosco and Its Commentators, ed. Lynn
Thorndike (Chicago: University of Chicago Press, 1949); on its role in navigation texts of this
period, see Pablo Emilio Perez-Mallaina Bueno, "Los Libros de Nautica Espafioles del Siglo
XVI y su Influencia en el Descubrimiento y Conquista de los Oceanos," in Ciencia, vida y
espacio en lberoamerica (Madrid: CSIC, 1989), esp. 477-478.
60. See, Martin Fernandez de Enciso, Suma de Geographia, 1519, ed. M. Cuesta Domingo
(Madrid: Museo Naval, 1987), 1I3-II8 (tables ff. 87-110); Falero, Tratado del Esphera part 2,
chs. 5-6; Chaves, Espejo de Navegantes, 149- 160 (book 2, treatise I, chapters 1-3).
61. AGI, Justicia, 768, N. 2; AGI, Indiferente, 1967, L. 16, ff. 3I2r-3I3f.
62. AGI, Justicia, I146, N. 3, R. 2, block 2, image 38 (Alonso de Chaves, 9 September
1544); block 2, image 380 (Pedro Mexia, undated January 1545); block 3, f. 20r (Pedro de Med-
ina, 3 September 1544)·
63· AGI, Justicia, 945, section 4, f. 70rv (Hernando Bias, 27 May 1549); ff. I03r-I04r (San-
cho Gutierrez, I I July 1549); f. II2r (Pedro de Medina, 17 July 1549). This was part of an
inspection of the Casa de la Contrataci6n by the Council of the Indies. Gutierrez died while
appealing the sentence .

lO6 Alison Sandman


64· AGI, Justicia, 945, section 4, f. II2r, 17 July 1549 testimony.
65. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 336, Alonso de Chaves, IO April 1545.
66. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 371, undated statement signed by about
fifty-fiv e pilots; AGI, Justicia, 792, N. 4, especially image 74, 1565 petition from the pilots
guild to have a certain lodestone made available for remagnetizing their compasses.
67. AGI, Justicia, 1146, N. 3, R. 2, block 3, image 186 (testimony from Martin Lopez),
image 203 (Pedro Camino), image 212 (Martin Sanchez), image 218 (Diego Sanchez
Colchero), image 223 (Francisco del Barrio), image 234 (Tome de la Isla), image 250 (Hernando
Bias), image 371 (statement from fifty-five pilots). All the witnesses cited here were pilots.
68. For being unable to use single-scale charts see the statements in AGI, Justicia, II46, N. 3,
R. 2, block 3, image 244 (Jeronimo Rodriguez), images 355-356 (six experienced pilots),
image 371 (about fifty-five pilots). For one pilot who could use both types of chart, see image
347 (Hernan Rodriguez).
69. See for example AGI, Justicia, 1146, N. 3, R. 2, block 3, images 339-340, description
by Alonso de Chaves.
70. AGI, Justicia, II46, N. 3, R. 2, block 2, image 33 (Diego Gutierrez, 28 August 1544);
block 3, image 119 (Sebastian Cabot, 9 September 1544)·
71. This charge was brought by Medina, and substantiated by the long-standing use of the
two-scale charts even after they were banned; AGI, Justicia, 945, ff. 109v-IIOV, 17 July 1549
testimony.
72. AGI, Indiferente, I093, N. 68, 5 April 1545 letter from the officials of the Casa de la
Contratacion to the prince.
73. AGI, Indiferente, 1963, L. 9, ff. 174v-176v, 22 February 1545 letter to the officials of
the Casa de la Contrataci6n.
74. AGI, Justicia, 1146, N. 3, R. 2, block 3, images 356-357, undated ca. 1544 petition from
six pilots.
75. Antonio Rumeu de Armas, El Tratado de Tordesillas, Colecciones MAPFRE 1492
Colecci6n America 92,12 (Madrid: Editorial MAPFRE, 1992), 141-150, 169-175.
76. The most notable skirmish during this period was probably the arrest of eleven Por-
tuguese sailors accused of being on the wrong side of the line of demarcation, which sparked
official inquiries into the location of the line. Rolando A. Laguarda Trias, El Predescubrim-
iento del Rio de la Plata por la Expedici6n Portuguesa de 1511-1512 (Lisbon: Junta de Investi-
gac;6es do Ultramar, 1973),91-113, esp. 96-97.
77. On Magellan, see Tim Joyner, Magellan (Camden, Maine: International Marine, 1992).
On the cartographic im portance of the debate over the Spice Islands, see Jerry Brotton, Trading
Territories: Mapping the Early Modern World (Ithaca, N.Y.: Cornell Univ. Press, 1998), II9-150.
78. For a brief description of the meetings, see Rumeu de Armas, El Tratado de Tordesillas,
222-224. He blamed the lack of results on the obstructionism of the Portuguese representa-
tives. Most of the expert opinions can be found in AGI, Patronato, 48, Ramos 12-17, and have
been published in Martin Fernandez de Navarrete, Colecci6n de los Viages y Descubrimientos
que Hicieron por Mar los Espanoles Desde Fines del Siglo Xv, Vol. IV: Expediciones at Maluco =
Viage de Magallanes y de Elcano (Buenos Aires: Editorial Guarania, 1945),296-335.
79. Ursula Lamb, "The Spanish Cosmographic Juntas of the 16th Century," Terrae 1ncog-
nitae 6 (1974): 51-64, 55; Rumeu de Armas, Et Tratado de Tordesillas, 223. Rumeu de Armas
characterized the Portuguese as obstructive throughout, while Lamb argued that the political
maneuvering hindered the experts' attempts to discuss freely all available data and suggested
that the differences over the charts involved a disagreement over projection.
80. AGI, Patronato, 48, R. 13, image 5, undated 1524 opinion signed by Hernando Colon,
Fray Tomas Duran, Doctor Salay a, Pero Ruiz de Villegas, Maestro Salazar, and Juan
Sebastian del Cano.

Mirroring the World


81. AGI, Patronato, 48, R. 17, image I, 17 April 1524 opinion of Hernando Colon.
82. Luis de Albuquerque, "0 Tratado de Tordesilhas e as Dificultades Tecnicas da Sua
Aplica\;ao Rigorosa," in El Tratado de Tordesillas y Su Proyeccion (Valladolid: Seminario de
Historia de America, Universidad de Valladolid, 1973), 131-132.
83. AGI, J usticia, 1146, N. 3, R. 2, block 3, image 359, undated spring 1545 statement from
Sancho Gutierrez. Emphasis mine.
84. Unsurprisingly, they all said no. AGI, Justicia, 1146, N. 3, R. 2, block 3, images 149ff.
(question 13 and answers by pilots).
85. AGI, Justicia, 1146, N. 3, R. 2, block I, images 13-14 (5 May 1545 statement from
Francisco Falero); block 3, image 344 (undated 1545 statement from Pedro Mexia); block 3,
images 336-338 (10 April 1545 statement from Alonso de Chaves); Chaves did not mention
the demarcation directly, but focused on the errors in distances and positions caused by the
double scale.

ro8 Alison Sandman


From Blowfish to Flower
Still Life Paintings
Classification and Its Images, circa 1600

CLAUDIA SWAN

A t its broadest span, this essay is about how early modern (pre-
Linnaean) natural history organized its experience of the natural
world. More narrowly, I am conerned with the visual organization
of the natural world by means of naturalistic figuration (mimetic pictures)
and schematic representation (grids). Much of the chapter will focus on nat-
uralistic representation in the classification of blowfish in the first decade of
the seventeenth century in the Netherlands, and on grids and schematic
tables in natural history of the period. I conclude by suggesting that the
coexistence of these two modes of representation is a crucial feature of early
modern natural history and that, taken together, they may help to explain
how botanical still-life paintings are structured, compositionally and episte-
mologically speaking. In other words, this essay treats the organization of
the natural world by images and the impact of natural history's modes of
visualization on the new genre of flower pictures ca. 1600.

CHAOS IN THE MICROCOSM: THE CASE OF ULISSE ALDROVANDI

Ulisse Aldrovandi, who died in 1605 at the age of 83, ranks among the most
renowned sixteenth-century natural historians. Aldrovandi was professor of
logic and philosophy and lecturer in simples (medical preparations from
plants and minerals) at the University of Bologna. From 1568 until his death
he served as director of the university's botanical garden, which he helped
found. During his lifetime, Aldrovandi was known to European medics, nat-
uralists, princes, clerics, pharmacists, and scholars as a collector who amassed
a truly staggering number of natural specimens, which he housed in a kind of
proto-museum, and many of which were illustrated in his massive volumes
on insects, birds, wood, metals, monsters, and other classes of naturalia. 1 His
fame endured: in 1750 the great naturalist Georges-Louis Leclerc, Comte de
Buffon (1707-88), dubbed him "the most diligent and knowledgeable of nat-
uralists" and philosopherlencylopedist Denis Diderot (1713-84) referred to
/
him. as "the most universal and complete modern naturalist."2
/ Diligent, knowledgeable, universal, and complete: these adjectives were
used even in the sixteenth century to describe Aldrovandi's efforts to observe
the natural world and to make it available for observation. Buffon and Diderot,
taken together, offer an apt, if antithetic, introduction to Aldrovandi's natural
history. Unlike his exact contemporary Carl von Linne/Carolus Linnaeus
(1707-78), Buffon did not believe in the necessity of systematic taxonomy.
Diderot in his turn drove the production of that immense and powerful
machine of calibrated knowledge, the Encyclopedie. While Aldrovandi
amassed, arranged, studied, taught, and published as many specimens of the
natural world as he could get his hands on, he did not find it necessary to offer
a systematic mode of organization for them. In the sixteenth century, natural-
ists resorted regularly and without apology to what subsequent natural histo-
rians would take to be arbitrary and convenient modes of organization. 3 The
alphabet, for example, was sufficient for Leonhart Fuchs, one of the three so-
called fathers of German botany and the author of a suite of volumes on the
plant world first published in the 1540s. Generally speaking, very broad mor-
phological or Aristotelian classes served as the brackets between which six-
teenth- and seventeenth-century naturalists arranged the stuff of nature,
which they so eagerly tracked down, studied, observed, dissected, dried,
bought, sold, taught, published, displayed, advertised. 4
In 1595, Aldrovandi described his collection as follows:

Today in my microcosm, you can see more than 18,000 different things,
among which 7000 in fifteen volumes, dried and pasted, 3000 of which I
had painted as if alive ("al vivo"). The rest-animals terrestrial, aerial and
aquatic, and other subterranean things such as earths, petrified sap, stones,
marbles, rocks, and metals-amount to as many pieces again. I have had
paintings made of a further 5000 natural objects-such as plants, various
sorts of animals, and stones-some of which have been made into wood-
cuts. These can be seen in fourteen cupboards, which I call the Pina-
cotheca. I also have sixty-six armoires, divided into 4500 pigeonholes,
where there are 7000 things from beneath the earth, together with various
fruits, gums, and other very beautiful things from the Indies, marked with
their names, so that they can be found (emphasis added).5

Contemporary observers were regularly stupefied by the contents of


Aldrovandi's collection-his microcosm, as he calls it here. One visitor wrote
of his heart being aflutter and his breath bated in anticipation of seeing all

IIO Claudia Swan


that Aldrovandi had amassed. 6 In the final years of his life, Aldrovandi
arranged to have ownership of what he even called the "eighth wonder of
the world" transferred to the city of Bologna, a gift accepted gladly by the
city's senators. 7 Aldrovandi's efforts are exemplary of what Paula Findlen
has called Renaissance curiosity (as opposed to Baroque wonder), of the
efforts to contain the infinite manifestations of nature in a single space-
also called a museum. 8 Curiosity, however, did not always sponsor recogniz-
able modes of classification.
One organizing principle for Aldrovandi's efforts, as for his possessions,
is to be found in the praxis of medicine. Some of the most noteworthy con-
tents of his museum and others of the time had or were thought to have
medicinal application-from bezoar stones to dragon skeletons to myriad
plants and spices. But if his own notes are any indication, to enter the space
of Aldrovandi's microcosm was to give way to what seems from the present
perspective of the life sciences relative chaos-chaos in pigeonholes and
armoires. Likewise, the headings under which he collated his working notes
were distinctly rudimentary from the point of view, say, of systematic taxon-
omy. They were arranged alphabetically, topically, and geographically. The
staggering number of images of the contents of the collection he had made
and that he stored in the collection itself is important too. Not because these
images contributed to the classification of the contents, but because the net
effect of their presence would have been to mirror, and to multiply, the vast
number of items at hand. It comes as something of a relief when Aldrovandi
tells us that the forty-five hundred pigeonholes in the sixty-six armoires in
which he had placed seven thousand dried specimens were "marked with
their names, so that they can be found."

THE ENDS OF NATURALISM: BLOWFISH

In the brief description of his collection cited above, Aldrovandi enumerates


the specimens he owns in the same breath as images of such objects, which
in their turn constitute a substantial portion of the whole. The images
Aldrovandi cites are continuous with the rest of the collection in a signifi-
cant regard: they are, as we know from the woodcuts made for publication
in the volumes he authored, naturalistic representations of natural speci-
mens. They were, as he specifies here and elsewhere, drawn or painted al
vivo. 9 What role do such images - purposefully morphological characteri-
zations-play in contemporary classificatory schemes and efforts?
The blowfish and, specifically, the blowfish as it came to be represented in
the context of Dutch natural history ca. 1600 may offer an answer to this
question. The blowfish was a widely popular specimen in natural history ca.

From Blowfish to Flower Still Life Paintings III


L'%~UW c../ Jacques de Gheyn 11, Blowjish, pen and ink drawing, ca. 1605
(Rijksmuseum, Amsterdam). Photo courtesy of the Rijksmuseum-Stichting, Amsterdam.

1600, prized for its exoticism.1° On the evidence of contemporary prints we


know blowfish to have been displayed prominently in a number of illustri-
ous collections, where they frequently were hung from ceilings." A single
specimen was also depicted by the Dutch artist Jacques de Gheyn I1 (1565-
1629). De Gheyn cultivated close contacts with members of the medical fac-
ulty at Leiden University during the years that he lived in the city of Leiden
and his drawing of a blowfish is coeval with published accounts by the great
botanist and fellow Leidenaar Carolus Clusius (1526-1609) of what Clusius
declared to be four different kinds of blowfish. Close analysis of de Gheyn's
and Clusius's verbal and pictorial descriptions of the blowfish is especially
revealing-of the difference between an artistic representation of a natural
historical specimen and a natural historical representation of the same, as
well as of the ways in which naturalistic representation served the ends of
contemporary classification strategies.
De Gheyn's drawing of a blowfish (Diodon hystrix; fig. +I), preserved in
its inflated form, is dated on stylistic grounds to the first decade of the sev-
enteenth century. It consists of two views of the dead fish, frontal and lateral,
and a lengthy inscription. In translation, the inscription reads:

I IZ Claudia Swan
Sea-Hedgehog-this fish is umber white and iron black grayish/ becom-
ing lighter from the back down to the belly/ this is white toward the tail it
is even browner and spotted/ with Cologne earth the spines are yellow
ochreish light gray/ the fins are umber and Cologne earthish against the
body yellow ochre/ and white somewhat red and also somewhat blueish in
color/ and also spotted with Cologne earth and the jaw/ rather umber-like
in color.12

De Gheyn may have observed this fish in a private collection, or among the
natural historical specimens belonging to Leiden University. In 1601, he was
commissioned by Dr. Pieter Pauw, professor of anatomy and botany, to
engrave a plan of the botanical garden (fig. 4.2).13De Gheyn's is a schematic
plan, intended to describe the layout of the garden. It contains individual
plots in which botanical specimens were planted and from which their iden-
tification and uses were taught to students of materia medica- the makings
of medicines. This practice is illustrated by the figure of the robed professor,

4.2 Jacques de Gheyn 11, Leiden University Hortus Publicus, engraving,


1601. Photo courtesy of National Herbarium of the Netherlands, Leiden.

From Blow$sh to Flower Still Life Paintings 1'3


surrounded by an attentive audience, in the distant center of the image; he
points to one of the plots in the garden, in much the same way as we know
the medical faculty to have taught from the contents of the garden in the
years around 1600. 14 In effect, de Gheyn's plan of the garden records its
physical characteristics as well as its function within the university curricu-
lum; like the anatomical theater, constructed in the mid-1590S, the univer-
sity's collection of naturalia served as a focal point in medical instruction.
In the gallery at the far, western edge of the garden, which was con-
structed in 1599, specimens of a variety of naturalia were housed. A carcass
of a blowfish is listed in the two existing inventories of the gallery, alongside
a number of rarities including coral, an imbricated shell, the beak of a for-
eign bird, an Indian ink pot, a crab, pygmy clothing, two Indian hammocks,
walrus teeth, and the like. IS That the blowfish was a prized item in the col-
lections of the Leiden University is also shown by its inclusion in another
plan of the garden, by William Swanenburgh and dated 1610 (fig. 4.3), by
which time the garden and the anatomical theater had become tourist attrac-
tions. 16 The blowfish was also recorded in a major natural history text writ-
ten by a close associate of de Gheyn's and the director of the university
garden-Clusius's Exoticorum Libri Decem (Ten books of exotica), a vast

HORTI I'\fBLlCI /\1:.1\ lJ£.MJA.

S3iffUYe IJ.,J Will em Swanenburgh after I.e. Woudanus, Leiden University Hortus
Publicus, engra ving, 16 I o. Photo courtesy of Rijk smuseum-Stichting, Amsterdam.

Claudia Swan
compendium of wonders of the natural world Clusius compiled in Leiden
and published there in 1605.17 That de Gheyn would have had access to a
specimen of the Diodon hystrix in Leiden seems clear; moreover, the status of
this fish as an object of curiosity at the time is amply documented. ls How de
Gheyn views the blowfish and the modes of verbal and visual description he
engages to describe it are what is crucial here.
De Gheyn scanned the surface of the prickly orb of fish before him, first
from in front of it and in line with its line of sight and then, in the render-
ing at the right of the sheet, from its right side. Finally, he scanned it again
in order to record, in the inscription below, the shifting colors of its body.
The continuity between the abbreviated and spiky forms of the drawing and
the forms of the text below is suggestive, and the pace with which he
describes the alterations of color from the back to the belly and from the fins
to the body (the text is entirely without punctuation) is consistent with the
sustained pattern of lines defining the fish above. Significantly, the descrip-
tive responsibilities of the verbal and the visual accounts de Gheyn provides
are distinct: the drawing of the fish conveys the form, and the inscription the
color. De Gheyn's written description does not reiterate what is made visible
in the accompanying image; it supplements it. And by describing the colors
of the fish in the colors of pigments (Cologne earth, umber white, iron black
grayish, yellow ochreish light gray) de Gheyn describes the fish as a picture.
De Gheyn's drawing offers a verbal description that is functionally dis-
tinct from the visual description it provides. His verbal description is of a
different order - the order of color, of a painter's colors. The inscription
does not allude to an external frame of reference beyond the palette, and in
this sense the description de Gheyn proposes is entirely self-referential:
"This fish" to which de Gheyn refers in the opening line of the inscription is
no longer the fish hanging in the Leiden gallery, or the fish returned to the
Netherlands after long voyages, the component of a collection of naturalia,
or this fish as compared to any similar or other fish. "This fish" is the fish of
de Gheyn's drawing, the fish of a picture in the making.
Reconstituted according to de Gheyn's indications-colored in-such a
picture would effectively convey the forms and colors of this exotic natural
specimen. To a very significant extent these were the characteristics according
to which distinctions of class and sort were made within contemporary nat-
ural history.19 That this is so is borne out in Clusius's accounts of the natural
world in general, and in his description of the Diodon hystrix in particular,
which is directly relevant to assessing the status of de Gheyn's description. In
his voluminous Exoticorum, Clusius documents the four fish he identifies as
blowfish in the course of four separate chapters of book 6. Of the first three
of the fish he describes, Clusius writes that he had observed them hanging in
"museums" belonging to individuals in Amsterdam and in Montpellier. 20

From Blowfish to Flower Still Life Paintings


Each of Clusius's entries on these four fish - what he calls the Histrix piscis
and three related fish, which, following Guillaume Rondelet, the French
icthyologist, he names the Orbis spinosus, the Orbis muricatus, and the Orbis
muricatus alter- is accompanied by a woodcut that represents the fish
described (figs. 4-4,4-5,4.6, and 4.7).21
In his text, Clusius makes reference to drawings that served him vari-
ously in the process of classifying the specimens he describes. One acquain-
tance, Jacob Plateau, donated drawings of two of the fish; 22 the Leiden
pharmacist Christian Porret is credited with having provided Clusius with
another. 23 At one point Clusius states that a drawing was made for him to
enable him to compare specimens. 24 Clusius's dependence on images in the
course of assembling his account of the Diodon hystrix is noteworthy; that
each of his entries is illustrated with a woodcut reflects a conviction, ampli-
fied in his text, that images convey information crucial to description and
some form of rudimentary classification. Within the context of late six-
teenth-century natural history, the combination of text and image here is
entirely conventional. But if Clusius's descriptive method exemplifies the
industry standard insofar as it corresponds to the manner in which verbal
and visual description are coupled throughout natural history writing of the
time, his account of these four fish also provides an excellent example of the
limits of contemporary classificatory strategies.
The degree to which Clusius relies on external, observable characteristics
in order to describe and classify the specimens he records is typical of con-
temporary natural history. Clusius's verbal description of the Histrix piscis,
for example (fig. 4-4), amounts to a meditation on the impenetrable surface
of the spiny fish. Its dimensions are given, measured from its shriveled lips
to the root of its tail and around its center, and then the specimen is, as it
were, fleshed out by a description that dwells on its superficial characteris-
tics. He seems to write as he scans the object: it is "without scales and cov-
ered merely with a whitish skin or hide, strewn with firm and sharp spines
on all sides." A description of its somewhat protuberant mouth and wrin-
kled lips and teeth follows, with conjectures about the predatory techniques
of the fish; from the eyes, with their raised eyebrows and four prickly spines,
Clusius moves along the body. The dimensions of the fins and the coloration
of the body are noted ("the skin of the belly is white, and the back is dark,
with many distinct dark spots") and, finally, the orifice through which the
fish is thought to breathe and the differing inclinations of its spines are
recorded .25 Ultimately, he informs the reader that he is unable to discuss the
internal structure of the fish, which was not native to local waters, because it
was available to him only in dried form. 26
If on the one hand the lack oflive blowfish in the Netherlands compounded
their exoticism, it also made for highly unstable classificationY Depending on

II6 Claudia Swan


. ERATpcmdab~emcmoow aa aoarpmam,llrc r~diocm,ri@bachrm
medio~ s a m b i t a v i g i n tnarcm
i mffus,fquamis a r m s , & cute tlvecorio a l k f ~
te dummxat t&s, uadiqua oc firmis & acuris fp~nisobfitus,quarum bafis in bins
1
bma B rub cute larentcs etincbat :oris apcrci diameter triurn uncirrumcnt, 8b
.
q~nntulumpromimbac, labra -6, bin34uc offi pro dentibus habeht, fupcrn2 onllm~
infcrnt alterurn, urrunlquc fomiatum, & anreriorcprcc nonnih~l ut oris riaus ;fm'

cS&+ 4.4 Anonymous, "HistrixPisicis, " woodcut in Carolus Clusius, Exoticorurn


Libri decem, Leiden, 1605. Photo courtesy of the National Herbarium of the Netherlands, Leiden.

A II emcsne ate (qaadhiambabebat, 6t biriis oLbus dentiurn loco prrdimm)


.. caa& pinrum of'que, fepccrn unciu cum fernire longuserat, decem verd cum Tern
cnfius, iden, d#um unciarurn cum fcmiffc Cht totiuscorporisambitur ;&dorfium, fi
coloris; m t a , d b i : roto anfern corporc brcvibus, fcd,firmisaculciscmtobfitos:latum
put habebat, elata fupmilia, qut mbuc aculeis muntta, & magnos oculos :quatuor f
Aphis, piiteream quzcxrrrmamuudm occu~bat.erac~radirus. u t m u c vidcli

cS&*- 4 . 5 Anonymous, "Orbisspinosus," woodcut in Carolus Clusius, Exoticorurn


Libri decem, Leiden, I 605. Photo courtesy of the National Herbarium of the Netherlands, Leiden.
ab
cut
an
bin
Tcd
adc
qu:
no:
sb i
' lor
. - 84
1 - brt
1 dorI :irm1aturn,Tip clam,& muticatisiimilitcr :
magnus uinis albis ofibw iiuc offeis hbris enr prdltus,
@(B.CCW 4.6 Anonymous, "Orbis muricatus," woodcut in Carolus Clusius, Exotico-

rum Libri decem, Leiden, 1605. Photo courtesy of the National Herbarium of the Netherlands,
Leiden.

the conditions under which they were imported, specimens were not always
intact, or they had been carelessly dried and so were deformed. To Clusius
and his contemporaries, variegation of form, however, pointed to differences
of biological sort or kind, rather than to the uncontrolled means of procuring
specimens. In fact, the four fish Clusius describes and names individually are
probably all of a single species-the Diodon hystrix. It is on the basis of exter-
nal, visible, quantifiable characteristics that Clusius isolated what he per-
ceived to be four separate kinds of fish, and his dependence on images in the
classification of varieties of the Orbis demonstrates this to a fault.
Throughout Clusius's Exoticorurn, as also in most contemporary natural
history publications, images served to complement verbal description; they
also, where they were the only available evidence, served as a basis for it, as
well as for classification. Insofar as the criteria for classification Clusius uses
are reducible to form and color, they amount to information an image can
impart. It is especially significant that, within Clusius's account of these
types of fish, the Orbis muricatus alter (fig. 4.7) is recorded as a variant solely
on the basis of an image provided by an acquaintance. It was not possible,
Clusius writes, for him to observe this particular fish, "but I received from
Jacob Plateau a colored picture of it." This image was later supplemented,

I 18 Claudia Swan
nen mcnk 3 me admaninn, hajus & fiiucntis m c n i i , longickdiacm &rmbii
fignantcnr, mittcbac.

?tscrs igitur,mjushac~iConemdamas,abexmmomdexdmracmda
dnndecim uncias e a t longus, cjus verbambitus mulm major, ut qui kdecim om
a mmfuram rrp~lcm,pa univdum cotpus muriotis fpints munitus, cotoris inda
5,& rnultisnigrismaculls confpcrfi,in vcntrc autem cincracci : fupcrcilia clam
~liqumtulurnprom~nulurn,&duobusalbisof?ibuspncditum,utin Fmrdcnte;m*
rtes ctiam introrfum habucrit ,me ktct: binas pinnas tn latcribus h a k l ~ non
r nk

4.7 Anonymous, "Orbis muricatus alter," woodcut in Carolus Clusius,


Exoticovum Libri decem, Leiden, 1605. Photo courtesy of the National Herbarium of the
Netherlands, Leiden.

Clusius notes, by information Plateau provided regarding the dimensions of


the fish.28Because the criteria for classification Clusius engages are more or
less exclusively descriptive and quantitative, images may even play a subver-
sive role: classificatory strategies were woven around images taken on faith.
That Clusius depended on images to supply him with the information
necessary for purposes of classification is evident in his works on the plant
world as well. Given Michel Foucault's arguments for the "epistemological
precedence enjoyed by botany" among the natural sciences of the classical
age, it is perhaps surprising that Clusius should have relied on images of fish
to the extent that he did. Foucault writes:

The area common to words and things constituted a much more accom-
modating, a much less 'black' grid for plants than for animals; in so far as
there are a great many constituent organs visible in a plant that are not so
in animals, taxonomic knowledge based upon immediately perceptible
variables was richer and more coherent in the botanical order than in the
zoological. . . . Because it was possible to know and to say only within a
taxonomic area of visibility, the knowledge of plants was bound to prove
more extensive than that of animals.29

From Blowfuh to Flower Still Life Paintings 119


In his works on plants, which culminated in the publication in 1601 of his
Rariorum Plantarum Historia, Clusius supplemented his descriptions of
flowers, for example, with considerations on relative scale, the time of the
year they blossom, and their provenance; but the characteristics most crucial
to their classification are those that can be observed in the immediate pres-
ence of the specimen. 30 Or, as we have seen in the case of the Orbis muricatus
alter, those that can be recorded pictorially. In his groundbreaking chapter
on tulips in the Rariorum, Clusius describes a variety of dwarf tulip within
the category of the "intermediates"; it blossoms between the "early" and the
"late" varieties. This class is described generally as follows:

The dwarf [intermediate) tulip is not more than a foot high, usually even
less, and in its leaves and flower it strongly resembles the early tulip. All
its segments are pointed, but the outer ones are much longer, externally
dull red but at the outermost margins greenish; the inner segments are of
a brilliant, fiery red throughout. The claws are yellow and radiating, but
marked with a jet-black patch in such a way that the latter appears encir-
cled by a mere golden aureole and bears some likeness to an eye; the fila-
ments and their anthers are blackish. It should be noted that its bulbous
root is woolly; the outer membrane enveloping and covering the substance
of the bulb is so tightly filled with an abundance of dense, white, soft
stuffing that it must form a very soft resting place for the bulb. 31

From the opening sentence of his description, Clusius moves the reader to imag-
ine the plant described . The dwarf tulip resembles the early tulip in its overall
appearance; and in the more specific rendering of the appearance of this flower,
we are led from part to part by gradations and shifts of color. At the center of the
plant, and of the description, we encounter in an almost specular manner (we
are looking into the tulip from above, observing the appearance of the golden
aureole) "some likeness to an eye." The sensual engagement with the object,
which culminates in an empathic description of the outer membrane of the bulb
("a very soft resting place" for it) is driven by a single organ-the eye.
One further example of the dwarf tulip is discussed by Clusius in the text
immediately following that cited above. Here again, as in the case of the
Orbis muricatus alter, Clusius incorporates this specimen into his account on
the basis of an image alone, and in the absence of actual experience of the
specimen. "Also another kind of dwarf tulips is found," he writes, "which,
however, I have not seen."

But I received a drawing in natural colors ["iconem suis coloribus expres-


sam"J of it in the year 1596 from the learned Johan de Jonge, Minister at
Middelburg, to which had been added the following description:

120 Claudia Swan


"I send you a picture ['contreftytsel'] of a certain tulip, drawn after the
plant itself, that is to say of natural size in regard to the plant as well as
to the stalk, the flower, the leaves (which should have been drawn
slightly longer and narrower) and the bulb, which I dug up in order to
enable the artist to properly draw it ... . "

The whole plant, then (as far as I have been able to gather from the draw-
ing), is not bigger than the palm of a hand, producing four narrow,
keeled leaves resembling those of the Montpellier tulip, from among
which arises a little stalk of the height of an inch or a little higher, leafless
(in contradistinction to the habit of other tulips), purplish green, and car-
rying on its top a flower consisting of six segments, externally somewhat
purplish, internally whitish, its center occupied by an oblong pistil fenced
in by six yellow little stamens .... That it has flowered in April I deduce
from the fact that my correspondent sent me the drawing by the begin-
ning of May.52

Sustained observation and morphological comparison are the means to


classification, and to the extent that these processes depend on the visual
aspects of the specimen to be classed, an image "drawn after the plant
itself' and "in natural colors" was deemed sufficient to supply the neces-
sary information.
Such images as Clusius cites and publishes mark the limits of his analysis,
which depended crucially on visually apprehensible information. The case
of Clusius is exceptional to a degree: unlike many of his contemporaries, he
had relatively little interest in the pharmaceutical properties of the plants
and other natural objects he described. In fact, it is Clusius who is most
renowned for having studied and cultivated rare and exotic species of flow-
ers-tulips, lilies, and other foreign bulbous varieties-as curiosities rather
than as remedies. This is not to say that the images he relied on were any dif-
ferent from the images his fellow "fathers of Netherlandish botany" Rem-
bertus Dodonaeus (I5I7-85) or Matthias Lobelius (I538-I6), for example,
included in their voluminous accounts of the plant world. Indeed, many of
the woodcuts in Clusius's publication were printed from woodblocks in the
possession of his publisher Cristoffel Plantin that were also used in Dodon-
aeus's and Lobelius's great herbals.33 Clusius's taxonomic efforts, though,
were driven by morphological rather than utilitarian (pharmaceutical) con-
cerns. And in this sense images could be said to playa distinct role in his
efforts, in principle if not always in fact.
There are two crucial differences between Clusius's verbal description of
the "porcupine fish" and de Gheyn's inscription on his drawing of the Zee-
Eeghel. De Gheyn shows no concern with the dimensions of the fish or its

From Blowfish to Flower Still Life Paintings 121


origins or the relation of this dried specimen to the living fish; and the
terms in which he describes its coloration are in effect painterly. It is in the
verbal, not the visual, information imparted by de Gheyn's image that we
can locate its functional prerogatives; it is only its inscription that distin-
guishes de Gheyn's image from those published by Clusius. It should be
clear from the foregoing that sixteenth-century natural history depended
on morphological description to such an extent that it would have allowed
for the assimilation of precisely this kind of image for scientific ends. Noth-
ing is intrinsically scientific about either de Gheyn's or any of Clusius's or
his acquaintances' images; they are capable of being impressed into service
of a scientific kind. They become the labels, in a sense, behind which names
may be stored.
This circular tale is intended to call attention to the ways in which natu-
ralistic representation served and furthered the ends of a natural history con-
cerned, as Foucault put it, with "the nomination of the visible."34 Elsewhere,
I have written at some length about the role of verifiably naturalistic
images-images that could stand in for what they represent-in natural
history of this period. is I want now to call attention to a different form of
visual representation common, not to say integral, to this natural history.
This is the grid, the schematic, rectangular representation that so very fre-
quently occurs in the context of the practice and publication of natural his-
tory in the sixteenth and early seventeenth centuries.

TABULATION

It is difficult, when examining the relations between visual representation


and the praxis of natural history in the early modern period, to overlook the
grid and its affiliate, the tabular diagram.if> Examples abound. Matthias
Lobelius's 1581 volume on plants, one of the most renowned botanical pub-
lications of its time, concludes with a striking section, "Vande Succedanea,"
which consists principally of a series of nearly twenty schematic grids (fig.
4.8).17 Lobelius's herbal shares the distinction, with the publications of his
contemporaries Dodonaeus and Clusius, of being copiously illustrated-
each of the roughly thirteen hundred pages of text contains at least two
woodcuts. The grids in the concluding section of the book come as some-
thing of a surprise, given the predominance of naturalistic images overall.
The title of this last section of Lobelius's voluminous publication specifies
that the individual tables illustrate which dried substances - herbs, roots,
flowers, seeds, resins, gums, stones, woods-may be substituted for others
for medicinal ends. Lobelius's grids or "tables," as he calls them, offer sug-
gestions for the organization of dried specimens in what Aldrovandi called

122 Claudia Swan


8.8 Anonymous, woodcut in Mathias Lobelius, Kruydtboeck, Antwerp,
1581. Photo courtesy of the National Herbarium of the Netherlands, Leiden.
"pigeonholes"-that is, compartments or drawers in larger pieces of furni-
ture. The pigeonholes schematically represented in the Lobelius woodcuts
were intended to correspond to the drawers of a pharmacist's cabinet; the
scheme according to which their contents are ordered is a functional-and
distinctly pharmaceutical- one.
The kinds of specimens itemized in these woodcuts were, as we have seen
in both Aldrovandi's case and the case of the Leiden botanical garden, col-
lected and studied in the immediate proximity of living specimens - plants
in particular. The basic units of the Leiden University garden (figs. 4. 2 , 4.3)
are similar to those of numerous early modern academic gardens: multiple
individual plots, which in Leiden were arranged in larger rectangular beds.
They are itemized and numbered in the plot at the lower left of de Gheyn's
plan and in the plot at the far lower left of the upper left quadrant (fig. 4-2).
Each small plot of the Leiden garden-in 1594, there were fourteen hun-
dred - contained five specimens at most, and generally speaking one or
twO. 38 None of those specimens, incidentally, is represented in de Gheyn's
plan of the garden. This is particularly interesting in light of the fact that
this engraving was produced in conjunction with Professor Pieter Pauw's
publication, in 1601, of a catalog of the "Hortus publicus" or public garden. 39
Pauw's is a strange catalog, for it consists of a text preface followed by pages
and pages of sets of rectangular boxes (fig. 4.9).40 Pauw explains in the pref-
ace that students of plants were to adapt the catalog to their own experi-
ence-to take it with them to the garden and to fill in the rectangles with
the names of the plants growing in the rectangular plots of the garden. The
space of phytographic experience was, indeed, the rectangular grid. Active
phytographs translated their experience of the plots de Gheyn represents in
bird's-eye view into the spatially coordinated charts, or tables, of the plants
the plots contained. This translation corresponds more or less directly to the
way in which the contents of Lobelius's herbal- a series of individual, nat-
uralistic descriptions of plants - are staged against the pigeonholes of the
schematic cabinet.
The assemblages of medicinal, protobotanical, zoological, icthyological,
ethnographic, mineralogical data such as we know Aldrovandi to have cul-
tivated, and medical professionals throughout Europe in the sixteenth and
seventeenth centuries to have studied, were consistently structured and rep-
resented by way of the grid. The great and widely traveled Dutch doctor
Bernardus Paludanus (1550- 1633), for example, amassed a collection of nat-
uralia (plant, animal, and mineral specimens) and artificialia (primarily
ethnographic specimens) at the end of sixteenth century that, although holed
away in the northern port town of Enkhuizen, was renowned throughout
Europe. 41 In a series of brilliant protocapitalist moves he sold, reconstructed,

124 Claudia Swan


&$- 4.9 Pieter Pauw, Hortus Publicus, Leiden, 1601, pp. 52-53. Photo courtesy of
the National Herbarium of the Netherlands, Leiden.

and sold again his curiosities, dried animals, minerals, fossils, plants, tusks,
and so forth. His collection was the focus of intense admiration on the part
of foreigners and native Hollanders alike; hardly a naturalist of the time
failed to mention, let alone pay homage to or visit, Paludanus's collection,
and the total number of visitors is in the thousands. An elaborate description
of Paludanus's collection was compiled by Friedrich of Wurtemberg (1557-
1608), one of the doctor's many distinguished guests. The future duke, who
was at the time of his visit expanding his own Wunderkammer in France,
published a catalog of the contents of the Enkhuizen collection in 1 6 o ~ . ~ *
Notably, it consists of several pages of grids, in which the contents of the col-
lection are itemized (fig. 4.10). These grids seem to correspond to the actual
storage of the specimens in the collection, in cabinets or drawers; in all like-
lihood, Friedrich transcribed the collection in this manner in order to trans-
pose it to his own.
The reason for the widespread use of grids may well be directly related to
the problem evident in Aldrovandi's description of his own collection:

From Blotvfsh to Flower Still Life Paintings '25


- I... -
-
-,
- -
.
I. -
d + n M * r a pap.. Fiw. -d
'?.I

L!&~*IW 4. fo WarhaffteBeschreibung zweyer Raisung, Tiibingen, 1603 photo courtesy


of Amsterdam University Library (ZKW).
Where to put it all? How to pigeonhole the vast ranges of data and speci-
mens that were accumulated? The grid asserts no necessary genetic or
relational order between the things it organizes and, crucially, it is infi-
nitely expansive. Nonhierarchical and nonchronological, the grid allowed
for precisely the kind of serial differentiation that drove natural history at
this moment in its development.

Generally speaking, the death of the naturalist Conrad Gessner (1516-65) is


lamented as untimely and said to have deprived contemporary natural history
of one of its greatest agentS. 43 In a sense, his death by plague was productive,
for it encouraged a posthumous production-edited by friends and hangers-
on-that invokes and represents Gessner's efforts in telling ways. In 1587,
Caspar Wolf (1532- 1601), a student of Gessner's and his successor as munici-
pal physician in Zurich, published a handbook under Gessner's name, called
De Stirpium Collectione Tabulae tum Generales, tum per duodecim menses . .. , or
"The general and annual tables for the collection of plants."44 This small vol-
ume consists entirely of a series of lists, called tables. Page after page contains
a typographically sparse accumulation of names and various qualities of
plants. Wolf's volume is divided into four parts, each of which amounts to a
different organizational grid, mapped onto the plant world. The first part,
called "The General Table of Plants," contains enumerations of the sorts and
parts of plants, which serve as means for differentiating them. 45 The qualities
surveyed range from "Substance," which covers both the type of plant (tree,
herb, fruit, legume, and so on) and its "constitution" (hard, soft, dense, frag-
ile) to "Quantity," and "Qualities of the Object," to "Location" (where it
grows), and, finally, "Virtues" and "Uses." The remaining sections of the
book are: "On the Collection of Plants in General," which provides general
instructions on when to pick and how to dry plants and seeds;46 "The First
Table of Plants, Flowers and Fruits, enumerated in alphabetical order," a list
of plants, by Latin name, with indications of when the plants flower, bear
fruit, and seed;47 and "The Second Table, containing plant names in German
and in Latin," which is organized according to the months of the year, and in
which the plants are listed according to when they flower, bear fruit, and
seed. 48 What do these "tables" add up to? Generally speaking, they provide
evidence of the extent to which the tabular model- the serial categorization
of entities - structured the experience of the natural world, for Gessner as
for Wolf. In the final section of Wolf's handbook, one rudimentary scheme of
classification (Tempus, or when plants mature) is crossed with another-the
alphabet. That overlay of schemes is indicative, of the consistency with which
use (and specifically medicinal or pharmaceutical application) structures nat-
ural historical experience. Wolfs preface to this volume contains the follow-
ing specification for "The Use of these Tables":

From Blowfish to Flower Still Life Paintings 12 7


These tables will be useful not only for those concerned with pharmacy, or
for apprentices to this science, but for all those who have an interest in the
study of plants. All those who enjoy, be it winter, summer, or fall, going
out in the countryside, following partly their own impulse toward knowl-
edge, and partly driven by the necessity of taking a break and exercising
their own bodies-let them take the opportunity, thanks to these tables,
to go looking for plants, those they know, and to hope to find new ones. 49

Here, as in the case ofPauw's interactive catalog of the Leiden garden, a tab-
ular model or grid structures experience of the natural world.
Very much remains to be said about early modern classification. Gener-
ally speaking, the period under discussion, the later sixteenth century, is just
about the time in which a medicinal, pharmaceutical, use-oriented botany
gives way to a more purely morphological botany. As mentioned earlier,
Carolus Clusius is often cited as a primary agent of this shift. What is
implicit in such claims for a transition from a more or less alchemical rela-
tion to the natural world -What are its intrinsic properties? What can it do
for me?-to a pre-Linnaean move to systematically account for the natural
world, is that the transition brings order with it. One of the aims of this arti-
cle has been to demonstrate that a certain order-visually represented by
the grid - was already operative; recuperating it and reconstructing its
applications are crucial in understanding early modern experience, and rep-
resentations, of the natural world. The coordinates of this order might be
reduced to serial differentiation; above all the grid allows for comparison of
specimens, which have been extrapolated from their "native" contexts and
offered up to the language of Gessner and Wolf, for example. Are such spec-
imens soft? Hard? Oily? Dry? Two-part? Three-part? Trees? Legumes?
The grid or tabular model of organization is artificial and schematic, but
not necessarily hierarchical. It is, rather, serial. What is perhaps most
remarkable about this mode of natural historical observation and schemati-
zation is that it is fundamentally nonconclusive. Just because the vanilla
bean is long, brown, and contains a certain number of seeds does not legiti-
mate placing it in any privileged relation to other beans, seeds, or foreign
plants. The market will do that. This brings us to a class of images I want to
adduce by way of conclusion. More or less contemporaneously with the
blowfish and the grids discussed here, a "new" genre of painting emerged
full force in northern Europe. The first decade of the seventeenth century
saw the production of a significant number of these paintings, which con-
form in technique and subject matter, and had not previously been widely
produced or generally marketable. By the mid-seventeenth century, they
were countless. Jacques de Gheyn II himself is credited with having painted
one of the first three . This "new" genre is the flower still-life painting which

I28 Claudia Swan


c.S3tf1~pJ'e 11.11 Jacq ues
de Gheyn II, Flowers
in a vase and small
animals, watercolor
and gouache on
vellum, 1600. Photo
courtesy of the Fondation
Custodia (Coli. F. Lugt),
Institut Neerlandais, Paris.

typically, in the early years, consists of numerous flowers gathered together


in isolation, and offered statically to the viewer's gaze. 50
The flower still-life attests to sustained interest in naturalistic representa-
tion of the blowfish kind; each painstakingly rendered specimen stands in
for its real counterpart and, taken together, they come to be referred to as
microcosmic representations of gardens (see fig. 4. I I). Simultaneously, it
makes sense to think about the structure of these pictures, and of their view-
ing, as pertaining to the tabular model. Consider, for example, that many of
the flower still-life paintings, or botanical portraits, produced in the early
seventeenth century consist of vases filled with more stems than could read-
ily be fit into the vases and that the flowers represented are more often than
not shown blossoming simultaneously, whereas in fact, or in nature, they do
not. The counternaturalistic impact of these pictures is crucial. If these
paintings were - initially at least, in the first decades of the seventeenth cen-
tury - painted and collected as "botanical portraits,"51 and served to record
and preserve the appearance of the individual specimens so carefully
arranged, to what degree does the sequential, nonnarrative structure of such

From Blowfish to Flower Still Life Paintings 129


pictures bear comparison with the grids and tables of natural historical expe-
rience? Conventional art historical interpretations have read these images-
the flower still-life paintings of the early seventeenth century -as allegories
of vanity and the brevity of life. 52 By way of a counterproposal, I want to cite
a passage from a philosophical dialogue written in the 1580s by the great neo-
Stoic Justus Lipsius (1547-1606), which goes some way in suggesting that
attempting to recuperate modes of experience might be more productive than
insisting on emblematic or allegorical readings of paintings of the natural
world. And that the apparently disjunctive range of scientific representation
I have cited-from naturalistic renderings of blowfish and tulips to
schematic diagrams of gardens-are not only compatible, but inseparable.
This passage is from Lipsius'sDe Constantia, a dialogue that takes place in
a garden, as befits a good Erasmian encounter. 53 The garden of De Constan-
tia may well be fictional, but the story of Lipsius's actual gardens is entirely
relevant to the foregoing. Lipsius, from the city of Louvain in the southern
Netherlands, was professor at the Leiden University from 1578 until 1591.54
When Lipsius left Leiden, he left two gardens behind, one of which was
impressed into immediate service as the university teaching garden before
the plots of that garden were dug in 1594, and the official garden opened. 55
Lipsius writes:

Observe for me these numerous flowers/ how they grow: how these are
brought out of their sheaths/ those out of their buds/ see how this one dies
suddenly and falls down! and another one grows on its stem. Finally/ see
how one sort of flower is distinguished from and can be compared to/
thousands of others/ solely on the basis of its form/ color, and appear-
ance/.... Now/ bring your scrupulous eyes here/ and regard for a
moment this sheen and the beautiful colors. See/ how this flower [is] a
beautiful purple in color! this one blood red/ this snow white/ this one
like a flame/ this shines like gold ... even the very best painter cannot
possibly replicate them ... Would that God would allow me to live
peaceably among these treasures .... among these flowers of the known
and the new unknown world. 56

For Lipsius, the garden is a spectacle, a source of wonder. He encourages


"scrupulous eyes" to follow the forms of the individual plants as they change
over time and as they vary from one to the next. The "treasures ... of the
known and new unknown world" offer their incomparable colors and sheen
for careful, sustained observation. The terms of Lipsius's description are
fairly pat; where he lapses into simile it is to compare the coloring of the
flowers to gold or to fire - the visible properties of these treasures exceed
the products of humans, and of the painter in particular. At the same time,

Claudia Swan
he invokes specific patterns of observation when he states that each sort of
flower can be distinguished from and compared to others on the basis of its
visually apprehensible, external characteristics. It is this comparative mor-
phology that motivates the use of naturalistic representations (which will
strive to equal the "sheen and beautiful colors" Lipsius records) and, at the
same time, drives the tabulation of the natural world.

Many thanks to Pamela Smith, Paula Findlen, and Peter Reill for the invitation to present
this material at the colloquium Commerce and the Representation of Nature in Early Modern
Europe (UCLA, October 1999), and for the helpful commentary they and many other partic-
ipants offered. I am also grateful to Mary Fissell and other members of the History of Medi-
cine, Science, and Technology Colloquium at Johns Hopkins University, for offering both the
opportunity to present these materials and a variety of productive responses; to the Depart-
ment of Geography at the Pennsylvania State University; to the History and Philosophy of
Science program at Northwestern University; to Cees Lut, Librarian, National Herbarium of
the Netherlands, Leiden; to Carla Teune, Hortulana, Hortus Botanicus, Leiden; and to
Londa Schiebinger, Amy Greenberg, Rich Doyle, Peter Parshall, Roelof van Gelder, Florike
Egmond, Paolo Bernardini, and David Freedberg.

I. See, principally, Sandra Tugnoli Pattaro, Metodo e sistema delle scienze nel pensiero di

Ulisse Aldrovandi (Bologna: Cooperativa Libraria Universitaria Editrice Bologna, 1981);


Giuseppe Olmi, L'inventario del mondo. Catalogazione della natura e luoghi del sapere nella
prima eta moderna (Bologna: II Mulino, 1992); Paula Findlen, Possessing Nature. Museums, Col-
lecting, and Scientific Culture in Early Modern Italy (Berkeley and Los Angeles: University of
California Press, 1994).
2. As quoted by Giuseppe Olmi, "Arte e Natura nel Cinquecento Bolognese: Ulisse
Aldrovandi e la Raffigurazione Scientifica," in Le Arti a Bologna e in Emilia dal XVI al XVIII
secolo. Atti del XXIV Congresso Internazionale di Storia dell'Arte, ed. Andrea Emiliani, 4 vols.
(Bologna: CLUEB, 1982),4: 151-173, esp. 151.
3. See, e.g., Pattaro, Metodo e sistema, 19: "Infine, come esempio del procedimento (modus)
col quale I' Aldrovandi tento di realizzare il proprio idea Ie erudito ed enciclopedico, si puo
portare la sua Selva universale della scienze 0 Pandechion epistemonicon [unpublished]. Quest'-
opera, che fu completata nel 1589, e una sorta di dizionario, in ottantatre volumi, ove Ie
materie pili disparate sone prese in esame per ordine alfabetico con amplissimo corredo di
riferimenti e d'informazioni, e fu concepita esplicitamente dal naturalista bolognese affinche
fosse di guida a chiunque desiderasse 'sapere 0 comporre sopra qual si voglia cosa naturale 0
artificiale,' on de trovare 'a quel proposito quel che n'hanno scritti i poeti, i teologi, i legisti, i
filosofi, gli istorici.'" More generally, on early modern classification, see F. S. Bodenheimer,
"Towards the History of Zoology and Botany in the XVIth Century," in La science au seizieme
sieele. Colloque de Royaumont [957 (Paris: H ermann, 1960), 285-296; Michel Foucault, The
Order of Things. An Archaeology of the Human Sciences (New York: Vintage Books, 1970);
David Knight, Ordering the World. A History of Classifying Man (London: Burnett Books,
1981), esp. chaps. 2 and 3; Scott Atran, Cognitive Foundations of Natural History. Towards an

From Blowfish to Flower Still Life Paintings


Anthropology of Science (Cambridge and Paris: Cambridge University Press, 1990); Brian W.
Ogilvie, Observation and Experience in Early Modem Natural History (Ph.D. diss., University of
Chicago, 1997), esp. 337-343. My thanks to David Freedberg for allowing me to read his
unpublished lecture, "Naming the Visible: Art and Natural History in the Circle of Galileo"
(Munich, 1991).
4. See references cited in previous note and Agnes Arber, Herbals, Their Origin and Evolu-
tion. A Chapter in the History of Botany (1470-1670) (Cambridge: Cambridge University Press,
1986; 1st ed. 1912); Allen J. Grieco, "The Social Politics ofPre-Linnaean Botanical Classifica-
tion," I Tatti Studies 4 (1991): 131-149, esp. 139ff.
5. As quoted by Lorraine Daston and Katharine Park, Wonders and the Order of Nature,
1150-1750 (Cambridge: MIT Press, 1998), 154; see also Findlen, Possessing Nature, 17-31 .
6. Findlen, Possessing Nature, 24; the visitor was Pietro Andrea Mattioli (1501-78), one of
the most famous Renaissance doctors and botanical authors.
7. Ibid.
8. Ibid., passim.
9. See Giuseppe Olmi, "Osservazione della natura e raffigurazione in Ulisse Aldrovandi
(1522-1605)," Annali dell'lstituto storico germanico italiano in Trento 3 (1977), 105-181; Olmi,
Inventario del mondo, passim; Claudia Swan, "Ad vivum, naer het leven, from the Life: Consider-
ations on a Mode of Representation," Word and Image I I (October-December 1995): 353-372.
10. Blowfish is the common name for the porcupine fish, which is of the order Tetraodon -
tiformes, and is most visibly characterized by spiny or plate-form scales. The stomach of two
families of this order-Diodontidae (porcupine fish) among them - is highly modified such
that it can inflate to enormous sizes; hence "blowfish." Inflation is caused by ingestion of
water into a ventral diverticulum of the stomach when the fish is frightened or annoyed;
deflation occurs when the fish expels the water. Inflation by air can also occur, when the fish
is removed from the water or on death. The bodies of all porcupine fish are covered with
sharp spines, which may become erect when the fish inflates. They generally have two fused
teeth. Joseph S. Nelson, Fishes of the World (New York: J. Wiley, 1984),379-386. Blowfish are
native to the Pacific and Indian Oceans.
I I. Examples of engravings showing the blowfish hanging from the ceiling of a collection

include "The Museum of Francesco Calzolari" by Hieronymus Viscardus after 10. Bapt.
Bertonus, in Benedictus Cerutus and Andrea Chiocco, Musaeum Franc. Calceolari (Verona:
Apud Angelum Tamum, 1622); anonymous engraver, "The Museum of Ferdinando Cospi,"
in L. Legati, Museo Cospiano . . . (Bologna: Giacomo Monti, 1677). These prints are frequently
reproduced in studies of cabinets of curiosities; see, for exa mple, Ellinoor Bergvelt et aI.,
Verzamelen. Van Rariteitenkabinet tot Kunstmuseum (Heerlen: Open Universiteit, 1993), figs.
64, 76, and 82.
12. "Zee eeghell dese vis is van omber wit en swart ijser graeu achtichl van den rugghen
neerewert allichter tot den buijckl die is wit nae de staert is hij noch bruijnder hij al gestip-
pelt! met keulse aerden de penne sijn geelenoocker achtich licht graeul de vinne sij omber en
keulse aerdeachtich teghen tlijf geleoockerl en wit wat root oock wat [illegible mark] blaeu
achtich gekolleureert! ende oock met keulse aerden gestippelt omden muijll wat omber
achtiger gecollereert." Rijksmuseum, Amsterdam, inv. no. A3971 (149 x 197 mm, pen and
brown ink on gray-brown paper). I. Q. van Regteren Altena,facques de Gheyn. Three Gener-
ations,3 vols. (The Hague: M. Nijhoff, 1983), vol. 2, cat. no. 896, pI. 370; see also vol. 2: 119.
Cf. K. G. Boon, Netherlandish Drawings of the F iJteenth and Sixteenth Centuries. Catalogue of
the Dutch and Flemish Drawings in the Rijksmuseum, 2 vols. (The Hague: Govt. Pub. Office,
1978), cat. no. 242; and Jacques de Gheyn II. Drawings, exh. cat. (Rotterdam and Washington:
Museum Boymans-van Beuningen, 1986), cat. no. 83: Two Studies of a Porcupine Fish (Diodon
hystrix).

Claudia Swan
13. Jan Piet Filedt Kok and Marjolein Leesberg, The New Hollstein. Dutch and Flemish
Etchings, Engravings and Woodcuts, ca. 145(}-1700 (jacques de Gheyn), 2 vols. (Rotterdam:
Sound and Vision, 2000), no. 213. On de Gheyn in Leiden, see Florence Hopper, "Clusius'
World: The Meeting of Science and Art," in The Authentic Garden. A Symposium on Gardens,
ed. Leslie Tjon Sie Fat and Erik de Jong (Lei den: Clusius Stichting, 1991), 13-36; Claudia
Swan, Jacques de Gheyn II and the Representation of the Natural World in the Netherlands ca.
1600 (PhD. diss., Columbia University, 1997).
14. See, inter alia, Leidse Universiteit 400. Stichting en eerste bloei 1575-ca. 1650, exh. cat.
(Amsterdam: Rijksmuseum, 1975); Swan,facques de Gheyn II, chapter 5, '''t Onderwijs der cruy-
den: The Leiden University Hortus 1587- 1600." The practice of teaching materia medica from
the garden became widespread throughout Europe at this time; see Karen Meier Reeds, Botany
in Medieval and Renaissance Universities (New York and London: Garland, 1981), passim.
15. For the inventories of the collection of naturalia, see Leidse universiteit 400, cat. nos.
D24-26; Erik de Jong, Natuur en Kunst. Nederlandse tuin- en landschapsarchitectuur 1650-
1740 (Amsterdam: Thoth, 1993), "Hortus Sanitatis . De hortus botanicus en de hortus
medicus als wetenschappelijke tuin," 190-234, esp. 202ff. In an appendix de Jong provides
transcriptions of the two most important inventories of the collection housed in the ambu-
lacrum (1617 and 1659) and cross-references them to Carolus Clusius, Exoticorvm libri decem:
Quibus Animalium, Plantarum, Aromatum, aliorumque peregrinorum Fructuum historiae
describuntur .. . (Leiden: Ex Officina Plantiniana Raphelengii, 1605) and to a copy hereof
that Clusius revised by hand, presently in the Leiden Universiteitsbibliotheek (UB nr. 755
A3). A version of this essay by de Jong was previously published as "Nature and Art. The
Leiden Hortus as 'Musaeum'," in Tjon Sie Fat and de Jong, The Authentic Garden, 37-60. A
very early source, to the best of my knowledge never cited, is the catalog by P. Pauw of the
Leiden garden; see below, note 38.
16. Hollstein, vol. 29 (1984), Swanenburg(h), H. 32.
17. See above, note 15; the relevant chapters are Clusius, Exoticorum 21-24: 137-140.
18. Cf. the passage from The Tempest in which Trinculo exclaims, on finding Caliban:
"What have we here? a man or a fish? dead or alive? A fish: he smells like a fish; a very
ancient and fish-like smell; a kind of, not of the newest, Poor-John . A strange fish! Were I in
England now, as once I was, and had but this fish painted, not a holiday fool there but would
give a piece of silver: there would this monster make a man: when they will not give a doit to
relieve a lame beggar, they will lay out ten to see a dead Indian (II.ii)." That fish and other sea
creatures were put on public display in Leiden and Amsterdam is amply recorded in Clusius's
Exoticorum.
19. See esp. Atran, Cognitive Foundations ofNatural History; Ogilvie, Observation and Expe-
rience, 337- 343·
20. Clusius writes that the Histrix piscis and the Orbis spinosus were available to him in a
"museum" of a merchant in Amsterdam; see Clusius, Exoticorum, 137-138. The third kind
he describes, the Orbis muricatus, he saw in Guillaume Rondelet's "museum" in Montpellier
(139). Cf. (written of the Orbis spinosus) "Exenteratus autem erat hic piscis, quemadmodum &
alij ejusdem generis, quos istic varic£ magnitudinis apud diversos mercatores videbam: satis
enim diligentes sunt in ea urbe rerum exoticarum conquisitores, quas a nautis ex sua naviga-
tione reducibis redimere solent" (139).
21. Clusius cites book 25 of Guillaume Rondelet, L'histoire entiere des poissons ... (Lyon:
Bonhomme, 1558); Clusius, Exoticorum, 139.
22. These are the Histrix piscis (Clusius compares the specimen he describes and illustrates
to the fish in Plateau's drawing) and the Orbis muricatus alter; see Clusius, Exoticorum, 138,
140. Plateau is also m entioned in Carolus Clusius, Rariorum Plantarum Historia (Antwerp:
Ex. Off. Plantiniana, apud Ioannem Moretum, 1601), passim .

From Blowfish to Flower Still Life Paintings 133


23. Namely, a drawing of the Orbis muricatus; Clusius, Exoticorum, 139. Porret is cited
throughout the Exoticorum; see F. W. T. Hunger, Charles de I'Esduse (Carolus Clusius) Neder-
landsch Kruidkundige 1526-1609, 2 vols. (The Hague: Nijhoff, 1927; 1943), Vol. I: 268.
24. The woodcut of the Orbi, spinosus is, Clusius explains, based on a drawing that was
made in order for him to compare this specimen with the Histrix pL,cis; images of the Orbis
spinosus and the Histrix piscis face each other head-on across the binding of the volume. "Ut
autem facilills utriusque differentia observari possit, illum, pcrmittente Mercatore, idem Vol-
cardus in meam gratiam delineabat, ego vera in adposita tabella deinde exprimi curabam"
(Exoticorum, 138).
25. Clusius, Exoticorum, 138.
26. Ibid. "De internis partibus nihil pronunciare queo, quandoquidem a recens capto fuer-
ant excmptce & abjectce, & corium dumtaxat it nautis funiculorum fragmentis suffarcinatum,
ut commodius resiccarent & conservarent, mihi fuit conspectum."
27. "In quo mari captus esset hic piscis, nemo certi quidpiam pronunciare poterat," he
writes of the Orbis muricatus; Clusius, Exoticorum, 140. On p. 137, however, he identifies the
Histrix piscis as having been captured in the "American Ocean."
28. "Horvm trium subsequentium Orbium historiam adec, exacte describcre non licebit,
ut superiorum IOrbis muricatus 1, quia ipsos pisces videre mihi non contigit, sed corum icones
coloribus expressas dumtaxat accipiebam 11 Iacobo Plateau, nullis adscriptis notis, e quibus
magnitudinis corporis & ejus partium conjecturam facere possem: post aliquot tamen menses
11 me admonitus, hujus & sequentis mensuram, longitudinem & ambitum designantem, mit-
tebat" (Clusius, Exoticorum, 140).
29. Foucault, The Order of Things, 137.
30. More generally, on the representation of variable qualities of plants, see also David
Freedberg, "The Failure of Colour," Sight & Imight. lcssays on Art and Culture in Honour of
E.H. Gombrich at 85 (London: Phaidon Press, 1994),245-262.
3I. Clusius, Rariorum, lib. 2: 147; as trans. by W. Van Dijk,A Treatise on Tulips by Carolus
Clusius ofArras (Haarlem: Enschede en Zonen, 1951),5°.
32. Clusius, Rariorum, lib. 2: 148; Van Dijk, A Treatise on Tulips, 52. The letter from
Johannes cle JonghC', dated 14 May 1596 and received by Clusius in Leiden on 2 June, is in the
Leiden University Library (Cod. Vule. roI); it and seven others written to Clusius by resi-
dents of Middclburg were transcribed and published by F. W. T. I-lunger, "Acht Brieven van
Middelburgers aan Carolus Clusius," Zeeuwsch Genoof.(chap der Wetenschappen (1925),
IIO-I33; for the letter from de Jonghe, see 111-113. See also Laurens J. 1301, The Bosschaert
Dynasty. Painters offlowers andfi'uit (Leigh-on-Sea: F. Lewis, 1980), 17-18, who suggests that
the Middelburg flower painter Ambrosius Bosschaert may have painted this (lost) drawing
and another drawing sent to Clusius by another Middelburg resident in 1597.
33. See F. de Nave et al., Botany in the Low CountrieJ (End of the 15th Century-ca. 1650),
exh. cat. (Antwerp: Plantin Moretus Museum, 1993); Swan, "Ad vivum, naer het leven."
3+ Foucault, The Order of Things, 132.
35. C. Swan, "Ad vivum, naer het leven."
36. By grid I mean a rectangular diagram, divided into small rectangles. On dichotomized
and bracketed outlines, and on class logic, see W. J. Ong, Ramus, i'vfcthod, and the Decay of
Dialogue. From the Art of Dw"Otme to the Art of Reason (Cambridge, Mass.: Harvard Univer-
sity Press, 1958), esp. chaps. 8 and 9; cf. idem, "From Allegory to Diagram in the Renaissance
Mind: A Study in the Significance of the Allegorical Tableau," Journal of Aesthetics and Art
Criticism 17 (June 1959): 42)-440.
37. Kruydtboeck oft Beschrtjulnghe Van allerleye Ghewassen, Kruyderen, Hesteren, ende Ghe-
boomten (Antwerp: Christoffel Plantijn, 1581): "Vande Succedanea, dat is te seggen! van
drooghen oft cruyden die by ghebreke d'een voor <l'ander ghebruyckt worden ... " (15 pp).

134 Claudia Swan


On p. I, Lobelius complains that prior publications on medicinal simples were "sonder eenige
ordeninghe! onderscheydt oft verstant" ("lacking all order, distinctions, and judgment").
"Vande Succedanea" and its tables were reprinted in Den Leytsman ende Onderwijser del'
Medicijnen, oft ordenlljcke uytdeylinghe ende Bereydingh-boeck vande Medicamenten, eds. Pieter
van Coudenberg and Matthias Lobelius (Amsterdam: Hendrick Laurentsz., 1614).
38. See Peter Pauw, Hortvs Pvblicvs Academiae Lvgdvno-Batava:. Eivs lchnographia, Descrip-
tio, Vsus. Addito quas habet stirpium numero, & nominibus (Leiden: Ex Officina Plantiniana
apud Christopher Raphelengius, 1601) fo!' 4r. Actual evidence of what was planted-and
grew-in the garden in 1601, 1602, 1604, and later years is provided by the copies of Pauw's
catalog, which were filled in by Pauw himself and others for presentation to the trustees of
the university; several of these are in the National Herbarium of the Netherlands, Leiden.
39. Pauw, Hortvs Pvblicvs.
40. There are a total of 176 pages in the catalog. In 1603, the catalog was printed in Leiden
by Ioannes Patus (Ex. Officina Ioannis Patii, Academ. Lugduno-Bat. Typographi), in revised
edition; the page size is smaller, and there are minor changes to the text. That de Gheyn's
engraving of the garden was intended for inclusion in the book when it was first published is
clear from the marginal note in the 1601 edition that reads, "ad ea quae sequuntur, inspicienda
erit Horti ichnographia, inserta pagina" (fo!. 7v.); this marginal note does not occur in the
1603 edition.
41. See F. W. T. Hunger, "Bernardus Paludanus (Berent ten Broecke) 1550-1633. Zijn
verzamelingen en zijn werk," in Itinerario voyage ofte schipvaert van Jan Huygen van Lin-
schoten 1579-1592 IIIe deel, ed. c.P. Burger and F. W. T. Hunger (The Hague: M. Nijhoff,
1934), 249-268; H.D. Schepelern, "Naturalienkabinett oder Kunstkammer. Der Sammler
Bernhard Paludanus und sein Katalogmanusckript in der Koniglichen Bibliothek in Kopen-
hagen," Nordelbingen. Beitriige zur Kunst- und Kulturgeschichte 50 (1981): 157- 182; E.
Bergvelt and R. Kistemaker, eds., De wereld binnen handbereik. Nederlandse kunst- en rariteit-
enverzamelingen, 1585-1735 (Zwolle!Amsterdam: Waanders Uitgevers!Amsterdams His-
torisch Museum, 1992); Roelof van Gelder, "Paradijsvogels in Enkhuizen. De relatie tussen
Van Linschoten en Bernardus Paludanus," in Roelof van Gelder, Jan Parmentier, and Vibeke
Roeper, Souffrir pour Parvenir. De wereld van Jan Huygen van Linschoten (Haarlem: Uitgeverij
Arcadia, 1998),30- 50, esp. 35-41.
42. Index Rervm Omnivm Natvralivm, a Bernhardo Palvdano, Medicina: Doctore, et Civitatis
Enckhvsensis Physico experientissimo, collectarum, in Warhajfte Beschreibung Zweyer Raisen
(Tiibingen: In der Cellischen Truckerey, 1603), 46ff. (24 unnumbered pages); see Van Gelder,
"Paradijsvogels in Enkhuizen," 36-38.
43. The classic study is Hans Fischer, Conrad Gessner. Leben und Werk (Zurich: Kommis-
sionsverlag Leemann, 1966); cf. Hans Fischer, G. Petit, J. Staedtke et aI., Conrad Gessner
1516-1565. Universalgelehrter, Naturforscher, Arzt (Zurich: Orell Fussli, 1967). See also the fac-
simile edition of the watercolors by Gessner of plants, which he died before publishing; H.
Zoller and M. Steinmann, Conradi Gesneri Historia plantarum. Gesamtausgabe, 2 vols. (Zurich:
Urs GrafVerlag, 1987-1991).
44- Caspar Wolf, De Stirpivm Collectione Tabula: Tvm Generales, Tvm per Dvodecim Men-
ses, cum Germanicis nominibus, & alijs hactenus a nemine traditis, olim per Conradum Gesnerum
conscriptaeac aeditae . .. (Zurich: Ch. Froschauer, 1587); Universiteitsbibliotheek Amsterdam,
613 H 27.
45. "Conradi Gesneri de Partibvs et Differentiis Plantarvm Physica Syn-nopsis ... in tab-
ulas methodice digesta," fols. 1r.-40v.; a header runs throughout these folios, identifying
them as "Tabulae stirpivm in genere."
46. "De Collectione stirpivm in genere," fols. 4u.-55v; the page header for this section is
"De Collectione in genere."

From Blowfish to Flower Still Life Paintings


47. "Tabvla Stirpium prima, alphabetice envmerans ... ," fols. 56r.-II6r.
48. "Tabvla secunda stirpivm nomina Latina et Germanica continens, quae singulis men-
sibus aut florent aut fructum maturant ... ," fols. 1 r6v.- I4 7v.
49. "Tabvlae istae non pharmacopolis tantum, tyronibus praesertim & minus exercitatis,
vtiles sunt futurce, sed omnibus stirpium notitice studiosis. Qui cum singulis fere mensibus
vernis, aestiuis & autumnalibus, partim cognitionis, partim animum remittendi & corpus
exercendi gratia, rusticatum exire soleant, occasionem ex hisce tabulis capient, qucenam eis
plantce potissimum quaerendae aut sperandae sint."
50. The literature is vast. See especially Beatrijs Brenninkmeyer-de Rooij, Roots of Seven-
teenth-Century Flower Painting. Miniatures, Plant Books, Paintings, ed . R. E. O. Ekkart; trans.
Ruth Koenig (Leiden: Primavera Press, 1996); Paul Taylor, Dutch Flower Painting 1600-1720
(New Haven, Conn.: Yale University Press, 1995); Sam Segal, Flowers and Nature. Netherlan-
dish Flower Painting of Four Centuries (The Hague: Government Publishing Office, 1990);
Norbert Schneider, "Vom Klostergarten zur Tulpenmanie. Hinweise zur materiellen
Vorgeschichte des B1umenstillebens," in G . Langemeyer and H. A. Peters, Stilleben in Europa
(Munster: Landschaftsverband Westfallen-Lippe, 1979), 294-312.
5I. Bol, The Bosschaert Dynasty, 46; see also, on this phenomenon in France, Antoine
Schnapper, Le geant, la licorne, et la tulipe. Collections et collectionneurs dans la France du XVlIe
sihle. 1 - Histoire et histoire naturelle (Paris: Flammarion, 1985),354 and 358-360, where
Schnapper speaks of "Ies 'portraits' de fleurs ou de fruits commandes par un amateur
desireux de peerenniser Ie souvenir des pieces les plus precieuses, mais essentiellement periss-
ables, de sa collection."
52. For a recent example of such an approach, see A. Chong, W. Kloek et aI., Still-Life
Paintings from the Netherlands 1550-1720, exh. cat., trans. R. Koenig et a!. (Zwolle: Waanders
Publishers, 1999), passim. For another counterproposal, the focus of which is on the social and
economic formations that inform still life paintings, see E. A. Honig, "Making Sense of
Things: On the Motives of Dutch Still Life," Res 34 (autumn I998), 166-I83.
53. De Constantia was first published in Leiden in 1584, and was reprinted in a variety of
European languages (Dutch, French, German, English, Italian, Spanish, and Polish) in as
many as eighty editions throughout the seventeenth century. The first Dutch translation, by
Jan Moretus, was published as Twee Boecken vande Standvasticheyt, Leiden, 1584 (Over stand-
vastigheid bij algemene rampspoed, trans. and annotated P. H. Schrijvers [Baarn: Amboboeken,
1983j). The Erasmian model of the humanist's garden is set out most famously in the collo-
quy The Godly Feast (Convivium religiosum, 1522); trans. and ed. C. R. Thompson, The Collo-
quies of Erasmus (Chicago: University of Chicago Press, 1965),46-78, esp. 46-47 and 51-52.
54. In 1578, at the age of thirty, Lipsius was made a professor of history and law at Leiden
University; he and his wife lived at Leiden for thirteen years, until 1591, at which time he
returned to Flanders (Lou vain) and to Catholicism, from which he had converted to
Lutheranism when he took a post at Jena in 1572.
55· Mark Morford, "The Stoic Garden," Journal of Ga/den History 7 (1987): 151-175, esp.
165- 167.
56. Trans. mine, from German (1st German ed., 1601); bk. 2, cap. 2.

Claudia Swan
((Strange" Ideas and
((English" Knowledge
Natural Science Exchange in Elizabethan London

DEBORAH E. HARKNESS

D uring the reign of Elizabeth I, a public well- the traditional


neighborhood locus for gossip, news, and information - stood at
the crossroads of Bishop's Gate and Threadneedle Streets in Lon-
don, near the Royal Exchange. Given the colorful occupations of many area
residents, and its proximity to Bedlem hospital, one can but imagine that
the quality of gossip there was high. Within a few square blocks lived an
extraordinary assortment of characters, many of whom made hands-on
investigations into the marvelous workings of nature. When John Dee was
in London he lived right in the thick of this neighborhood, along with his
mathematics pupil Sir William Pickering, and Sir Thomas Gresham,
builder of the Royal Exchange. Several members of the Royal College of
Physicians also made Bishop's Gate their home, such as the botanically
inclined Peter Turner, and the Venetian-born Dr. Caesar Aldemare, who
had been trained at the famous medical school in Padua. Aldemare was one
of the many foreigners in Bishop's Gate, and the voices of Bedlem would
have had to be very loud to drown out the cacophony of tongues generated
by the Flemish, French, Dutch, German, Italian, and Spanish residents.
These "strangers" included many natural science practitioners: instrument
makers, surgeons, midwives, alchemists, and distillers. Neither a humanis-
tically informed natural philosophy, nor a university-taught Galenic medi-
cine, nor a hands-on skill in technology is sufficient to describe the range of
interests and activities within the metropolis of London. And so I fall back
on "science," which was used commonly in the period to describe those
things that required knowledge, but not exclusively theoretical or exclu-
sively practical knowledge. As John Securis explained in his Detection and
querimonie ofthe daily enormities and abuses co[mlmited in physick, "science is
an habite, ... [a] ready, prompt and bent disposition to do any thynge, con-
firmed and gotten by long study, exercise, and use." Securis's emphasis on

I37
study, exercise, and use defines how most Elizabethan science practitioners
interested in the natural world occupied their time and made their living. l
To locate these English and alien practitioners-many of whom did not
publish - we must turn to the minutiae that historians of science wade
through when they conduct their research: the tortured book prefaces that
introduce nearly every scientific work of the period, the roll books of the Bar-
ber-Surgeons and College of Physicians, the annals of Oxford and Cambridge
Universities, and the diaries of those few men like Dee who left written
remains of their practices. But we must also look beyond these sources to the
minutiae that historians of science seldom consult: parish registers, the
addresses on the back of state papers, petitions for patents and monopolies
addressed to the queen, probate cases, and the censuses that were intended to
account for the names and occupations of every Stranger living and working
in London. 2 Together, these documents can provide historians of science with
a richer and more textured sense of natural science practitioners in Eliza-
bethan London - a mapping of who they were, whence they came, how they
were educated, and where and with whom they lived and practiced.
While this mapping process would, in and of itself, constitute valuable
factual information for historians of early modern science, it also sheds light
on the daily practice, the economic imperatives, and even the contemporary
conception of science in the period. The evidence also provides an intriguing
glimpse into the exchange of ideas and the importance of intellectual com-
munity in Elizabethan London. We learn from the mapping process that
distinct neighborhoods of science sprang up throughout the city, some hous-
ing mainly medical practitioners, one constituting the instrument-making
center of London, and still others providing communities for chemical dis-
tillers, alchemists, compass makers, and gardeners. While we might think of
science practitioners in Elizabethan London as a few scattered individuals, it
is clear that one had only to walk down the street to St. Martin Ludgate to
have a discussion about anatomy,3 visit St. Antholin's parish to debate the
significance of comets with John Dade and Richard Forster,4 mingle with
the gardeners of St. Giles Cripplegate to learn how to propagate olive trees,5
or procure a novel remedy for sciatica from the Paracelsian apothecaries and
innovative physicians and surgeons of St. Benet Paul's Wharf. 6
Natural science neighborhoods and practitioners in London were not
exclusively English, however, but were international in composition and
outlook. It is thus important to recognize that there was no purely English
natural science in the period, but only a natural science practiced in England
by a variety of individuals both native and foreign, some university trained
and others barely literate. As early as 1571, the official census of strangers
indicated that the city of London was home to 4,850 non-native workers,
who made up approximately 4.9 percent of the total population. 7 More than

Deborah E. Harkness
75 percent of that number were from the Low Countries; the remainder
included French, Italian, Spanish, Scottish, Portugese, Danish, Greek, and
even Turkish men and women. 8 The numbers of Strangers in the city only
increased during the remainder of the century, mostly because religious and
political conflicts on the continent forced many people to seek protection in
Protestant England.
A healthy proportion of these immigrants possessed skills that brought
them to the attention of London citizens interested in the properties of
nature. The 1571 census lists twelve Strange physicians, for example, only a
few of whom were licensed to practice by the authorities. Sixteen men and
women from Burgundy, Antwerp, and Amsterdam professed to being sur-
geons, while three apothecaries-one Italian, one from Flanders, and a
woman from Holland-also worked in the English medical marketplace.
Eight professional gardeners from other countries lived in the city in 1571,
working both inside and outside the city gates in the garden plots of the
wealthier citizens. In addition, many skilled Strangers engaged in trades
that supported natural science - glassmakers who made alchemical vessels,
potters who shaped apothecary jars, and clock-makers who could craft any
number of mechanical marvels such as intricate clocks and astrolabes.
The presence of so many Strangers among London's natural science prac-
titioners suggests that a wide-ranging intellectual atmosphere pervaded the
streets and neighborhoods of the city. London provided a lucrative and vital
environment for natural science practitioners, and the sheer number of prac-
titioners who lived and worked in the City during the period was much
higher than the number who were maintained at, or even circulated
through, the royal court. 9 Yet the influx of "strange ideas" was bitterly
resented by some because alien natural science practitioners drew clients
away from citizens in the tight London market, and because their very unfa-
miliarity made them stylish to urban consumers.
English natural science practitioners complained frequently about the
inroads their alien competitors were making into London commerce. One
medical practitioner wrote of the "runners about called cutters for the
stone," Strangers who

have suche a great name at their first coming. But after ... their work be
tried and then the proof of them seen: the people for the moste parte are
wery of them .... Such is the foolish fantasyes of our English nation that
if he bee a Straunger: he shall have more favourers then an English man,
though the English mans knowledge doo far passe the others .... "10

A physician charged that a few of London's popular medical practitioners


were actually feigning alien behaviors in order to gain a clientele: "Some-

"Strange" Ideas and "English" Knowledge 139


times, [popular physicians] fain themselves to be of some straunge countrey,
and wyll counterfayte their language."!!
English and alien practitioners competed with each other for economic
survival by relying upon a variety of time-tested strategies to carve out a
niche for themselves. Some resorted to advertising to draw potential clients to
their doors, such as surgeon Edward Parke (fl. 1564-88), who in 1568 erected
a sign outside his shop that described him inaccurately as "the skoller
[scholar] of St. Thomas of Willyngforde." Parke was competing with three
surgeons in the parish of St. Dunstan in the West, and though he undoubt-
edly knew that the Barber-Surgeons would demand he take down the sign,
the opportunity to set himself apart from his commercial competition by
claiming some education was too great to resist. 12 Other practitioners favored
more theatrical demonstrations of their knowledge in the homes of clients or
in the open, as did the surgeon John Smythe (fl. 1556-73) who was admon-
ished by the Repertory Court of Aldermen in 1573 to "make open show" of
his surgical skills "against his own house and dore and not elsewhere."13
Others settled in a neighborhood already reputed to offer the latest
designs in compasses and other scientific instruments, chemical prepara-
tions, or surgical techniques, as did a host of foreign clock, instrument, and
watch makers who flooded into the former ecclesiastical Liberty of the
Blackfriars following in the footsteps of the well-known engraver, instru-
ment maker, and medical empiric Thomas Gemini (fl. 1540- 1562). Most of
London's natural science neighborhoods were based, like the Blackfriars, on
one or more parishes, and were anchored by a parish church where the Eng-
lish and some immigrant households worshipped. Though many of the
Strangers attended their own Protestant churches-either the French,
Dutch, or Italian congregations-all those residing in the parish at the time
of death were recorded in the parish's register of births, marriages, and
deaths. The size of London's parishes varied widely from the minute St.
John the Evangelist not far from St. Paul's Cathedral to the suburban sprawl
of St. Botolph Aldgate. Even the smallest-and some were less than one
square acre-could house a surprisingly large number of residents by
today's standards.!4
Because of trade restrictions that were enforced by the City's guild and
livery companies, many immigrants gravitated toward those areas in the city
where corporate control was at its weakest, namely the former ecclesiastical
Liberties, the suburbs, and the areas within the city walls just adjacent to the
old gates. What these areas had in common was greater freedom, greater
space, and a more equitable mix of English and Strangers than could often
be found in the center of the City in neighborhoods that were largely con-
trolled by the guilds and livery companies. There the organizations fostered
their own sense of community that was based not on neighborhoods, but on

Deborah E. Harkness
allegiance to the company, and thus acted as a counterweight to neighbor-
hood associations. IS
The masters of London's guilds and livery companies, who were charged
with the task of regulating commercial behavior and restricting nonmem-
bers from engaging in their trades, were keen to apprehend both Strangers
and English citizens who impinged on their privileges. The Royal College of
Physicians, for example, vigorously sought out the French medical practi-
tioner Charles Cornet (fl. 1555-98), whom they described as "an Ignorant
Fleming and a most shameless buffoon," after he put up bills of advertise-
ment "on all the Corners of the City." The college punished Cornet by seiz-
ing "his feigned and unwholesome remedies" and throwing them into a
bonfire in Westminster's public market. 16 Such efforts were a particular
hardship to the alien practitioners, who were technically restricted from
guild membership except in special circumstances: at the request of the
queen or high-placed noble, for example, or if the alien practitioner met
guild standards through an examination or demonstration. Those few
Strangers who managed to become foreign members of the City's guilds and
livery companies paid steeply for the privilege.
As a result, many Stranger practitioners resorted to higher governmental
authorities, such as the City's Repertory Court of Aldermen and the queen,
for permission to advertise their services and practice their natural sciences.
Peter van Duran (fl. 1559-84), a brewer in St. Olave Southwark who was
also known by his colorful nickname "Pickleherring," for example, satisfied
the Aldermen that he "professethe ye knoledge & science of surgery," and he
was given permission in 1563 to "sette up bylles upon posts in such p[ar]ts of
the Cytye as to him shall seeme good to give the people knowledg of his said
science."17 The Dutch empiric Margaret Kemmex (fl. 1576-83), following
persistent efforts of the Royal College of Physicians to close down her med-
ical practice, successfully appealed to the queen and Sir Francis Walsingham
for protection. In 1581, Walsingham found it necessary to remind the Col-
lege that "it was her highness pleasure that the poore wooman should be
permitted by you quietly to practise and mynister for the curing of diseases,
and woundes, by the meanes of certaine simples."18 Walsingham cited two
reasons why Kemmex should be allowed to practice unmolested, one
knowledge-based and one shaped by economic imperatives: "god hath given
her an espetiall knowledge [of simples] to the benefit of the poorer sort," and
also for "the better maintenaunce of her impotent husband and charge of
family, who wholy depend un the exercise of her skill."1 9 Unwilling to let
Walsingham have the last word on such an important matter, the College
responded on 22 December 1581 that Kemmex's "weaknes and insufficiency
is suche as is rather to be pitied of all, then either envied of us or maintayned
of others."2o

"Strange" Ideas and "English" Knowledge


Royal patents often became the vehicle for immigrants to secure the right
to practice their skills despite guild and livery company restrictions, and for
the queen to ensure that some English citizens were trained in the process.
Patent petitions often contain evidence of heated intellectual debates
between Strangers and English citizens during which expertise and knowl-
edge were called into question. It was in the best interests of the crown, how-
ever, to arrange a collaborative detente. The responsibility for this kind of
brokerage and peacemaking fell to Elizabeth's secretary of state, William
Cecil. For decades Cecil vetted and supervised natural science projects like
an early prototype of today's National Science Foundation, judging the mer-
its of each proposal and, whenever possible, turning would-be competitors
into scientific collaborators, as will be seen below. One factor always gov-
erned his decisions: the profit that the patent holders would be able to bring
to the crown.
The spirit of commerce is vividly captured in many of the exchanges that
took place between natural scientists, their clients, and the crown in Eliza-
bethan London. Given London's competitive markets, we should not be sur-
prised that natural science ideas and expertise, be they English or alien, had
definite economic worth. Few doubted that some profit could be gained by
more highly trained mathematicians who might keep your accounts in
order, for example, or from a more fuel-efficient furnace, and most were
prepared to invest money in far riskier schemes to transmute metals, mine
for precious minerals, and construct water mills on London Bridge. Natural
science practitioners thus competed in a commercial world in which ideas
and materials were quickly transmuted into merchandise. This transforma-
tion was part of a general trend toward merchandising of which John
Wheeler, in his Treatise of Commerce (r60r), complained: "There is nothing . ..
so ordinarie, and naturaJl unto men, as to contract, truck, merchandise, and
traffike one with an other, so that it is almost impossible for three persons
to converse together two houres, but they wil fal into talk of one bargaine
or another." Commerce and exchange, Wheeler noted, were no longer the
sole province of merchants, but preoccupied everyone, high and low: "The
Prince with his subjects, the Maister with his servants, one friend and
acquaintance with another ... the Husband with his wife, [and] women
with and among themselves." Wheeler regretted the emphasis on com-
modities that resulted when his world went mad for merchandising. "[A]II
the world choppeth and changeth, runneth & raveth after Marts, Markets
and Merchandising, so that all thinges come into Commerce," Wheeler
lamented, "[T]his man maketh merchandise of the workes of his owne han-
des, this man of another mans labour, one selleth words ... [and] all that a
man worketh with his hand or discourseth in his spirit is nothing els but
merchandise."21

Deborah E. Harkness
The ways in which natural science practitioners struggled for economic
survival in this mercantile atmosphere provide us with important insights
into the conditions of intellectual exchange and the dynamics of commercial
competition in Elizabethan London. For some, as we will see in the exam-
ples below, economic survival could best be fostered through collaboration
with Strangers that would blend their alien ideas and practices with English
traditions. This was particularly true for practicing alchemists and engineers
who preferred syncretism, in which the resulting ideas or products repre-
sented an innovative blend of English and non-English knowledge and
practice. Other English and alien practitioners, especially clock makers and
other instrument makers, preferred to work apart from each other in dis-
tinct neighborhoods. For medical practitioners who faced enormous compe-
tition, however, there was a blend of conflict and collaboration that could
bitterly divide some practitioners while bringing others more closely
together in the face of their detractors. When taken as a whole, these exam-
ples are evidence of the complicated ways in which natural science practi-
tioners faced challenges in an urban, mercantile environment.

PARACELSIAN THERAPEUTICS AND THE MEDICAL MARKET:


COMPETITION, COLLABORATION, AND CONFLICT

Because medical practitioners were by far the largest group of natural sci-
ence practitioners within the Elizabethan City, encompassing physicians,
empirics, surgeons, and barber surgeons, midwives, "cutters for the stone,"
occulists, dentists, midwives, and nurses, understanding the challenges that
faced these practitioners and the negotiations that took place between alien
and English ideas represents an important contribution to our understand-
ing of urban-based natural science practice. In addition, medicine often pro-
vided the means by which English and immigrant practitioners could
engage in other, less lucrative branches of natural science such as natural his-
tory, astrology, and alchemy. It is all the more important, therefore, that we
understand the tensions between English and Strange practitioners, how
they were resolved, and the conditions under which intellectual exchange
could take place.
Medical practitioners represented a wide array of educational back-
grounds, from illiterate empirics who learned most of their skills from
demonstration and hands-on experience, to university-trained members of
the Royal College of Physicians. Despite these differences, there were often
strong links among what might seem on the surface to be entirely different
types of practitioners. The evidence suggests that physicians, surgeons, and
apothecaries did indeed form strong friendships through shared patients,

"Strange" Ideas and "English" Knowledge 143


remedies, and neighborhood ties. The close working relationships that could
occur between physicians and apothecaries can be seen in the medical and
alchemical papers of the apothecary Ed ward Barlow (fl. 158 I - 94), which
mention which physicians he provided with drugs and medicines on a regu-
lar basis, including Strangers Johann Vulpe (fl. 1581 - 89), Hector Nonez (fl.
1553-92), and John Shoring (fl. 1592/3) and Englishmen Thomas Penny (fl.
1569-89), Richard Forster (c. 1545-16), and Walter Baylie (fl. 1580-91/92).
Barlow's notes also indicate that English and alien physicians shared cases, as
did Christopher Atkinson and Hector NonezY The relationships that Bar-
low forged with members of the Royal College of Physicians are all the more
interesting because he was reprimanded in 1581 for practicing medicine
without the organization's consent.23
It was not uncommon for husband-and-wife partnerships to exist among
the surgical practitioners in Elizabethan London, or for surgeons to marry
midwives and establish a joint practice. Such is the case with surgeons
Hugh and Ann Vellam, immigrants who operated a joint practice in the
late 1560s which continued to be active after Hugh's death in 1568.24
William Baxter, a member of the Barber-Surgeons Company, was married
to Emma Philipps (fl. 157! - r603), whose brother, Edward Philipps, was an
apothecary. Emma Philipps was a medical empiric, and drew the ire of the
Royal College of Physicians, which described her as "an ignorant and bold
woman" and committed her to prison. "5 Guillaume Alaertes, a Stranger
surgeon, was married to a midwife, Lieven Alaertes, who was described,
much like Emma Baxter, as "an ignorant old woman" by the Royal College
of Physicians. 26
Given such a variety of backgrounds and expertise it is not surprising that
there was both fruitful collaboration and tense competition amongst medical
practitioners. Though many English practitioners resented the success of the
Strangers, for others the presence of alien ideas presented an opportunity to
share new ideas, books, and techniques. Surgeon George Baker (fl. r577-
r607), for example, praised the diverse backgrounds and expertise of Lon-
don apothecaries, and complimented the work of "Maister Kemech an Eng-
lish man," "mayster Geffray, a French man," and John Hester, whom he
called "a paynfull traveyler in those matters, as I by proofe have seene."27 But
Baker did not approve of all alien ideas: in his book on the preparation of
oleum magistrafe he scathingly criticized the London followers of Paracelsus
for putting their patients' health at risk. 28
John Hall's (fl. 1565) translation of Lanfrank's Chirurgia parva (1565)
recounts a fascinating tale of exchange which took place during his walk
along Bucklersbury, the London street famous for its grocery, spice, and
drug stores. 29 There he met a "woman [who] came to sell hearbes, to the
Apothecaryes." She offered the apothecaries maidenhair, but Hall was

Deborah E. Harkness
aghast when the herb gatherer produced "Nothinge agreeinge with that
whiche she named: But only it had rounde leaves, standinge in good order
on eche syde [of] the stalke, as maiden heare hathe." Hall took a sample of
the plant, in case he "might meete with anye, that knew it, and so to attaine
the name therof." Within an hour Hall was encouraged to consult with an
alien physician by his friend "master Gale Chirurgien of London" who lived
nearby on Lime Street in the parish of St. Dionysius Backchurch.
Lime Street was full of foregin-born practitioners with excellent intellec-
tual credentials, including the physician and botanist Matthew L'Obe1. 30
While we cannot be sure that Gale and Hall consulted L'Obel, it is certainly
possible. Whoever the physician was, he exchanged "divers communica-
tions" before meeting with Gale and Hall to view the troublesome specimen.
Hall and the alien physician soon parted company after the Stranger "sayde
it made no matter to be so precise in the knowledge of herbes." This single
anecdote is laced through with all sorts of exchanges: between the herb
woman and the apothecaries, between the apothecaries and Hall, between
Hall and his friend Thomas Gale, and between Gale, Hall, and the immi-
grant population of physicians and surgeons. In this case, the potential col-
laboration between the English surgeon and alien physician did not come to
fruition because the physician's "strange ideas" were incompatible with the
English surgeon's beliefs and practices.
It was Paracelsus and his remedies, however, that most polarized the
English and alien medical practitioners. One of the most notorious and well-
documented showdowns between Paracelsian advocates and opponents
involved Valentine Russwurin ofSchmalkald. 31 Self-described as a "Medicus
spagirirus opt[halamistus]/2 Russwurin was made denizen by Elizabeth in
1574 at the same time that he was practicing his Paracelsian remedies on the
London population.33 Russwurin's London career became problematic when
he took up the cure of Helen Currance, a musician's wife, on 3 April 1574.
In the presence of witnesses, Russwurin "did attempt with his instruments to
have taken out of her bladder a stone." The witnesses later alleged that "find-
ing none there, privily he tooke a stone out of the pocket of his hose ... con-
veyed it into a spunge ... [and] forst it in Pudendo."34 When this procedure
failed to relieve her discomfort, Russwurin sent her a powder that made it
impossible for her to urinate. Uncomfortable side effects from the powder
included blisters in her mouth, nose, face, and "inward parts of her bodie,"
which rendered her unable to eat.
Russwurin continued to treat patients for bladder stones, and branched
out to treat various illnesses of the eye, including cataracts. Most egregiously
affected was Mr. Castleton, a scholar of Cambridge, who still retained some
vision when he contracted for a cure with Russwurin. Shortly after their
agreement, Valentine "by his rustical dealings, put out his eyes cleane, and so

"Strange" Ideas and "English" Knowledge


deprived him of all his sight." Castleton had Russwurin arrested at the Royal
Exchange, "wher he did display his banners and wares ... being in the mid-
dest of his pontificalibus."35
Finally the matter was brought to the Court of Aldermen on 22 April
I574 who heard "certen complants and objections" from Russwurin's
patients as well as surgeons George Baker and William Clowes. 36 According
to Clowes, Russwurin left behind him a very long list of dead patients-
twenty-three in all-from all walks oflife including Master Mace, a grocer,
the servant of goldsmith Master Dummers, and two Strangers. l7 The court
put together a committee of two aldermen and two physicians, with instruc-
tions that they should call upon the expertise "of the Discretist and best
skylled surgeons of this cytie" to judge Russwurin's "knowledge & skill in
s[ur]gerye."38 Safely incarcerated in the Newgate Prison, Russwurin was
examined by a new, largely English committee on IO May I574. 39
No account survives of Russwurin's trial, and so we must rely on one of his
fiercest critics, William Clowes, for our insights into what transpired. Accord-
ing to Clowes, Russwurin had only one defender: a "proud bragger ... of the
foresaide Adders broode ... a man oflittle skill, and lesse honestie ... [who]
practiseth Chirurgerie, without all order or aucthoritie." An unascribed note
in a sixteenth-century hand in the British Library's copy of Clowes's work
states that this was "John Hester Alchymist at Paul's Wharf," who was a known
supporter of Paracelsian ideas. 40 Hester claimed that "Velentine Rasworme was
a wise Alchymist, " and that Clowes and Russwurin's other opponents were
"ignoraunt fooles and asses."41 Clowes felt unable to judge Russwurin's alchem-
ical skill, but did report that "I doe know wise Alchimistes, of mine opinion,
that accounts him in deede, an arch coosener, and loper, and Quacksalver."42
According to a treatise on the chemical analysis of urines and other
opthamalgic matters addressed to William Cecil, Russwurin may well have
impressed John Hester as a "wise Alchymist" because of his knowledge and
incorporation of Paracelsian ideas into his medical practices. Russwurin
chemically analyzed Cecil's urine, weighing it carefully to find that it was
"eyght ounces and a lytle more, wherein it hath no difference from a sound
man his water at all."43 He also discussed the problems Cecil's mother was
having with her cataracts, focusing on their hard "Tartar."44 These preoccu-
pations put Russwurin's practices well within the concerns of Paracelsian
therapeutics and would have resonated with anyone who had read works by
the author. 45
While much more needs to be done to tease out the full implications of
the Russwurin case for a better understanding of the influence of Paracel-
sian medicine in London, it is clear that such an analysis must take into
account the connections that were made between commercial success and
alien ideas within the city.46 For there was more at stake than just theoretical

Deborah E. Harkness
concerns when an English practitioner decried a Paracelsian cure - eco-
nomic and nationalistic matters were of at least equal importance. Eleven
medical treatises making reference to Paracelsus were printed in the six
years following the Russwurin case, marking a definite spike compared to
print trends prior to 1574. This evidence suggests that Russwurin, no matter
his fate at the hands of the English authorities, had an effect on the com-
mercial exchange of ideas in Elizabethan London. 47 While conflict might
overshadow collaboration and influence when viewed from the distance of
nearly five hundred years, the Russwurin case reminds us how complicated
intellectual exchange could be in the early modern period.

MECHANICAL MARVELS: STRANGE IDEAS AND ROYAL PATENTS

While medicine was a feature of life for nearly all residents in Elizabethan
London, the city was also a feast for the eyes of anyone interested in
machines, engineering, and visual displays of technological prowess. Those
interested in mechanical marvels could visit the windmills and glasshouses in
St. Giles Cripplegate just northwest of the city walls, or feel the heat of the
brick kilns and gun foundry in St. Botolph Aldgate to the northeast of the
city center. Smaller-scale industries also thrived within London's walls,
including clock- and watchmakers, mathematical instrument makers, and a
variety of smiths working on copper and iron. Public interest in viewing such
marvels is evident in the Court of Aldermen's decision on 17 October 1588 to
exhibit in the Guildhall an "artificial motion" devised by Henrick Johnson
from Utrecht in the Netherlands. The Guildhall was the symbolic center of
the City's civic power and judicial prerogatives, and it is striking that the
Aldermen permitted a Stranger to display his invention there. The Alder-
men did so at the request of Sir Thomas Heneage, Queen Elizabeth I's cham-
berlain of the household, with the stipulation that the display be for "such
inhabitants of this city [and] others who shall be willing to see the same."48
Instruments, artifical motions, clocks, and watches were much in vogue
in the period, their importance raised to new prominence through England's
naval expeditions and the influx of European-style watches and clocks that
accompanied the immigrants when they entered England. While the desire
to possess such items did not always indicate the purchaser's interest in the
finer points of natural science, the expense and stylishness of mechanical
marvels indicate their cultural currency. William Bourne, a well-known
author of navigational and mathematical texts, explained that prices for
some instruments put them out of reach for those natural science practition-
ers who could make most use of them, such as the "Mariners heere in Eng-
lande for that the charges is so muche in the making of them." Bourne was

"Strange"' Ideas and "English·· Knowledge


especially thinking of equinoctial dials, which indicated "the houre of the
day, & to shew the true shadowe of the Moone," which were not "used by any
English Master or Pylot, but only by one man, which person had not it for the
proper use therof, but rather had it, to say that he had suche an instrument as
no English man had the like, & to bragge that he had such an instrument that
he could do great feates therewith in the going of long viages."49
The relative rarity of instruments and their high asking price provided
skilled English and non-English technicians with the economic incentive to
produce more and more of the highly desired watches, clocks, and mathe-
matical instruments. Yet unlike the medical practitioners - for whom there
is so much evidence of collaboration and conflict-alien and English instru-
ment makers in Elizabethan London appear to have had little contact except
in two neighborhoods: the areas just outside the western walls of the city
between the parishes of St. Clement and St. Dunstan in the West, and the
neighborhood around St. Bartholomew's Hospital, which included parish
churches of St. Bartholomew the Less, St. Bartholomew the Great, and St.
Sepulchre. 50 Otherwise, the two groups tended to settle in different neigh-
borhoods in the city, and few English practitioners worked for Strangers
and vice versa.
This division of instrument makers into separate districts mirrors a strik-
ing division in workshop production: Strangers dominated the clock- and
watchmaking industries, while English makers took preeminence in the
field of mathematical instrumentation such as quadrants, astrolabes, staffs,
balances, and other navigational instruments. Such a firm division of labor
may well have had long-term implications for the development of a highly
instrumental natural philosophy in seventeenth-century England. 51 Despite
these differences, instrument makers shared some characteristics with med-
ical practitioners, namely that family ties and interconnections featured
strongly in their communities.
Two neighborhoods dominated by English instrument makers were the
printing district centered on St. Paul's Cathedral Churchyard and the extra-
mural parish of St. Botolph Aldgate. Five English instrument makers settled
in St. Botolph Aldgate, which is striking because so many foreign-born sur-
geons, apothecaries, physicians, and aquavita distillers lived there. Three of
these men-Richard Stevens (fl. 1569), Thomas Hearne (fl. 1592), and John
White (fl. 1602h603)-made compasses. 52 Yet the parish, in addition to
housing many Strangers, also was home to a number of sailors, mariners,
and shipwrights, which may indicate that there was a ready market of
potential consumers. In addition, the distribution of active dates may be evi-
dence of a single workshop.
Two notable instrument makers, James Kynvin (fl. 1570- 1610) and
Humphrey Cole (fl. 1568-91), lived around the precincts of St. Paul's Cathe-

Deborah E. Harkness
dral. Kynvin was highly recommended by William Bourne. Gabriel Harvey,
a staunch supporter of experiential knowledge who considered Kynvin "A
fine workman, & mie kinde frend," noted in his copy ofJohn Blagrave's The
Mathematical Jewel (1585) that the paper dials which Blagrave set forth in his
treatise could now be purchased in brass from his shop.53 Some members of
Elizabeth's court purchased instruments from him, including the Earl of
Essex, who bought from him a combined compass and sundial enclosed in a
box in 1593. 54 Humphrey Cole was a steady producer of mathematical
instruments including pocket compendiums, navigational instruments,
astrolabes, armillary spheres, ring-dials, sectors, gunners' scales, and theodo-
lites. 55 Even Elizabeth purchased instruments from Cole, who like Kynvin
received special mention by Gabriel Harvey who described him as a "Math-
ematicall Mechanicia[n]" in Pierces Superogation. 56
Alien instrument makers, alternatively, preferred to settle in two for-
mer ecclesiastical Liberties: the Blackfriars and St. Martin Ie Grand. Of the
two the Blackfriars was the more dynamic, with thirteen instrument mak-
ers known to have lived there during the Elizabethan period. The first
alien instrument maker to settle in the Blackfriars was Thomas Gemini,
whom Leonard Digges recommended to his readers in A Bake Named Tec-
tonicon (1556), stating that Gemini was "dwelling within the blacke Friers
... [and] is there ready exactly to make all the Instrumentes apperteynyng
to this booke."57 Another instrument maker, Eloy Mistrell, was one of the
Blackfriars' more notorious residents. Mistrell was a French goldsmith
who was arrested for counterfeiting and went on to be employed at the
Royal Mint after receiving a patent for his novel machine for stamping
coins. 58 Other instrument makers residing in the Blackfriars included
three members of the Vallin family of clock- and watchmakers from
Brussels; three members of the Noway family; Francis Rozean; Peter de
Hind; Laurence Dauntenay; Thomas Tiball, a balance maker; and Mark
Sara, a scale maker. It is difficult to assess their full importance in the his-
tory of Elizabethan instrumentation because so few of their works sur-
vive, but it is in clockmaking that the neighborhood appears to have made
its reputation. 59
Because of their expertise with clocks, many Stranger instrument makers
found additional employment in London's parish churches, whose clocks
were in perpetual need of repair and opened up a bottomless pit for parish
revenues. The clock of All Hallow's Staining was fixed repeatedly by a num-
ber of experts between 1558 and 1579, including the royal clockmaker
Nicholas Orshawe. 60 The clock at St. Helen's Bishopsgate was similarly
troublesome, and French clockmaker John De Mellayne kept busy continu-
ally from 1565 to 1569 mending, keeping, and, most important, oiling it. 61
Similar positions in parishes all over the city provided the clock makers with

"Strange" Ideas and "English" Knowledge 149


an additional source of revenue, and heightened English residents' aware-
ness of their skills and abilities.
Large-scale engineering and mechanical feats gave foreign-born techni-
cians an opportunity to exercise their technical skills by bringing continental
inventions into England. Elizabeth, who could be difficult in patronage
matters, was unusually appreciative of the hard work that went into many
inventions, as demonstrated by the case of William de Berger (fl. I535-67),
a coppersmith born in Utrecht. In I559 he was given a seven-year license to
make and sell a unique corn mill whose invention had "cost him much
money and study during the past seven or eight years."62 James Acontius, an
Italian engineer from Trent, even received one of Elizabeth's rare annuities
in exchange for his service to the crown. 63 Acontius's exemptions from guild
control which were implicit in the patent became an issue for contention in
I566 when representatives from the Masons, Tilers, and Blacksmiths were
called to the Repertory Court of Aldermen to be reminded of "the hole con-
tents and effects of the Quenes Ma[jesJtyes hir patentes ... in consideracon
of a certon & goodly and wyttye devise by hym fyrst found and taught to her
hygh[nJes subiectes for the makyge of furnesses." 64
The importance of engineering and inventions to the state is demon-
strated by the role that Elizabeth 1's chief minister, William Cecil, played in
supervising the patent process. Cecil's efforts ensured that there was an
emphasis on output, adherence to a timely schedule, and ultimate English
proprietorship over the skills of the inventors. To ensure that the crown's
investment in the project was fruitful, patents were typically granted only if
the work commenced within a proscribed time frame. When George Gilpin,
an English merchant, and Peter Stowghberghen, a Stranger, were given the
monopoly on "making ovens and furnaces after a new pattern, more eco-
nomical of wood and other fuel, which they have invented," they were
required to begin work within two months.i» Failure to produce the antici-
pated results within another stipulated time frame was grounds for Eliza-
beth to pull her support entirely, as is vividly illustrated in the patent granted
to Philip Cockerman, mercer, and John Barnes, haberdasher, who took over
a monopoly on the manufacture of saltpeter once held by a German mineral
expert, Garrard Honricke. They were granted a twenty-year monopoly, but
were warned that they had only one year to demonstrate the usefulness and
profitability of the process or their exclusive rights would be revoked. 66
The influx of new skills and techniques that accompanied the Strangers
into London was highly regarded by the City and the crown. Unlike other
forms of natural science practice, instrumentation and engineering caught
the attention of both common citizens and high-placed officials in Eliza-
beth's government. I would argue that such a wealth of technical expertise
specifically caught the eye of Francis Bacon, who lived a stone's throw from

Deborah E. Harkness
the St. Clement-St. Dunstan instrument-making neighborhood from 1576
when he enrolled in Gray's Inn. Though Solomon's House in the New
Atlantis (1627) has always been seen as a prescriptive for scientific practice, it
is clear from the examples above that Bacon did not need to actually dream
up the displays of ingenuity and inventiveness that he described there. Eliz-
abethan London-Bacon's London-had its own "engine-houses, where
are prepared engines and instruments for all sorts of motions," its neighbor-
hoods that produced "divers curious clocks," and its mathematical houses
"where are represented all instruments, as well of geometry as astronomy,
exquisitely made."67

ALIEN ALCHEMISTS AND THE ENGLISH ALCHEMICAL


TRADITION IN LONDON

While Elizabethan instrument makers and engineers made visual spectacles


of their inventions for the delight of the populace, London's alchemists
labored in more private quarters conducting experiments and expounding
upon theories that also promised great profit and rewards. Seventy-four
alchemists are known to have practiced in the city during the reign of Eliz-
abeth, and because this number is based on written remains it may well
under represent the number of actual practitioners. Alchemical practitioners
came from many occupations, including medical, metallurgical, apothecar-
ial, and distilling trades. The alchemical papers that survive reveal a wide
range of approaches, from the traditional to the Paracelsian.
No one neighborhood marked the center of alchemical practice; instead,
alchemists were distributed throughout the city, from its crowded center to
its more spacious suburbs. The highest concentrations of alchemists
occurred in St. Botolph Aldgate, a relatively industrial area outside the city
walls to the northeast, boasting four alchemists. St. Helen's Bishopsgate,
which was just inside the city walls near St. Botolph Aldgate, had three
alchemists. Most other parishes in London had one or two alchemical prac-
titioners. Alchemists may have spread out in this fashion because their prac-
tices were so likely to annoy their neighbors with smoking stills, fires that
were kept hot all day and night, noxious smells, and regular explosions.
There were instances, however, when the normally secretive practice of
alchemy was performed on the public stage. One such incident involved a
Polish alchemist, Cornelius Alnetanus, who defaulted on an agreement to
transmute lead into gold after Elizabeth gave him a stipend and access to
raw materials at the Tower of London's Mint. Hounded by William Cecil,
who was determined to make him produce something useful given the sub-
stantial investments made by the Queen, Alnetanus became so concerned for

"Strange" Ideas and "English" Knowledge


his safety that he tried to escape the crown's clutches by fleeing the country
with the bankrupt Princess Cecilia of Sweden. 68 Even more public, however,
were the alchemical controversies surrounding Frobisher's gold.
While much scholarly attention has been paid to this episode in the his-
tory of English navigation, the events surrounding the Frobisher's gold
assays have not been investigated for what they can tell us about alchemical
practice and the interactions between English and alien alchemists in Eliza-
bethan London. The episode does indeed reveal rich information about
these topics and highlights the place that alchemy could have in the English
commonwealth. In addition, though alchemists lived all over London, the
key players in the Frobisher assays all lived in close proximity in three adja-
cent areas of northeast London. Within the streets of the Bishopsgate,
Aldgate, and Tower Hill were John Dee, Sir Thomas Gresham, the Mus-
covy Company, the Mint, the Royal Exchange, and virtually all of the
alchemists involved in the controversy.
When Martin Frobisher returned from the New World with a dead
Eskimo, a black rock, and some other curiosities, he had no idea of the
excitement he was about to unleash. In October 1576 Frobisher gave
Michael Lok, a London merchant and member of the Cathay Company, a
small black stone that was discovered in present-day Baffin Bay.69 Lok
handed it to his wife, who, in a gesture of disdain for Frobisher's adventure,
threw it into the fire, where it began to burn. Her curiousity stirred, Mrs.
Lok retrieved the stone from the fire, washed the ashes away with vinegar,
and discovered that it glittered like gold. 70
Michael Lok was prompted by this discovery to give samples of the ore to
three investors in the Frobisher voyages: Queen Elizabeth I's assay master at
the mint and two members of London's powerful Goldsmith's company.
When all three failed to achieve anything as encouraging as had his wife,
Lok turned for assistance to a Venetian alchemist living in London, Gio-
vanni Baptista Agnello, who may have been introduced to him through
their mutual friend, John Dee.71 Agnello was able to produce "a very little
powder of gold" from the stone in early 1577.72 Lok expressed astonishment,
but Agnello assured him that his methods could be trusted because he knew
"how to flatter Nature.'m While Agnello continued to coax golden powder
from the black rocks in subsequent trials, gossip began to circulate in Lon-
don and the court about the mysterious substance. As the gossip swelled, so
too did the reported value of the ore and the quantity thought to be available
in the New World.
By the second week of January 1577, controversy erupted over the nature
of Frobisher's gold. The controversy coalesced around three related issues:
First, was the ore of any value? Second, who should be trusted with the
heavy responsibility of assessing and then certifying the ore's value? And

Deborah E. Harkness
finally, what method or methods should be used to extract anything of value
that might be embedded in the ore? Once three English goldsmiths, a
housewife, and a Venetian alchemist arrived at three conflicting assessments
of the ore's value, it became a matter for open speculation within the metal-
lurgical and alchemical communities of London, and separate trials of the
ore began. The queen's master of ordnance, Admiral William Winter, set his
own metallurgist, a Saxon named Jonas Shutz, to work on the ore assisted by
his alchemically minded friends Sir John Barklay and Sir William Morgan. 74
The queen's notoriously suspicious advisor, Francis Walsingham, alarmed at
the growing involvement of powerful people in London, sent his own sam-
ples to "certayne very excellent men" who reported that there was "nothing
therein, but ... a little sylver."75 Walsingham's experts included the courtier-
poet Sir Edward Dyer, who conducted his trials under Walsingham's skep-
tical eye; and a French alchemist living near the Tower, Geoffrey Le Brum.
While the ore was sent to various experts, Elizabeth became convinced of
its value through the promising results achieved by a collaboration of
Agnello and Shutz. Working together in furnaces at William Winter's
house on Tower Hill, the two men "by ... meanes of the learning of the sayd
Baptista in alchimia and the knowledge of the said Jonas in myneralls and
metalls handling," repeatedly gleaned gold from the ore. 76 While Jonas con-
tributed the practical metallurgical knowledge for which Saxony was
famous, Agnello's alchemical "learning" is more difficult to characterize, but
it seems to have been a deft combination of medieval notions with a sprin-
kling of Paracelsianism for good measure. 77
Working with metallurgical and alchemical techniques, Agnello and
Jonas convinced Elizabeth and her Privy Council of the richness of the ore
and were put into partnership. Disputes soon surfaced about the rightful
place of alchemical practices in the trials of the ore, but a compromise
method was reached: Agnello would handle the ore before it was put in the
furnaces and supervize the chemical additives that would make the melting
process easier; Jonas would then complete the process of melting and refin-
ing the gold in furnaces he had invented himself.78 With Agnello tagged as
the "chemical" man, and Jonas as the "furnace" man, the two men should
have been able to work together in Winter's house toward their common
objective.
The compromise failed when a second alchemist, an Englishman named
George Woolfe, was brought in to assist Agnello. Lok was forced to admit
that Agnello, Shutz, and Woolfe had irreconcilable differences in method
and approach which threatened the progress of the trials. In late November
I577, Lok wrote: "the iij workmasters cannot yet agree together, eche is
jelous of [the] other" and fears "to be put out of the work." What had been
productive collaboration had turned into competition, and Lok reported

"Strange" Ideas and "English" Knowledge I53


that the men were now "lothe to shew their conynge or to use effectuall con-
ferens" with each other. 79
While the disagreements raged on, other voices were raised. Urged by an
unknown party to lend his experiences to the project, the queen's German
physician, Dr. Burcot, "assayed and proved" that Frobisher's gold wasn't as
rich as all had been led to expect. Instead of Agnello's gentle, alchemical flat-
tery Burcot advocated a more aggressive and controlling approach to the
materials in which the "roughe wyeld and forrayne" ore would "be well hus-
banded by a skyllfull and expert man."80 Frobisher, smelling the first whiffs
of disaster at the gap between Agnello's alchemical courtship and Burcot's
metallurgical marriage, quickly threw in his lot with the queen's physician,
and tried to sabotage the proceedings by spying on Jonas and Agnello.
At court, the Privy Council became concerned and sought the advice of
"the goldesmithes and goldefyners of London and manye other namyd
counynge menn," all of whom "had made many prooffes of the ewer and
could fynde no whitt of goolde therein."81 In response, Frobisher urged yet
another collaboration-this time between Burcot and Jonas Shutz-and
the controversy surrounding the ore and its parting turned into a tale of
espionage and skulduggery. Throughout December 1577 and into the first
months of 1578 the two Germans exchanged insults: Jonas accused Burcot of
"evell manners" and of ignorance in "divers points of the works" while Bur-
cot responded by announcing that "yf Jonas had any couninge" it surely
should have yielded gold by now. The collaboration ended a few weeks later
when Shutz and Burcot refused to have anything to do with each other,
leaving only one line of communication between the two camps: the English
goldsmith Robert Denham (d. 1605). 8" After all the controversy and con-
tention, only Denham was able to turn the dross of the situation into a more
profitable career. Denham was probably spying on both Jonas and Burcot
and reporting directly to Elizabeth's Privy Council, resulting in his appoint-
ment as chief assayer on Frobisher's third voyage in the summer of 1578.
Later, Denham became the director of operations in the royal mines.
As the evidence from the medical, instrumental, and alchemical practices
makes clear, in the Elizabethan period natural science practitioners inter-
ested in the production of knowledge and the exchange of ideas were
increasingly caught up in the international commercial environment that
could be found at the Royal Exchange, where national independence was
tempered by foreign contributions. The building itself reflected the blend of
alien ideas and English knowledge that characterized natural science during
the period much more than the local well at Bishop's Gate ever could. For
the Exchange - supposedly the symbol of English economic indepen-
dence-was built with English workers and European know-how and
materials. The architect was Flemish; the stones came from Flanders; the

154 Deborah E . Harkness


windows and wainscoting came from Amsterdam; and the design was mod-
eled after the Bourses of Antwerp and Venice. While the Exchange may
have been intended to announce that England had fully arrived on the inter-
national economic stage, it also served as a reminder of the country's ongo-
ing relationship with the Continent and the Strangers in its midst.
Approaching the history of science from below helps us to focus on the com-
plex ways in which alien ideas and English knowledge were negotiated in
the neighborhoods of London and then transformed into commodities to be
exchanged. In doing so, it adds a new dimension to our understanding of
scientific practices and practitioners in the early modern period.

I. John Securis, A Detection and Querimonie of the daily enormities and abuses co{m}mitted in

physick, Concernyng the thre partes thereof that is, The Physitions part, The part of the Surgeons,
and the arte of Poticaries (London, 1566), sig. Biiiiv.
2. Elizabethans called residents who came from abroad "Strangers" or "Aliens," and those

who came from parts of England outside of London "Foreigners." One of the greatest
resources for studying these individuals is the four-volume Returns ofAliens Dwelling the City
and Suburbs of London from the Reign of Henry VIII to that ofJames I, ed. R. E. G. Kird and
Ernest F. Kirk (Aberdeen: Huguenot Society of London, 1900-08).
3. St. Martin Ludgate was home to nineteen surgeons between 1558 and 1603, including two
masters of the company, Thomas Bird (fl. 1577-1607) and Edward Griffin (fl. 1563-96). One
physician who lived in the parish during the Elizabethan period, William Harvey, was deeply
interested in anatomical studies and went on to set forth theories on the circulation of the blood.
4. Richard Forster (or Foster, ca. 1546-1616) and John Dade (fl. 1589-1614) were both
physicians. Forster taught Sir Christopher Heydon astrology, and published an ephemerides
dedicated to Robert Dudley, earl of Leicester, titled Ephemerides Meteorographica (London,
1575) which is an extremely well-designed blend of technical and narrative information. His
astrological manuscripts can be found in British Library MS Sloane 1713, ff. 1-9.
5. Forty-eight people who identified themselves as gardeners lived in the parish during
the period from 1558 to 1603.
6. St. Benet Paul's Wharf was a small parish on the Thames where novelty and innovation
were prized . Residents included foreign-trained physician John Osborne (fl. 1577-93), physi-
cian Richard Caldwell (ca. 1505-84) who possessed a number of instruments for use in sur-
gical practice, and John Hester (fl. 1570-93), whom Gabriel Harvey described as "the
alchemist of London" in his annotations on Hester's broadsheet These Diles, Waters, Extrac-
tions, or Essence Salts, and other Compositions (London, 1585?), now in the British Library.
7. Population figures for the period are extremely difficult to state with great precision,
but most historians estimate that London's population grew from about seventy thousand in
1550 to two hundred thousand in 1600. In 1571, the total population of London was approx-
imately seventy-three thousand. Most historians use the figures generated by Robert Finlay,
Population and Metropolis: The Demography of London 1580-165° (Cambridge: Cambridge
University Press, 1981). See Finlay, p. 68, for estimates ofthe Stranger population, and p. 53
for figures regarding the London population more broadly.

"Strange" Ideas and "English" Knowledge ISS


8. There is a large and growing literature on the Stranger population and its significance
in Elizabethan London. See Joan Thirsk, Economic Policy and Projects: the Development of a
Consumer Society in Early Modern England (Oxford: Clarendon Press, 1978); Steven Rappa-
port, Worlds within Worlds: Structures of Life in Sixteenth-Century London (Cambridge: Cam-
bridge University Press, 1989); Ian Archer, The Pursuit of Stability: Social Relations in
Elizabethan London (Cambridge: Cambridge University Press, 1991); Andrew Pettegree, For-
eign Protestant Communities in Sixteenth-Century London (Oxford: Oxford University Press,
1986); Huguenots in Britain and Their French Background, [550-[800, ed. Irene Scouloudi
(Totowa, N.J. : Barnes & Noble Books, 1987); Raingard Esser, "Germans in Early Modern
Britain," in Germans in Britain since [500, ed. Panikos Panayi (London: Hambledon Press,
1996), 17-27; Edward Chaney and Peter Mack, eds., England and the Continental Renaissance:
Essays in Honour ofj. B. Trapp (Woodbridge, Suffolk: Boydell Press, 1990); J. Arnold Flem-
ing, Flemish Influence in Britain (Glasgow: Jackson, Wylie, 1930), Vol. I; Henri Gorain, Les
Fran~'ais a Londres (Paris: La Vague, 1933); J. Van Dorsten, The Anglo-Dutch Renaissance:
Seven Essays, ed. J. van den Berg and Alastair Hamilton (Leiden: E. J. Brill, 1988); J. Van
Dorsten, The Radical Arts; First Decade of an Elizabethan Renaissance (Leiden: Leiden Univer-
sity Press, 1970).
9. My research to date has uncovered information regarding 1450 natural science practi-
tioners who lived in London and its immediate suburbs who were active during the period
from 1556 to 1603. Because of the patchy survival of Elizabethan records (whole volumes of
state papers, parish records, and court cases have not survived), no accounting of practition-
ers can ever hope to be complete. Still, this number is much higher in every category (such as
female practitioners, medical empirics, and alchemists) than one might expect.
ro. George Baker, The composition or making of the moste excellent and pretious Oil, called
Oleum Magistrale (London, 1574), 44v.
11. Securis, A Detection and Querimonie, sig. Ciiiv.
12. The three surgeons consisted of two English surgeons, Robert Clarke and Richard Wis-
towe, and foreign-born surgeon James Markady. On 16 March 1567/68 Parke was ordered by
the company to remove this flattering self-description and "to sette his signe as other Surgeons
do without any Superscription." Records of the Barber Surgeons, now London Guildhall MS
52571r, f. 52r. Hereafter, London Guildhall manuscripts will appear as GH.
13. Repertory Court of Aldermen, Corporation of London Records Office Rep. 18, f. I07V.
(hereafter CLRO Rep.)
14. Parishes that were one acre or less included St. John the Evangelist, All Hallows
Honey Lane, and St. Mary Mounthaw. Finlay, Population and Metropolis, 170.
15. There is a vast literature on London's livery companies and guilds, but Joseph Ward's
Metropolitan Communities: Trade Guilds, Identity, and Change in Early Modern London (Stanford,
Calif.: Stanford University Press, 1997) is the finest account of the efforts of these organizations
to foster community within the City. See also A. L. Beier, "Engines of Manufacture: the Trades
of London," in London 1500-1700: the Making of the Metropolis ed. A. L. Beier and Roger Fin-
lay (London: Longman, 1986), 141 - 167. Studies of specific organizations relevant here include
Joyce Brown, Mathematical Instrument-Makers in the Grocers' Company, [688-1800, with Notes
on Some Earlier Makers (London: Science Museum, 1979); Michael A. Crawforth, "Instrument
Makers in the London Guilds," Annals of Science 44 (1987): 319-377. Harold J. Cook, "Good
Advice and Little Medicine: The Professional Authority of Early Modern English Physicians,"
Journal of British Studies 33 (1994): I - V; Raymond S. Roberts, "The London Apothecaries and
Medical Practice in Tudor and Stuart England" (Ph.D. thesis, University of London, 1964).
16. The Royal College of Physicians, London has a series of unpublished manuscript vol-
umes of its proceedings in the period, known as the Annals. For references to Cornet see
Annals I: 8a.

Deborah E. Harkness
17. CLRO, Rep. IS, f. 156r. The Royal College of Physicians attempted in that same year
to shut down his practice. Annals I: 22b.
18. Annals 2: 7a.
19. Ibid.
20. Ibid., 2: 7b- 8a
21. John Wheeler, A Treatise of Commerce (London, 1601),6.
22. For Barlow's medical prescription books of 1588-90 see Bodleian Library MS Ash-
mole 1487. The manuscript also contains a description of his library, which contained 172
books and his original manuscripts.
23. See Annals 2: 6b.
24. See Ann Vellam's will proven and registered in the Archdeaconry Court of London on
14 February 157olr, now Guildhall MS 9171/31269' In her will Ann bequeaths a case of knives
to the Barber-Surgeons Company.
25. Annals I: 33a.
26. Passing mention is made of Lieven Alaertes in Margaret Pelling and Charles Webster,
"Medical Practitioners," in Health, Medicine and Mortality in the Sixteenth Century (Cam-
bridge: Cambridge University Press, 1979) 179 (where she appears as Lieven Allette) and in
Thomas Forbes, Chronicle from Aldgate: Life and Death in Shakespeare's London (New Haven,
Conn.: Yale University Press, 1971), 194. Margaret Pelling has done more than any other his-
torian to uncover the importance of empirical practitioners in early modern England. In
addition to numerous pathbreaking articles she is the author of The Common Lot: Sickness,
Medical Occupations, and the Urban Poor in Early Modern England (New York: Longman,
1998) and coeditor, with Hilary Marland, of The Task of Healing: Medicine, Religion, and Gen-
der in England and the Netherlands, 1450-1800 (Rotterdam: Erasmus Publications, 1996), both
of which are germane to the subject of this paper.
27. George Baker, The newe Iewell of Health, wherein is contayned the most excellent Secretes
of Phisicke and Philosoph ie, devided into fower Bookes (London, 1576), sig. *iiiv, sig. [*ivrJ and
p. 187v.
28. George Baker, Oleum Magistrale (London, 1574), sig. Ciir.
29. John Hall's translation of Lanfrank's A Most excellent and learned woorke of chirurgeri,
called chirurgia parva Lanfranct~ Lanfranke ofMylayne his briefe ... (London, 1565), sig. [Miiir-
MiiiirJ.
30. For a more detailed discussion of the Lime Street community, see Deborah E. Hark-
ness, "Living on Lime Street: The Anatomy of a Natural History Community in Elizabethan
London," forthcoming. For more on L'Obei and his significance in the history of botany, see
Edward Lee Greene, Landmarks of Botanical History, 2 vols. (Stanford, Calif.: Stanford Uni-
versity Press, 1983), Vol. 2: 876-937; and A. Louis, Mathieu De L'Obel 1538-1616. Episode de
l'Histoire de la Botanique (Ghent-Louvain: Story-Scientia, 1980).
31. The spelling of Russwurin's name appears to have given Elizabethan writers an unusu-
ally hard time. He is also known Valentyne Rawnsworm and Valentine Rushworm. Evidence
surrounding the Russwurin case must be pieced together from the Patent Rolls, records of the
Repertory Court of Aldermen, a treatise on the chemical analysis of urine and ocular medicine
by Russwurin (British Library MS Landsdowne 101/4) which is undated but which must have
been written by 1587, given its references to William Cecil's mother, who died in that year, and
William Clowes's A briefe and necessarie Treatise touching the cure of the disease called Morbus
Gallicus, or Lues Veneres (London, 1585). In modern scholarship I have found only passing ref-
erences to Russwurin in R. Theodore Beck, The Cutting Edge: The Early History of the Surgeons
of London (London: Lund Humphries, 1974), and Charles Webster, "Alchemical and Paracel-
sian Medicine," in Health, Medicine and Mortality in the Sixteenth Century, ed. Charles Webster
(Cambridge: Cambridge University Press, 1979),301 - 334,317. For a more detailed analysis of

"Strange" Ideas and "English" Knowledge 157


this incident, see Deborah E. Harkness, "Paracelsian Therapeutics in Elizabethan London:
The Case of Valentine Russwurin ofSchmalkald," forthcoming.
32. British Library MS Landsdowne lOl/4, f. ISV
33. Webster, p. 305, where he cites Calendar of Patent Rolls Elizabeth 6: 261, 25 February
1574. Russwurin was made denizen only a few weeks before he ran afoul of Mrs. Currance.
34. Clowes, Briefe and necessarie Treatise, lOr.
35. Ibid., II r.
36. CLRO Rep. IS, ff. 196r-v.
37. Clowes, Briefe and necessarie Treatise, I If-V.
3S, CLRO Rep. IS, f. I 96r-v. The doctors consulted were Peter Symons and the Italian
physician Julio Borgarucci.
39. CLRO MS Rep. IS, f. 21 If. The second committee was far more English in its compo-
sition: Dr. Smith of Oxford, Dr. Smith of Cambridge, Dr. Waller, Dr. Gyfford, Dr. Bor-
garucci, and the Portuguese physician Hector Nonez.
40. Clowes, Briefe and necessarie Ii'eatlse, 12r. John Hester is a fascinating Elizabethan sci-
ence practitioner who is well deserving of further study. William Eamon, in Science and the
Secrets of Nature (Princeton, N.J.: Princeton University Press, 1994) refers to Hester in rela-
tionship to his association with the Italian practitioner Fioravanti on 254-255. Allen G.
Debus also refers to Hester in Engizsh Paracelsians (Cambridge: Cambridge University Press,
1965), lOI.
4 I. Clowes, Briefe and necessarie Ii'eatlse, 12r.
42. Ibid., 12V.
43. British Library MS Landsdowne lOl/4, f. 12r. Though the treatise is undated, Russ-
wurin's claim that "from my first co[mlminge into this lande ... there hath not escaped, as I
am credibly enformed, a meale or meetinge where any of the universitye Doctores have bene
present, wherin I have not been backbytten, sclaundered, and also impudently ... belyed," is
in keeping with the general tenor of the events of 1574, f. 8r.
4+ British Library MS Landsdowne 101/4, f. 14r.
45. For the Paracelsian emphasis on the chemical analysis of urine, see Allen G. Debus,
The Chemical Philosophy: Paracelsian Science and Medicine in the Sixteenth and Seventeenth Cen-
turies, 2 vols. (New York: Science History Publications, 1977), Vol. I: 59, 109-IIO. For the
medical implications of tartar in Paracelsian medicine, see Vol. I: 107.
46. The historiography surrounding the adoption of Paracelsian ideas in Elizabethan
England has split between the Kocher/Debus approach (which emphasizes the "compromise"
between Galenic and Paracelsian ideas that was made by medical practitioners who disagreed
with Paracelsian theories while being attracted to Paracelsian therapeutics) and the approach
of Charles Webster, who emphasizes the enormous range and importance of Paracelsian
ideas that circulated within England during the period. I believe that the evidence presented
here and included in my larger study of natural science practice in progress points more in the
direction of Webster's approach. See P. H. Kocher, "Paracelsian Medicine in England: (ca.
1570-1600)," Journal of the History of Medicine II (1947): 451-480; Allen G. Debus, The Eng-
izsh Paracelsians (Cambridge: Cambridge University Press, 1965); Charles Webster, "Alchem-
ical and Paracelsian Medicine," in Health, Medicine, and Mortality in the Sixteenth Century, ed.
Charles Webster (Cambridge: Cambridge University Press, 1979),301-334.
47. I have been unable to find any references to the outcome of the trial. No mention of
Russwurin appears in the records of the Barber-Surgeons Company, and the Annals of the
Royal College of Physicians for that year does not survive. For a list of works published in
London that discuss Paracelsian theories and therapeutics between 1574 and 1580 see Web-
ster, "Alchemical and Paracelsian Medicine," 333, Appendix 2.
48. CLRO, Rep. 15, f. S98v.

Deborah E. Harkness
49. William Bourne, A Regiment for the Sea: Conteyning most profitable Rules, Mathemati-
cal experiences, and perfect knowledge of Navigation, for all Coastes and Countreys: most needefull
and necessarie for all Seafaring men and Travellers, as Pilotes, Mariners, Marchants &c. Exactly
devised and made by William Bourne (London, 1574), 58. For the importance of such recom-
mendations to the London instrument-making trade, see D. J. Bryden, "Evidence from
Advertising for Mathematical Instrument Making in London, 1556- 1714," Annals of Science
49 (199 2): 301 - 33 6
50. The area from St. Dunstan in the West to St. Clement contained the highest number of
instrument makers, including four Strangers (Awdrian Gaunte, Robert Grynkin, Peter Della-
mare, and Lawrence Fortuna) and English practitioners Thomas Brome, James Ilsberye,
Richard Blunte, Charles Whitwell, John Modye, and Bartholomew Newsam. The St.
Bartholomew's neighborhood had fewer practitioners, but they were relatively well-known and
influential, such as Maryan de Lander and Michael Nowen (both immigrants) and Christopher
Paine and John Reade. For a brief notice of Nowen, see Brian Loomes, The Early Clockmakers
of Great Britain (London: NAG Press, 1981),415. For brief notices of Christopher Paine and
John Reade see E. G. R. Taylor, The Mathematical Practitioners of Tudor and Stuart England
(Cambridge: Cambridge University Press, 1954), 189-190 and 185, respectively. Taylor's mon-
umental contribution to the history of English mathematics was one of the first works to
attempt to map mathematical practitioners (including instrument makers) within London.
51. The most recent and influential studies of the importance of instrumentation in Eng-
land during the Scientific Revolution are Steven Shapin and Simon Schaffer, Leviathan and
the Airpump (Princeton, N.J.: Princeton University Press, 1989) and Steven Shapin, A Social
History of Truth: Civility and Science in Seventeenth-Century England (Chicago: University of
Chicago Press, 1995).
52. Information taken from the will of Richard Stevens in the Archdeaconry Court of Lon-
don, now GH MS 9I7r131225, and the parish registers of St. Botolph Aldgate, GH MS 9221.
53. Virginia F. Stern, Gabriel Harvey: A Study of His Life, Marginalia, and Library (Oxford:
Oxford University Press, 1979),85,202. Harvey's copy of Blagrave is now at the British Library,
and has copious notes. The notes on Kynvin were probably made between 1585 and 1590.
54. Loomes, Early Clockmakers, 346.
55. Taylor, Mathematical Practitioners, 171 - 172.
56. For further information on Kynvin see Loomes, Early Clockmakers, 158; Taylor, Math-
ematical Practitioners, 187.
57. Leonard Digges, A Bake Named Tectonicon, briefelye shewynge the exacte measurynge,
and speady recenynge all maner Lande, squarted Tymber, Stone, Steaples, Pyllers, Globes, &c.
(London, 1556), title page.
58. See PRO State Papers Domestic, Elizabeth: I2!r9/35, I2!r9/46, 12122122, 12/22/52. Cal-
endar of Patent Rolls Elizabeth 2: 153 (a 1561 pardon for Mistrell's 1558 offense of counterfeit-
ing) and British Library Landsdowne MS 14/8 (his 1572 petition for patents on new and
improved machines for stamping coin).
59. Most of the information on these practitioners must be pieced together from extremely
fragmentary evidence. Nicholas Vallin and Michael Noway are the exceptions, as a few items
from their workshops do survive. There are references to the Vallin and Noway families in
Loomes, as well as in George White, The Clockmakers of London (Hants: Midas, 1998), 1-8.
60. See GH MS 495612, f. 84v- I32r for repeated references. In addition to Nicholas
Orshawe or Urseau, John Portar, John Skrewens, Bruce Awsten, John Goldar, and John
Newsam all worked on the clock between 1558 and 1579.
61. See references to John De Mellayne in GH MS 6836, f. 2r-9v.
62. Calendar of Patent Rolls Elizabeth I: 39.
63. Ibid. I: 254.

"Strange" Ideas and "English" Knowledge 159


6+ CLRO Rep. 15, f. 502r.
65. Calendar of Patent Rolls Elizabeth 2: 470.
66. Ibid. 2: 98.
67. Francis Bacon, The New Atlantis (1627), ed. Jerry Weinberger (Arlington Heights, Ill.:
Harlan Davidson, 1989),79-90.
68. James Bell provides an account of the scandalous visit of Princess Cecilia to England,
and her connections to Alnetanus in Queen Elizabeth and a Swedish Princess: Being an Account
of the Visit of Princess Cecilia of Sweden to England in 1565, ed. Ethel Seaton (London: Hawle-
wood Books, 1926). The state papers surrounding the case are scattered between the PRO, the
British Library, and Hatfield House. For further information on Alnetanus see Deborah E.
Harkness, "Queen Elizabeth's Alchemists," forthcoming.
69. Many of the state papers pertaining to the Frobisher voyages and the assays associated
with them have been reproduced in Richard Collinson, The Three Voyages of Martin Frobisher
(London: Hakluyt Society, 1867). For ease of reference, I will give references from Collinson
unless the manuscript consulted was not included in his collection. For references to the
exchange of the stone, see Collinson, 91. As for the stones themselves, they appear to have been
some form of hornblende pyroxenite. See Stuart K. Roy, "The History and Petrography of Fro-
bisher's 'Gold Ore'," Geological Series of the Field Museum of Natural History 7 (1937): 21-38.
70. George Best, A True Discourse (London, 1578),98.
71. For Agnello, see Deborah E. Harkness, John Dee's Conversations with Angels: Cabala,
Alchemy, and the End of Nature (Cambridge: Cambridge University Press, 1999), 204; Webster,
"Alchemical and Paracelsian Medicine," 307, and Harkness, "Queen Elizabeth's Alchemists."
From 1547 to 1549 "J. B. Agnelli & Co." were authorized by Wriothesley and Peckham to
import gold bullion for use in the royal mint (see PROEhol/303/9)' For more information on
this incident see C. E. Charles, The Tudor Coinage (Manchester: Manchester University Press,
1978), 181. Agnello's alchemical treatise Apocalypsis spiritus secreti (London, 1566) was later
translated by R[obertJ N[apierJas A revelation of the secret spiriti of alchymie (London, 1623).
72. Collinson, Three Voyages, 92.
73. Ibid., "Bisogna sapere adulare la natura."
74· Ibid., 97- 98.
75· Ibid., 97·
76. Ibid ., 174-175.
77. There are scattered clues to Agnello's alchemical views. Agnello gave John Dee a
Venetian book on alchemy-Giovanni Pantheus's Voarchadumia contra alchimiam-in 1557.
See Julian Roberts and Andrew Watson,fohn Dee's Library Catalogue (London: Bibliograph-
ical Society, 1990), #D16 and p. 157. Though relatively unknown today, Pantheus, who pro-
posed that practitioners return to a true, cabalistic alchemy, was one of those "most famous"
alchemists mentioned by Agricola. Dee's marginalia in the Voarchadumia indicates that it was
instrumental in the genesis of his Monas Hieroglyphica (1564). See Deborah E. Harkness,john
Dee's Conversations with Angels, 88- 89. Agnello's own alchemical text, Revelation of the Secret
Spirit, is consistent with the ideas expressed in medieval alchemical texts, but in the second
part, Agnello emphasizes the importance of salts to alchemical processes. Robert Napier, who
translated the text from Italian and Latin into English in the first quarter of the seventeenth
century, enthused about Agnello's "practical search ... Ifor 1that Chrystalline ... Salt" which
was so prominent in the writings of Paracelsus.
78. Collinson, Three Voyages, 175
79· Ibid., 192.
80. Ibid., 194
81. Ibid., 176.
82. Ibid., 176-178, 181.

160 Deborah E. Harkness


NETWORKS OF KNOWLEDGE
Commerce and the Representation of Nature
This page intentionally left blank
Local Herbs, Global Medicines
Commerce, Knowledge, and Commodities in Spanish America

ANTONIO BARRERA

O n 14 September 15 01 , Diego de Lepe, a resident of Palos, Spain,


received a royal license for trading in the New World with gold, sil-
ver, copper, mercury, and other metals, jewelry, and gems. In addi-
tion, the royal license acknowledged the possibility of valuable unknown
entities by granting Lepe the right to trade in "plants and animals of any
quality, fish, birds, species and drugs, and any other thing of any name and
quality even if they are of a higher value than those already mentioned."!
This royal license assumed that there would be new entities of unknown
qualities and names, with potential commercial value in the European mar-
ket. If there were new entities of unknown qualities and names, how to dis-
cover their qualities and economic value? In general, how did the Spaniards
study the New World entities and how did they find the qualities of these
new entities? What practices did they establish for organizing and using
information about the nature of the New World, and with what purpose?
Spaniards needed to learn about these things not only for commercial
purposes, but for their own health and living conditions (food, construction,
ornaments). Health issues were particularly important because many
Spaniards who arrived in the New World soon became sick, melting in
their European clothes. Many would die in the Caribbean islands, weak and
sick after a long and difficult voyage. 2 They passed on their diseases to the
Native Americans, who would die in massive numbers, weakening the sur-
vivors' ability to halt the new invaders.3 In the early phases of contact,
health issues might have constituted a common concern for both indigenous
people as well as for Spaniards. Moreover, sometimes the medicines the
Spaniards brought with them were already dated, or did not last long in the
new environment. 4 It is not difficult to imagine an apothecary or a physi-
cian asking the native inhabitants of the Santo Domingo port (the main port
of entrance to the New World in the early sixteenth century) for their med-
icines and herbs-not only to replace their own medicines but for commer-
cial purposes. This article explores the interactions between commerce and
knowledge in the production of empirical practices as well as the roles
played by the Spanish state and entrepreneurs in shaping those practices
through the case study of a drug found in the Hispaniola, the Santo
Domingo balsam. s
The story of the Santo Domingo balsam belongs to the intersection
between the history of European state formation, the history of science, and
the history of the Atlantic encounter. As the Spanish Empire established
institutions and laws in the American kingdoms and as commercial interests
brought New World commodities and curiosities to Europe, empirical prac-
tices developed for taking possession of nature in the New World. 6 In the
case of Spain, these empirical practices were institutionalized first at the
Casa de la Contrataci6n (f. 1503) and later at the Council of the Indies (f.
1524), as well as at the royal court, the viceregal courts in the American
kingdoms, and the Royal Academy of Mathematics in Madrid (f. 1584).7
The history of science and the history of the New World are intimately
related. The European understanding of the New World's nature passed
through a rapid process of transformation after Columbus's landing. s It
shifted from an image of paradise as described by Columbus to an overtly
pragmatic image deriving from personal accounts such as that of Dr. Chanca
to the city officials of Seville in 1494, or the letter of Michele de Cuneo to his
friend Hieronymo Annari, both of which discussed natural resources in con-
crete and practical terms. 9 This shift occurred largely as a result of the mate-
rial culture that shaped the European encounter with the New World. The
encounter of new lands provided merchants and royal officials with
untapped commercial possibilities. Europeans explored the new lands to
find new routes to the East and commodities for the European market. In
the process Europeans occupied and colonized the New World. The main
purpose of these explorations was to find new sources of revenues for the
state. During the explorations, some European notions of nature and experi-
ence were displaced from their traditional literary contexts. Classical notions
did not account, for example, for the size of earth, a new continent, life in
the Torrid Zone, manatees, and guayacan (a New World medicine, later
used to treat syphilis).
Experience gained in exploration and .in contact with other cultures
increasingly displaced classical sources as the authority for knowledge. The
natural products of the New World lacked referents in the classical sources.
An avocado was nowhere to be found in Pliny or Aristotle, thus the empiri-
cal information about avocados-or pineapples, iguanas, mountains, rivers,
or herbs - became an alternative and more reliable source of knowledge
than the imperfect knowledge contained in classical sources. Thus, the tra-

Antonio Barrera
ditional notion of experience, bound to Aristotelian texts, gained a new
autonomy within the commercial and imperial needs of the encounter,
where experience assumed a new role in validating knowledge.
The need to control faraway lands brought together royal bureaucrats,
merchants, pilots, and cosmographers in an effort to produce practical
knowledge that could be used to govern the new lands and profit from its
resources. 10 This effort led to the development at the Casa de la Contrataci6n
of intensely scrutinized and increasingly standardized mechanisms for gath-
ering, producing, and distributing useful knowledge about the New World.
The institutionalization of these offices and practices created a veritable
Chamber of Knowledge at the Casa de la Contrataci6n: a set of offices and
professionals in charge of collecting navigational and geographical knowl-
edge about the Indies, systematizing this knowledge, disseminating it
(teaching), and making new tools with this knowledge (instruments and
charts). The practices developed there reshaped the status and application of
personal experience in the creation of authoritative knowledge about nature,
particularly in the guise of cosmography and navigation. The use and devel-
opment of navigational techniques and instruments, and the establishment
of juntas of theoretical experts and practical people at the Casa de la Con-
trataci6n played as important a role in those domains as it did in the realms
of natural history and medicine. II Years later, the Council of the Indies
would implement similar empirical practices also with the aim of establish-
ing valid knowledge in natural history and medicine.
At the center of these empirical practices was the need to control human
and natural resources in the New World and, in particular, to control the
search for things that could bring profits in the Old World. The case of the
Santo Domingo balsam shows how empirical practices emerged from entre-
preneurial, imperial, and commercial contexts in sixteenth- and seven-
teenth-century Europe. It also illustrates how empirical practices that
emerged from commercial interests influenced the development of knowl-
edge production practices regarding resources from the New World.

TESTING NATURE

Balsam was an old and celebrated classical medicine. According to


Diosc6rides, Judea and Egypt produced balsam, but only in very small quan-
tities; a circumstance that encouraged the selling of fake balsam. The liquor
of balsam was better than its fruit or wood. The liquor "was very effective,
for its very hot quality."'2 It was used to cure vision problems, to purge, to
provoke menstruation and childbirth. It also helped to heal wounds, to pro-
voke urine, and to mitigate fatigue; it was a good antidote against poison.

Local Herbs, Global Medicines r65


Balsam was almost an all-purpose medicine in this period and in high
demand. However, the production of Egyptian balsam had already stopped
by the early fifteenth century as the traveler Pero Tafur, to his consternation,
discovered when he visited Matarea, Egypt, in the late 1430S.13
The garden of Matarea had been, according to legend, the place where the
Virgin Mary had found water when she and her son escaped to Egypt. This
water irrigated the balsam of Matarea. A well-informed humanist such as
Peter Martyr (1457- 1526) did not know that the production of the Matarea
balsam had come to a halt until he visited this garden in 1502 and found that
the fountain of water was already dry and the production of balsam had
stopped. 14 When Antonio de Villasante, a resident of Santo Domingo,
claimed to have found a similar product in Hispaniola and that he was will-
ing to exploit it with royal help, the crown supported him. There was already
a high demand for balsam in the international drug market, and the discov-
ery of balsam in Hispaniola came just in time to supply this market.
The discovery of the Hispaniola balsam, however, came not only as the
result of commercial demand but also as the result of an informal search
for curiosities in the New World. In 1525, the humanist Peter Martyr
obtained a royal decree that ordered ships' masters to bring animals and
plants such as parrots, "turkeys from Tierra Firme," "other strange birds,"
fruits, iguanas, chilies, cinnamon, roots, blue stones, amber, or "any thing"
that the officials from Hispaniola would want to send him. IS At the same
time, Charles V asked Gonzalo Fernandez de Oviedo to write a natural
history of the New World; Oviedo published a Sumario de la Historia Nat-
ural in 1526. It was in this context of interest in curiosities from the New
World that Antonio de Villasante would a few years later present his
report on balsam.
We know little about Antonio de Villasante. The scholar Ernst Schafer
maintains that Villasante was already a resident of Santo Domingo by
1514.16 He received thirty-five indigenous people, Tainos, in encomienda. 17
He married Catalina de Ayahibex, a chief, or cacica, who had converted to
Christianity. He became friend of the viceroy Don Diego Colon and later
obtained a license from him to exploit balsam and other drugs on the island.
Villasante then traveled to Madrid to secure a monopoly for the exploitation
of this and other drugs (escormonea, rhubarb root, aturbin, polipondyo,
acubebas, atiribinro, and myrrh) in the Caribbean.
Early in 1528, Villasante had obtained from the crown the right to exploit
balsam on the condition that he present before the Council of the Indies "a
long and very complete report about the tree to obtain the already men-
tioned liquor, and what its shape is and where this tree is found and what
method is used to obtain the liquor; and similar [information] about other
drugs."19 The crown requested an empirical and practical report about balsam

166 Antonio Barrera


and other drugs, and a few months later Villasante complied, presenting a
report on Santo Domingo balsam before the Council of the Indies. 19
Villasante told the council that his wife, whom he described as "an
Indian, cacica, and Christian," and her family had taught him about the
properties and uses of diverse medicinal plants from the island, one of which
was balsam. 20 Villasante thus established the authority of his native, female
informant. Because Catalina de Ayahibex was a cacica and a native, she
knew about the resources of her island. She may have been one of the last
survivors of the island by this time. 21 Moreover, she was Christian and thus
trustworthy. Villasante declared that he would state everything he knew
about the plants, so the king or his officials could, eventually, find and
exploit them. The issue of secrecy became irrelevant, for Villasante's purpose
was to obtain a commercial monopoly over balsam. It was to his benefit to
provide all the necessary information for its commercialization and sale.
After establishing the authority of his sources and information, Villasante
continued with a description of balsam. He explained that he knew from
experience that in Hispaniola, near Santo Domingo, there was a tree called
balsam in Spanish and boni, guacunax, or canaguey in the native language,
depending on the province. He described the tree, its height, girth, color,
and leaf shape, the color and smell of the bark, type of fruit, and habitat.
The tree was three yardsticks tall, and grew around rivers and wet areas;
it was about as thick as a human arm. The mature ones were bushy-topped
trees; the leaf was very green and in the shape of a rhombus. He then pro-
vided the royal officials with a very schematic drawing of a leaf. The bark
smelled and looked like cinnamon and tasted good, although it was a little
hot and sour. The tree produced a fruit like the pepper tree, he concluded,
but thicker. 22
Villasante knew that the credibility of his report depended on the author-
ity and knowledge of the indigenous people and on his direct access to that
knowledge. And the council was by then well aware that Spaniards needed
the knowledge of the indigenous people to survive and move around the
New World; in this area, as in many others, indigenous people had the
advantage. Villasante told the council that "the indigenous people affirm
that there are many other beneficial trees and drugs in the Indies," and he
promised to send reports on these as he learned about them.23
Villasante, however, relied not only on indigenous knowledge to support
his information but also on the language of commerce. His intended audi-
ence, the Spanish bureaucracy and his commercial partners, shaped his
description of balsam. The very name balsam that he used to translate the
indigenous names boni,guacunax, or canaguey, was connected with an exist-
ing commercial drug. But to make his point even more explicit, Villasante
compared the Santo Domingo balsam with pepper and cinnamon, two valu-

Local Herbs, Global Medicines


able spices in the international trading system at this time. The search for
spices and medicines, which had begun with Columbus, continued with Vil-
lasante and with the efforts of merchants and the crown to find new drugs
and spices as well as plants and animals "of any quality and name" in the
New World for commercial purposes. 24
Villasante confirmed his wife and her family's knowledge and the com-
mercial possibilities of the balsam by performing tests with it - what he
called "esperieneias." After describing the tree, Villasante provided the
method he used for preparing balsam brew. Villasante cut the branches with
a knife and took off the leaves and seeds with his hand; using his hands, he
shredded the pruned branches together with bark from the trunk; he
pounded this mixture with rocks; and finally he chopped it into pieces with
a knife. He then warmed the mixture in a clay pot with water. Once the
mixture had soaked awhile, he took it out of the clay pot and squeezed it to
obtain a liquor. Villasante finally heated the liquor in a small pot, which was
inside a bigger one full of ashes, until it was reduced to a thick liquor. 25
Sometimes he would put the liquor to dry under the sun with the same
results . He also mentioned that he once had cut the trunk of the tree with a
knife and that a "liquor" came out. This sap, "as it was coming out, hard-
ened like gum," he said, "and of this hardened [substance] I did not make
any other test or experience."26
According to Villasante's "experiences" in the New World, in Seville, and
at the court, balsam was effective for healing wounds in a short time. It was
also useful for healing all types of abrasions, and for relieving stomach pain.
Balsam was therapeutic for the liver and gallbladder, for treating gout, and,
finally, for relieving toothache. Villasante expected that the knowledge
about his balsam would increase, accumulate, and become more perfect with
time and new tests since, as he himself reported:

This [balsam], by experience, shows already that it is beneficial for the dis-
eases that I have mentioned. With time it may be shown by experience or
reports from physicians whether it might be beneficial for other things,
and they could also reveal the method for the perfection of this liquor and
balsamY

Villasante assumed that knowledge about balsam was cumulative and


would be based on the experience of physicians. This was a common ten-
dency in the production of knowledge related to the New World, and echoed
the model used at the Casa de la Contrataci6n. Only through the accumula-
tion of information and the correction of previous information through new
empirical information could physicians, cosmographers, and natural histori-
ans complete their study and understanding of the New World.

r68 Antonio Barrera


Villasante called on his own experience in the preparation and uses ofbal-
sam to, first, mediate the transfer of knowledge from the indigenous people
to the Spaniards, and, second, validate his commercial interests. Villasante
based his knowledge about balsam upon the expertise of Catalina de
Ayahibex and, more important, upon his own tests with balsam in the New
World as well as in the Old World. Note Villasante's method of presentation
here. He did not place balsam within the Galenic framework, for instance,
as the physician Garciperez Morales would do a few years later:

Of this precious liquor, commonly called balsam, which is brought from


Santo Domingo of the Indies: its first virtue is hot in the second grade, or
a little less; dry in the first metha of the third [grade], or a little more. 28

Garci perez Morales would write his treatise at the request of the crown in
1530. His audience consisted of royal physicians at the court and regular
physicians in Spain, such as his student Nicolas Monardes, who would
become well-known for his research on American plants. 29 Morales framed
his treatise in classical and traditional terms familiar to him and to his audi-
ence. Villasante, by contrast, framed his account in empirical terms for his
audience of Council of the Indies members and his commercial partners.
The difference in audiences, with their diverse interests and backgrounds,
explains the difference between Villasante's and Morales's approach to the
Santo Domingo balsam, a difference not unlike that between the practical
pilots and learned cosmographers at the Casa de la Contrataci6n. In addi-
tion, Villasante had firsthand experience of the New World and knowledge
provided by indigenous people while Morales had neither this experience
nor this knowledge.
Villasante's empirical approach to nature was not new to the Spaniards in
the New World. Since the mid-fifteenth century, both nominalists and
humanists had emphasized the collection of empirical evidence to solve
internal problems in their textual sources. What was new to Villasante and
the Spaniards in the Indies was the intense use of empirical evidence in
describing the things they encountered, the elaboration of their reports out-
side the traditional frameworks of knowledge, and the institutional role
played by the Spanish monarchy in this not yet formalized project of
research. The relevant framework here was the exploitation of commodities
for the European market.
The interest of the crown in the commercialization of balsam, for
instance, shaped its decision regarding not only the production of balsam,
but also the validation of empirical knowledge about balsam. For the pro-
duction of balsam, the crown granted Villasante, his heirs, and whomever
else he deemed appropriate a complete monopoly on the Santo Domingo

Local Herbs, Global Medicines


balsam as well as on the other drugs he would find in the New World. 3D Vil-
lasante also obtained, in perpetuity for himself and his heirs, the alcaldia of
the fortress of Santo Domingo, Indian labor, tax exemptions, and other pre-
rogatives. 3l The exploitation of balsam seemed to be ready. Other experts in
the field, however, soon challenged Villasante's report.

TESTING EXPERIENCE

In 1529 a competing report about balsam came to Spain from the Hispaniola
physician Licenciado Barreda, challenging Villasante's report. Barreda, who
has been the Inquisition's physician, left Spain for Hispaniola with Pedrarias
Davila's expedition to Panama (1513-14).32 In December 1513, the crown
had ordered the Cas a de la Contrataci6n to pay 12,000 maravedis to Barreda
for his travel expenses. 3l In Hispaniola, he held the title of royal physician
for some time until 1519, when the crown suspended his title. 34 In 1526,
Barreda was appointed official physician of Santo Domingo. 35 By the time
Barreda wrote his report on the so-called balsam of Santo Domingo in 1528,
he had been in the New World almost fifteen years.
In his report, Barreda argued that the crown had been deceived by the
physicians who "approved as balsam the liquor that the aforementioned Vil-
lasante" took with him to Spain. 36 Barreda claimed that the royal support for
this drug, "approved" as balsam by physicians in Spain, would harm the per-
son and property of the crown's subjects. He criticized the fact that the
physicians in Spain did not discuss the matter of the balsam with their col-
leagues in Santo Domingo:

[Spain's physicians] know or should know that they [Santo Domingo's


physicians] do not lack letters, nor extensive experience, nor knowledge of
the tree, its fruit and leaves and the methods to apply the aforementioned
liquor that comes from this tree. 37

For Barreda the lack of interest on the part of the Spanish physicians in
sharing their opinions and in consulting their learned and experienced coun-
terparts in the New World were an offense and a great mistake, which led
the crown into a dangerous deception. He emphasized the possible financial
and health consequences of using a fake balsam for the kingdom. Barreda's
report helps to uncover the link between knowledge and political and eco-
nomic power in sixteenth-century Spanish America. Knowledge pertaining
to the New World, claimed Barreda, had to be articulated by those with
direct experience of the New World. Physicians in the Old World, despite
their "letters," did not have this experience. For this reason, they needed to

170 Antonio Barrera


consult with their counterparts in the New World. Otherwise political and
economic decisions regarding the New World could harm the subjects of the
king. In the interest of the common good, personal experience was a better
source of knowledge than "letters" alone. The crown and its institutions, the
Council of the Indies and the Casa de la Contrataci6n, adhered to and sup-
ported this view. The issues at stake concerned who would control the pro-
duction of knowledge-the crown in collaboration with entrepreneurs such
as Villasante, or physicians such as Barreda?
Once Barreda had established his authority based on his expertise and
personal experience, he moved into the description and uses of Villasante's
liquor and compared it to balsam. Villasante had applied the same empirical
model to his own account of the balsam, but he did not compare his liquor
with the original, Old World balsam; he assumed that they were similar.
Personal experience is always fragmented and based on the personal back-
ground of the informer or informers. How could these two accounts, both
based on experience, be reconciled? Which one was more reliable?
Both Villasante and Barreda argued that their respective accounts were
true because each one was based on personal experience and experimenta-
tion. What mechanism could be established to determine the truth of the
matter? In the case of the pilots and cosmographers, the crown established
the mechanism of juntas to determine the truth of different accounts by con-
sensus. In these juntas the experience of pilots together with the formal
knowledge of cosmographers came together to produce new knowledge
about the New World. 38
The crown established a similar solution in the case of the New World
balsam - perhaps taking a hint from Villasante's suggestion that "with time
it may be shown by experience or reports from physicians whether it might
be beneficial for other things." The crown, after learning of Licenciado
Barreda's account and other similar accounts, requested that different Span-
ish physicians and hospitals carryon experiments with the balsam. Barreda
had provided a convincing case for his own account. He noted the differ-
ences between classical balsam and the "liquor that Villasante" took to
Spain:

[The1main virtue of this liquor is to restrain the blood in fresh wounds by


pressing it over them, and [to restrain 1the flow of blood from below [rec-
tum], this virtue, either called opilativa ... or constritiva ... , in what
books does it appear that balsam has this virtue? 39

Certainly classical texts on medicine such as Diosc6rides's Materia medica


did not list this virtue among those attributed to balsam. 40 Barreda also com-
pared the trees and the different methods to obtain the liquor from each type

Local Herbs, Global Medicines I7 I


of tree before concluding that Villasante's liquor was not the authentic bal-
sam. By the mid-sixteenth century, such scholars as Gonzalo Fernandez de
Oviedo, Andres Laguna, Nicolas Monardes, Pedrarias de Benavides, and
Conrad Gessner would agree with him.4l Nevertheless, Barreda found that
this "liquor has other virtues experimented by me [par mi spimentadas ]."42 He
found that the Santo Domingo balsam was efficient for healing rheum, and
kidney and stomach "passion."43
Dr. Barreda was not alone in his criticisms. In I530, the crown claimed
that there were already "some physicians, surgeons and other people who,
without complete information on the balsam recently discovered in our His-
paniola and without yet having made any experience with it, have published
and continue to publish some publications (ynpreciones),,44 against it. More-
over, people had decided not to buy the new balsam because of these publi-
cations, which "harms the health of the sick and wounded, and our royal
treasury."45 Such publications indicated that criticism had merit, but the bat-
tle for true knowledge about the balsam had just begun.
The crown sought to control and discipline this group of dissident physi-
cians by ordering that

physicians, and surgeons of any city, town and place of our kingdoms and
possessions should have unequivocal information [cierta noticia] about this
balsam before they talk or publish works, and when, by experience or by
other method, they find out that it is harmful for wounds or any other ill-
ness, they should declare and reveal it to our local magistrates. 46

Meanwhile, local magistrates should try to foster the sale of balsam "in
the best way they see fit."47 The crown sought to control dissident physicians
by ordering them to speak or publish only after they had made experiments
with the balsam, for which they had to buy it. Furthermore, they had to
bring their experimental findings before local magistrates, who would send
them to the crown. By asking physicians to experiment with balsam and
then to show their reports to royal officials, the crown controlled the pro-
duction of knowledge about the balsam. This situation shows the interplay
between the production of new medical knowledge and the political and
economic interests of the crown in controlling this knowledge and its prod-
ucts. In this particular case, controlling knowledge about balsam amounted
to controlling the possibility of its commercialization.
With the I530 decree to royal officials, the crown established a protocol
for the articulation of empirical information about Santo Domingo balsam,
namely, experimentation with samples and the recording of findings. Again,
this model resembles that of the Casa de la Contrataci6n case in which cos-
mographers and pilots appointed by the crown organized empirical infor-

172 Antonio Barrera


mation about navigation and geography provided by pilots. In the interplay
between the interests of the crown and the interest of individual subjects
there arose a scientific practice based on empirical experimentation ("experi-
encias") and the collective articulation of the resulting information.
The crown, however, not only attempted to discipline dissident physi-
cians into experimenting with the balsam; it also ordered particular physi-
cians, surgeons, and hospitals to carry out experiments with the New World
balsam. The crown had listened to the dissidents and sought to produce
accurate knowledge about the balsam. Following its own protocol, the
crown sent samples of balsam to physicians and hospitals for experimenta-
tion. In one case, the crown sent a sample of balsam, useful to "cure injuries
and many illnesses," to the hospital of the cardinal in Toledo for use on
patients chosen by the physicians and surgeons of the hospital. The crown
requested the hospital administrators "to be attentive to inform us of the
cures and experiences realized in the hospital with this balsam."48 Hospitals
in Seville, Burgos, Galicia, and Granada received similar orders. 49
Furthermore, the crown brought particular physicians into the project for
testing the Santo Domingo balsam. The physician Andres de Jodar, for instance,
a resident of Baeza, received the order to use balsam for those "cures and expe-
riences" that he would deem appropriate. 50 Moreover, whatever he found, by
means of "art" and experience, "certain and true," he should "put in writing,"
"sign" his report, and send it to Villasante's partners in Spain. Twenty-two
physicians and surgeons in different cities of Spain received similar orders.51Vil-
lasante's partners, Franco Leardo and Pedro Benito de Basniana, would use
these reports for the commercialization of balsam in Spain. They hired some
physicians and surgeons to help them in the commercialization of balsam. 52
By I532, information was already arriving to the court. A certain Juan de
Vargas had been using the "balsam from the Indies" to heal the sick. 53 He
seemed to have been quite successful, for the crown ordered the officials of
Cuellar to collect information from patients who had been healed with
Santo Domingo balsam. The scribe of Cuellar, Melchor de Angulo, received
the information and sent it to the crown. He received I08 reals for the eigh-
teen days he worked on this assignment. 54 The crown also requested Juan de
Vargas to come to the court, which he did in late 1532 or early I533. 55 Dur-
ing his stay there, he tested the balsam and was paid for his work. 56 Still
some medical practitioners opposed the use of this balsam and maintained it
was fake. In I539, the physician and apothecary of the village of Amusco
denounced Vargas for using the New World balsam. The authorities of
Amusco arrested him and took his balsam. He was later released; the crown
asked the authorities of Amusco to explain the matter. 57
In the end, the crown could not dismiss Barreda's contention that the His-
paniola balsam was not authentic. However, the only thing that mattered to

Local Herbs, Global Medicines


the crown was the fact that this balsam was especially good, as both Villasante
and Barreda had argued, at treating wounds. The crown sought, first, to
develop the right method to use it; second, to end the confusion between New
World balsam and classical balsam; and, finally, to convince other physicians
that it was a worthy medicine. The dissident group of physicians in Santo
Domingo and Spain were controlled and disciplined by requesting them to
experiment with the balsam and to send their results to royal officials-they
could not publish or discuss their findings without royal approval. Simulta-
neously the crown and Villasante's partners hired a group of physicians to
legitimize the use of balsam in their practice. The name balsam, given by
Villasante, was a propagandistic device to sell this new drug. The physician
Monardes, years later, commented that the liquor "received that name
because it produces great effects and cures many illnesses,"58 as had the clas-
sical, Old World balsam.
The economic possibilities of balsam shaped the research on it. This
research was characterized by, first, empirical observation, that is, knowl-
edge about products of the New World came from "experiences"; second,
professionals and experts collectively articulated this knowledge; and,
finally, the crown arbitrated the outcome of disputes about knowledge in
light of its economic and political goals. In its role as knowledge broker, the
crown established a protocol for research, which fostered economic and
commercial interests. From these interests emerged the empirical practices
that characterized the long-distance control of the New World. The balsam
episode helps to understand the significant emphasis placed on empirical
approaches to natural products of the New World, approaches that resulted
from the commercial and imperial activities of Europeans outside Europe.
The encounter with the New World slowly displaced European notions
of nature and experience, which had been closely tied to textual practices. 59
Physicians such as Barreda and entrepreneurs such as Villasante as well as
cosmographers and pilots at the Casa de la Contrataci6n, natural historians,
and explorers reexamined those notions to accommodate the increasing flow
of new knowledge circulating between Spain and America. Certainly, initial
information about new drugs came from books, but it was the testing of new
drugs, for instance, that provided final knowledge about them. European
notions about nature were adapted thus to incorporate discrete local settings,
soon to become gardens of knowledge, into an emerging global framework
of communication and trade. 61l
Simultaneously, local natural settings were adapted to fit European objec-
tives and strategies with regard to trade and the exploitation of natural
resources. Contact with the New World accelerated this process of trans-
forming and exploiting nature. In this emerging global context, nature was
largely secularized and approached in empirical terms. Nature became a con-

I74 Antonio Barrera


tingent reality adaptable to human plans and needs and a collection of com-
modities, such as balsam, or curiosities ready for exploitation or collection.
From about 1500S through the 1560s, this secular or practical approach to
the natural world fostered informal, nonsystematic empirical research into
the natural products indigenous to the New World. During the sixteenth
century nature was described and studied increasingly through empirical
terms as more groups joined this collaborative enterprise. The balsam
episode illustrates this development. By the late six tenth century there was
already an international network of scholars, including such well-known
figures as Carolus Clusius and Antonio Recchi, studying American nature
through Spanish gardens, books, and collections. 6 ! Nature's commodities
were tested at the courts, hospitals, and gardens; nature's curiosities were
collected, studied, and described. These empirical practices would first
become practical knowledge at the Casa de la Contrataci6n and at the Coun-
cil of the Indies before becoming institutionalized as science in the form of
natural history, cosmography, geography, and choreography.
The balsam case constitutes just one episode in the establishment and
institutionalization of empirical practices for controlling natural resources in
the New World. As the Spanish crown faced particular problems about
understanding and exploiting new products, it sought to create systematic
methods for the production of knowledge about the New World. As the
American enterprise developed, and as more groups became involved in it,
empirical information became increasingly more relevant and significant in
the production of knowledge. But one report, even if it is based on direct
personal experience, does not constitute knowledge. Consequently, the
crown sought to establish conditions for the production of several reports
based on testing and experimentation. The balsam case illustrates this
method. The first notice comes from Antonio de Villasante, who learned
about this medicine from the Tainos, in particular, from his native wife.
Soon, competing reports challenged Villasante's characterization of this
medicine as balsam, and the crown sent samples of the balsam to several hos-
pitals and almost two dozen physicians in Spain for testing the medicine on
patients. Whether or not this medicine was placed in a Galenic theoretical
framework, the issue at stake remained an empirical one: the practical uses
of the medicine, which could be found only through experience, as was the
case with the navigation to and mapping of the New World, the exploitation
of other resources, and the control of its human resources. The development
of empirical practices are at the center of the American enterprise; they
made possible the conquest and commercialization of the new resources.
Such practices are also at the center of the modern world.

Local Herbs, Global Medicines I75


I would like to thank the staff at the Archivo General de Indias (Seville, Spain) for their help
locating some of the material for this article, in particular, Pilar Lazaro de la Escosura and
Socorro Prous Zaragoza; my advisor, Paula Findlen, for her intellectual and personal support
during my research; and Pamela Smith for her advice and comments.

1. Cedula Real a Diego de Lepe on 14 September 1501, Archivo General de Indias, Indifer-
ente, 418, L. I, ff. 29v-32v. From now on I refer to the Archivo General de Indias as AGI.
2. Columbus was particularly concerned with this problem. In his second voyage many of
his people arrived sick or became sick afterward, and there were not enough medicines for
them. See the so-called Torres Memorandum of 30 January 1494 in Cristobal Colon: Los cua-
tro viajes, testamento, ed. Consuelo Varela (Madrid: Alianza Editoral, 1986), 209, 211. The
expedition of fray Nicolas de Ovando, who was appointed governor of the Indies in 1501, was
well provided with medicines and some medical instruments. See the list of medicines in
Angel Ortega, La Rtibida. Hlstoria documental crftica (Seville: Impresora y editorial de San
Antonio, 1925),315. From Santo Domingo, however, Ovando asked the crown to send more
medicines and apothecaries, a request that was granted. See Cedula Real a Fray NicoLls de
Ovando, I I November 1505, Salamanca. AGI, Indiferente, 418, L. I, ff. 185v- 186r. On occa-
sion the crown sent orders to treat those arriving in the Indies. Thus, in 1544, the crown
ordered the officials of Hispaniola to cure the friars arriving on the island because they would
probably arrive sick. See Cedula Real a los oficiales de la Espanola, 23 February 1544. Val-
ladolid. AGI, Santo Domingo, 868, L. 2, f. 208v. There was a name for a particular sickness
that the Europeans suffered as they arrived in the Indies: chapetonada. According to the sur-
geon Pedro Arias de Benavides, those who survived it would live for many years. See Arias
de Benavides, Secretos de Chirurgia, especial de las enfermedades de Morbo galico y Lamparones y
Mirrarchia, y assi mismo la manera como se curan los Indios de llagas y heridas y otras passiones en
las Indias, muy util y provechoso para en Espana y otros muchos secretos de chirurgia hasta agora no
escriptos (Valladolid: Impresor Francisco Fernandez de Carbona, 1567).
3. On the role played by disease in the conquest of the New World see Alfred Crosby, The
Columbian Exchange: Biological and Cultural Consequences of 1492 (Westport, Conn.: Green-
wood Press, 1972); and Noble David Cook, Born to Die: Disease and New World Conquest,
1492-1650 (Cambridge: Cambridge University Press, 1998). Cook also discusses the conse-
quences of diseases for Europeans as they arrived in the New World, or soon afterwards. See
Born to Die, 29ff.
4- In 1538, the crown was informed that many of the medicines taken to the New World
arrived, or became, "corrupted" there. The crown ordered its officials in Tierra Firme to
check the medicines and destroy those that were decayed. See Cedula Real a los oidores de
Tierra Firme. 16 April 1538. Valladolid. AGI, Panama, 235, L. 6, ff. 195v-196r.
5. When I began my research on the Hispaniola balsam in 1996 there were very few refer-
ences to it in the secondary literature, and only an article on Antonio de Villasante by Ernesto
Schafer, "Antonio de Villasante, descubridor droguista en la isla Espanola," lnvestlgacion y Pro-
greso 9, no. I (1935): 13-15. In 1996, I presented a paper at the Escuela Libre de Investigadores
(Seville, Spain), and, with few changes, at the Consejo Superior de Investigaciones Cientificas
(Seville, Spain) which discussed the balsam case and its commercialization. Later Esteban Mira-
Caballos published an article that discusses the Hispaniola balsam. See his article "La medicina
indigena en la Espanola y su comercializaci6n (1492- 1550)," Asclepio 44 (1997): 185 - 198.
6. On this subject see Antonello Gerbi, Nature in the New World (Pittsburgh: University of
Pittsburgh Press, 1985), and Raquel Alvarez-Pelaez, La Conqulsta de la Naturaleza Americana
(Madrid: Consejo Superior de Investigaciones Cientifias, 1993)'

Antonio Barrera
7. For the establishment of the Casa de la Contrataci6n see the royal decree of I4 Febru-
ary I503, AGI, Contrataci6n, 5784, I.I, ff. IV-2. See also Antonio de Herrera, Historia General
de los Hechos de los Castellanos en las Islas i Tierra Firme del Mar Oceano. Esaita par Antonio de
Herrera Coronista (sic) Major de Su Magestad de las Indias y su Coronisata de Castilla, ([Madrid:
Imprenta Real, I60~-I51; Madrid, I730)' decada I, p. I44; Joseph de Veitia Linage, Norte de la
Contratacion de las Indias Occidentales ([Seville, I672], Buenos Aires: Publicaciones de la
Comisi6n Argentina de Fomento Interamericano, I945), 4-5; Jose Pulido-Rubio, El Pilato
Mayor de la Casa de la Contratacion de Sevilla: Pilotos Mayores, Catedraticos de Cosmografia y
Cosmografo (Seville, I950); David C. Goodman, Power and Penury (Cambridge: Cambridge
University Press, I988), pp. 74ff.; J. H. Parry, The Spanish Seaborne Empire (Berkeley and Los
Angeles: University of California Press, I990), 54f.; and Clarence Henry Haring, Trade and
Navigation between Spain and the Indies (Cambridge, Mass.: Harvard University Press, I9I8),
chap. 2. The Casa de la Contrataci6n awaits its modern historian. On the establishment of the
Council of the Indies see Ernesto Schafer, El Consejo Real y Supremo de las Indias, 2 vols.
(Seville: M. Carmona, I935) as well as his article, "El Origen del Consejo de Indias," Investi-
gacion y Progreso 7 (5 May I933): I4I-I45. For an overview of the scientific activities of the
Casa de la Contrataci6n, the Council of the Indies, and the Royal Academy of Mathematics,
see Ursula Lamb, "Cosmographers of Seville: Nautical Science and Social Experience," in
First Images ofAmerica: The Impact of the New World on the Old, ed. Fredi Chiappelli (Berke-
ley and Los Angeles: University of California Press, I976), vol. 2: 675-686.
8. See Columbus's development through his diaries in J. Cecil, The Four Voyages ofCloum-
bus, 2 vols. (New York: Dover, I988). See also Valerie I. J. Flint, The Imaginative Landscape of
Christopher Columbus (Princeton, N.J.: Princeton University Press, I992); Stephen Greenblatt,
Marvelous Possessions (Chicago: University of Chicago Press, I99I); and Gerbi, Nature.
9. See the letter of Dr. Chanca (1494) in Martin Fernandez de Navarrete, Coleccion de los
viages y descubrimientos que hicieron par mar los Espanoles desde fines del siglo XV can varios docu-
mentos ineditos concernientes a la historia de la Marina Castellana y de los Establecimientos Espanoles
en Indias (Buenos Aires: Editorial Guarania, I945); and the letter of Michele Cuneo (I495) in
Samuel Eliot Morison, Journals and Other Documents on the Life and Voyages of Christopher
Columbus (New York: Printed for the members of the Limited Editions Club, I963), 209ff.
ro. On this topic see John Law, "On the Methods of Long-distance Control: Vessels, Nav-
igation, and the Portuguese Route to India," Sociological Review Monograph 32 (I986):
234-263; Steven J. Harris, "Confession-Building, Long-Distance Networks, and the Organi-
zation of Jesuit Science," Early Science and Medicine I (I996): 287-318 as well as his article
"Long-distance Corporations, Big Science, and the Geography of Knowledge," Configura-
tions 6 (I998): 269-304'
I I. The name "Chamber of Knowledge" is my own characterization of the scientific
aspects of the Casa de la Contrataci6n. By Chamber of Knowledge I mean the offices and
practices developed and institutionalized within the Casa for collecting and disseminating
information about the New World, for training lay people (pilots) in the new navigational
techniques, and for hiring professionals (cosmographers and pilots) for research and teaching
activities. Haring calls it a "Hydrographic Bureau and School of Navigation, the earliest and
most important in the history of modern Europe" (Trade and Navigation, 35).
I2. Andres Laguna, Pedacio Dioscorides Anazarbeo, acerca de la materia medicinal, y de los
venenos mortiferos. Traduzido de la lengua Griega en la vulgar Castellana, e illustrado can claras y sub-
stanciales Annotaciones, y can las figuras de unnumeras plantas exquisitas y raras par el doctor ... ,
Medico de Julio III, Pontifece Maximo (Anvers, I555), ff. 26ff. On the importance of balsam for
sixteenth-century naturalists, see Paula Findlen, Possessing Nature: Museums, Collecting, and
Scientific Culture in Early Modern Italy (Berkeley and Los Angeles: University of California
Press, I994), 270ff.

Local Herbs, Global Medicines I77


13. Pero Tafur, Andan,as e viajes de Pero Tafur por diversas partes del mundo avidos (1435-
1439) (Madrid, 1874),85-86.
14. Ibid., 575 n .: btilsamo
IS. Cedula Real del Rey don Carlos. January 29, 1525. Madrid . AGI, Contrataci6n, 5787,
N. I, L. 1. ff. 33-34v.
16. Ernesto Schafer, "Antonio de Villasante," 13. Villasante's name appears in a document
signed in Santo Domingo in February ISIS. In this document, Villasante was proposed as a
witness (together with other residents) to answer questions about Rodrigo de Albuquerque's
activities on the island in 1514. AGI. Justicia 1003, transcribed in Luis Arranz-Marquez,
Repartimiento y Encomiendas en la Isla Espanola (EI Repartimiento de Albuquerque de 1514),
(Santo Domingo: Ediciones Fundaci6n Garcia Arevalo, 1991).
17. Arranz-Marquez, Repartimiento, 560. On the Tainos, see Irving Rouse, The Tainos:
Rise and Decline of the People Who Greeted Columbus (New Haven, Conn., Yale University
Press, 1992). Cook suggests that the Taino population around 1492 might have been half a
million, by 1518- 19 the numbers had fallen to around eighteen thousand and by 1542 the
native population was less than two thousand. Villasante's information came from a group
that was disappearing from the earth. See Cook, Born to Die, 23-24.
18. Provisi6n Real proponiendo un asiento con Antonio de Villasante sujeta a la pre-
sentaci6n de un reporte de Villasante sobre el balsamo y otras drogas. 4 April 1528. Madrid.
AGI, Indiferente, 421, L. 13, ff. 85r-86v.
19. Fernandez de Oviedo comments that this balsam is not the real balsam but something
different that Villasante called balsam. Oviedo also says that Villasante either learned the
secret of balsam from his cacica wife or from an Italian physician who went to the Indies in
ISIS and died there. See Gonzalo Fernandez de Oviedo y Valdes, Historia General y Natural
de las Indias, 5 vols. (Madrid: Biblioteca de Autores Espanoles, 1959), vol. 2: 11.
20. Relaci6n de Antonio de Villasante, n/d, but it was probably presented in mid-I528.
AGI, Indiferente, 857. On 4 April 1528 the king ordered Villasante to present a report before
the council. By 14 June 1528 he had already submitted his report, see Indiferente, 421, L. 13,
ff. 2I3v-2I4r. Ernst Schafer thinks that this document dates from around 1526, for, accord-
ing to him, Villasante was in Spain in 1525, see Ernst Schafer, "Antonio de Villasante," 14.
Perhaps Villasante was in Spain in 1525 or 1526 and at that time sought support for his pro-
ject. The call number given by Schafer for this Villasante's report, Indiferente, 856, is a mis-
take; it is Indiferente, 857; see also a document in Colecci6n de documentos ineditos relativos al
descubrimiento, conquista y colonizaci6n de las posesiones espanolas en America y Oceania,
(Liechtenstein: Kraus Reprint, 1966), series 2, 14: 31.
21. See note 17.
22. Villasante, Indiferente, 857.
23. Ibid.
24. The Spaniards also took spices and medicines for agricultural and commercial pur-
poses to the New World. Thus, Don Francisco de Mendoza, son of the first Mexican viceroy,
signed in 1558 two capitulations with the princess Dona Juana (approved by Philip II in 1559)
to cultivate ginger, sandalwood and pepper, cinnamon, and clover. See Maria Justina Sarabia-
Viejo, Don Luis de Velasco, virrey de Nueva Espana (1550 - 1564) (Seville: Escuela de Estudios
Hispano-Americanos, 1978),403-405. The Council of the Indies considered this project
unfeasible. See Consulta del Consejo. 21 March 1559. Valladolid. AGI, Indiferente, 738, N . 47.
25. Villasante, Indiferente, 85T "La mana que hasta agora yo he tenydo en el sacar dellicor
con otros cosas de estos arvoles asy lo? que con un cuchillo cortados los rramos destos arboles
con su hoja y grano y con la mano arrancaba los granos y tambien la hojarada cosa por sy y
tomaba los rramos asy mondos y tambien tomava de la corteza de 10 grueso del arvol hacia el
tronco y 10 desmenuzava y ... taba? y 10 majava encima de unas piedras 0 losas con otras piedras

Antonio Barrera
o madero despues de picado con cuchillo y asy majado 10 ponya en un as vasijas de barro de?
m ... ? de barreno?nes? 0 labrillos? y ... ? calentaba en un caldero con una cantidad de agua
competente y la echava en el dicho barreno? y desde a un poce despues de enpapado y enbevido
en el agua 10 apretavba en un tornyllo de madera y sacaba dello todo el ~umo y .. . d? que tenyz
y 10 colava y colado 10 ponya en un caldero pequeno y despues tomaba otro caldero grande lleno
de ceniza hasta la my tad del . . . ?/ y dentro de aquel caldero de ceniza ponya y asentaba el otro
caldero pequefio con el dicho licor del balsamo colado y ponya fuego debajo del caldero de la
ceniza de mafia que el calor dela ceniza consumyese el agua que estaba en dicho licor hasta tanto
que se espesava y tornaba del color y mafia que yo ho he tenydo y entregado a su magi ... ".
26. Villasante, Indiferente, 857.
27. Ibid.: explained that balsam was "en la verdad provechoso asi en las Indias donde 10
experimente muchas veces como algunas en estos reinos en sevilla y en la corte y pues para
estas enfermedades que he dicho ha parecido por experiencia ser provechoso adelante podra
parecer por experiencias 0 por relacion de los medicos si aprovechara a otras cosas y tambien
ell os diran la forma que se podra tener para mas perfeccion del dicho licor y balsamo y otras
cosas del dicho arbol."
28. Garciperez Morales, Tratado del Bdlsamo y de sus utilidades para las enfermedades del
cuerpo humano. Compuesto por el Doctor . .. catedrdtico de prima en el colegio de Sancta Maria
de Jesus de la ciudad de Sevilla. Dirigido al yllustrissimo senor don Pedro Giron Duque y Conde de
Urena (Seville, 1530), ff. 2r.
29. Nicolas Monardes was a physician and entrepreneur very interested in the natural
resources of the New World. His father, Niculoso de Monardis was a Genovese bookseller
established in Seville. It is unclear when Nicolas Monardes was born; he died in 1589.
Monardes obtained his B.A. in art and philosophy in 1530 and a B.A. in medicine at the Un i-
versidad Complutense in 1533. In 1547 he obtained the licenciatura and doctor's degree from
the Colegio-Universidad de Santa Maria de Jesus de Sevilla. Monardes's medical practice in
Seville was very successful. He worked with the doctor Garciperez Morales and, in 1537,
Monardes became Morales's son-in-law when he married Morales's daughter Catalina.
Monardes was also very successful in his commercial activity in the Indies. He obtained many
plants and herbs for his medical practice and his garden through his commercial contacts in
the Indies. See Juan Jimenez-Castellanos y Calvo-Rubio, prologue to Historia medicinal de las
cosas que se traen de nuestras Indias Occidentales que sirven en medicina ... por Nicolris Monardes
(Seville: Padilla Libros, 1988), v to xi. See also Nicolas Monardes, {Primera y Segunda y Tercera
partes de la} Historia Medicinal de las cosas que se traen de nuestras Indias Occidentales que sirven
en Medicina ([Facsimile edition, 15741 Seville: Padilla Libros, 1988); for a discussion of
Monardes's work see Jose M. Lopez Pi fiero, "Las 'Nuevas Medicinas' Americanas en la Obra
(1565-1574) de Nicolas Monardes," Asclepio 42, no. I (1990): 3-67. The work of Monardes
was translated into English (Nicolas Monardes, Ioyful! Newes Out of the Newe Founde Worlde
[London, 1577)) as well into French, Latin, and Italian.
30. Provision Real a Antonio de Villasante. 20 April 1528. AGI, Indiferente, 421, L. 13, ff.
rror- I I If.
3I. Ibid. 22 April 1528. AGI, Indiferente, 421, L. 13, ff. rrIf- II2r; Real Provision a Anto-
nio de Villasante. 14 June 1528. AGI, Indiferente, 421, L. 13, ff. 2I3v-2I4r.
32. Cedula Real a los oficiales de la Casa de la Contratacion. 16 December 1513. Madrid.
AGI, Panama, 233, L. I, f. I26r.
33. Ibid.
34- Cedula Real allicenciado Rodrigo Figueroa, juez de residencia de la isla Espafiola. 26
July 1519. Barcelona. AGI, Indiferente, 420, L. 8, f.97v.
35. Cedula Real a los oficiales de la Espafiola. 14 September 1526. Granada. AGI, Indifer-
ente, 421, L. I I, ff. 202V- 203r.

Local Herbs, Global Medicines


36. Carta dellicenciado Barreda al rey Carlos V. 26 October 1528. Santo Domingo de la
Espanola. AGI, Patronato, '74, R. 43 ·
37. Carta de Barreda, AGI, Patronato, 174
38. See Alison Sandman's contribution to this volume, "Mirroring the World: Sea Charts,
Navigation, and Territorial Claims in Sixteenth-Century Spain"; and Ursula Lamb's fasci-
nating articles, "Science by Litigation: A Cosmographic Feud," Terrae Incognitae I (1969):
40- 57, and "The Spanish Cosmographic Juntas of the Sixteenth Century," Terrae Incognitae
6 (1974): 51- 6+
39. Carta de Barreda, AGI, Patronato, 174: "Ia virtud mas principal que se halla en el dicho
licorl es restrenir la sangre en las lIagas frescas sobre elias aplicadol y dado por la boca el fluxo
de sangre por abaxol dest avirtud agora se opilativa que sua viscositate aut g?oficie inplendo
venari orificia rectineat sanguyneusl agora sea constrictiva que sua frigiditate Ir? stiticitate?
constringat venas. digo que entanta manera aprieta que puesto sin ligadura parece el miembro
estar atadoll pues donde se vido ni en que libros se hallo tener el balssamo esta virtud antes de
todo en todo contra ria en 10 qual por ser muy manifiesto dexo de ser prolixoll."
40. Laguna, Dioscorides, 26-27.
4I. See Andres Laguna, Pedacio Dioscorides Anazarbeo, acerca de la materia medicinal, y de
los venenos mortiferos. Traduzido de la lengua Griega en la vulgar Castellana, e illustrado con
daras y substanciales Annotaciones, y con las figuras de unnumeras plantas exquisitas y raras por el
doctor . .. , Medico de Julio III, Pontifece Mdximo (Anvers, 1555),26 and 27; Nicolas Bautista
Monardes, Historia Medicinal, ff. 9ff; Conrad Gesner, Evonymus C. Gesneri Medici de Remedis
secretis, Libel' Physicus, Medicus & partim etiam Chymicus, & Oeconomicus in vinorum diversi
saporis apparatu, Medicis & Pharmacopolis omnibus praecipue necessarius, nunc primum in lucem
editus, nip, n/d [This seems to be the edition from Zurich, c. 1565; for the date and place see
Klaus Wagner, Catdlogo abreviado de las obras impresas del siglo XVI de la Bibliotec'a Universi-
taria de Sevilla (Seville: Universidad de Sevilla, '988)], ff. 13rr-v; Pedrarias de Benavides,
Secretos de Chirurgia, especial de las enfermedades de Morbo galico y Lamparanes y Mirrarchia, y
asimismo fa manera como se curan los Indios de llagas y heridas y otras passiones en las Indias, muy
util y pravechoso para en Espana y otras muchos secretos de chirurgia hasta agora no escritos (Val-
ladolid, 1567), ff. 3ov-31r.
42. Carta de Barreda, AGI, Patronato, 174.
43. Ibid.
4+ Cedula Real de la Reina a las justicias de Sus reinos. 5 April 1530. Madrid. AGI,
Indiferente, 422, L. 14, f. 67v.
45. Ibid.
46. Ibid ., 67v-68r.
47. Ibid., f. 68r.
48. Cedula Real a los visitadores del Hospital del Cardinal de la ciudad de Toledo. 5 April
1530. Madrid. AGI, Indiferente, 422, L. 14, f. 72V.
49. Cedula Real a los visitadores de varios hospitales. 5 April 1530. Madrid. AGI, Indifer-
ente, 422, L. 14, f. 72V.
50. Cedula Real al bachiller Andres de Jodar medico, vecino de Baeza. 5 April 1530.
Madrid. AGI, Indiferente, 422, L. 14, ff. 73r-74v.
5I. Cedula Real a varios medicos y cirujanos. 5 April 1530. Madrid. AGI, Indiferente, 422,
L. 14, ff. 73r-74v.
52. Cedula Real a Pedro Benito de Basniana y Franco Leardo para que puedan subir los
salarios asignados a los medicos que contribuyen a la propaganda del balsamo. 12 July 1530.
Madrid. AGI, Indiferente, 422, L. 14, ff. 102r- I03r.
53. Cedula Real a los oficiales de Cuellar. 16 October 1532. Madrid. AGI, Indiferente, 422,
L. IS, ff. 197v-198r.

ISO Antonio Barrera


54. Ibid.; Cedula Real a Diego de la Haya para que pague a Melchor de Angulo. 27
November 1532. Madrid. AGI, Indiferente, 422, L. 15, f. 199v.
55. Cedula Real a Juan de Vargas para que venga a la corte. 21 November 1532. Madrid.
AGI, Indiferente, 422, L. 15, f. 199r.; and Mandamiento a Diego de la Haya para que pague a
Juan de Vargas por haber estado en la corte. 27 February 1533. Madrid. AGI, Indiferente,
422, L. 15, f. 199r.
56. Mandamiento a Diego de la Haya para que pague a Juan de Vargas por haber estado
en la corte. 27 February 1533. Madrid. AGI, Indiferente, 422, L. 15, f. 199r.; Cedula Real a
Diego de la Haya para que pague cierta sum a a Juan de Vargas. 3 October 1533. Monzon .
AGI, Indiferente, 422, L. 16, f. 43V., Real Cedula a Juan de Vargas. 18 April 1534. Toledo.
AGI, Indiferente, 422, L. 16, f. 75v.
57. Cedula Real a los alcaldes ordinarios de la villa de Amusco. 23 May 1539. Toledo. AGI,
Indiferente, 423, L. 19, ff.247-248.
58. Monardes, Historia, ff. 9r.
59. See Flint, Christopher Columbus; Richard H. Grove, Green Imperialism (Cambridge:
Cambridge University Press, 1995); Steven Greenblatt, Marvelous Possessions; Gerbi, Nature; and
John H. Elliott, The Old World and the New (Cambridge: Cambridge University Press, 1970).
60. On this topic see Grove, Green Imperialism, 32ff.
61. In 1564 and 1565, the botanist Carolus Clusius (1526-16°9) visited Spanish botanical
gardens, such as Simon de Tovar's well-known garden. Later he would receive samples of
plants and curiosities from Spain for his books. For Clusius's visit to Tovar's garden, see Car-
olus Clusius, Rariorum Plantarum Historia (Antwerp, 1601), 50: 2. 173. For his contacts in
Spain, see Asso, Hispaniiensium atque Exterorum Epistolae cum praefatione et notis Ignatii de
Asso (1793),53-7°. The physician Nardo Antonio Recchi brought the work of Dr. Francisco
Hernandez on American plants and animals to Italy in the late sixteenth century. Recchi's
summary of Hernandez's work was published by the Academia dei Linceii between 1630 and
1651. See Raquel Alvarez-Pelaez, "La obra de Hernandez y su recuperacion ilustrada," in La
Real Expedici6n Botanica a Nueva Espana, 1787-1803 (Madrid: Consejo Superias de Investi-
ganes Cientificas, 1987), 156 n. I, as well as her article, "La historia natural en la segunda
mitad del siglo XVI: Hernandez, Recchi y las relaciones de Indias," in Nouveau Monde et
Renouveau de L'Histoire Naturelle, Vol. 3, ed. Marie-Cecile Benassy et al. (Paris: Presses de la
Sorbonne Nouvelle, 1994).

Local Herbs, Global Medicines 181


Merchants and Marvels
Hans Jacob Fugger and the Origins of the Wunderkammer

MARK A. MEADOW

TRADE, TRAVEL, AND THE PROCUREMENT OF CURIOUS OBJECTS

I n the cultural world of sixteenth-century Europe, few institutions offer a


more compelling venue to study the intersection of art, nature, science,
and economics than the Kunst- and Wunderkammern (commonly in Eng-
lish: curiosity cabinets) of such figures as the Wittelsbach Duke Albrecht V of
Bavaria or the Habsburg Holy Roman Emperor Rudolf II ofPrague. 1 These
Wunderkammern served many functions within the great courts of transalpine
Europe, being not only instruments of diplomacy and display, but also prag-
matic tools of economic statecraft, repositories of ready funds for unexpected
wars and disasters, sites of cultural and technological production, and active,
functional, and practicallahoratories for a variety of crafts and disciplines.
In order to think ahout the role of trade objects in these collections, let
me begin by posing a scenario and a question. I n these great collections of
Albrecht V, Rudolf II and others, visitors encountered a vast and marvelous
range of naturalia (natural objects) from across the globe: narwhal tusks
from near the Arctic, camel bezoars from the East, ivory and ostrich eggs
from Africa, birds and featherwork from the New World. Artificialia
(works of human craftsmanship) were also present in various forms, includ-
ing Limousin enamel, majolica pottery, and Venetian glass. Other human
artifacts came from well beyond the European world: Indian and Turkish
carpets, Tunisian textiles, carved African ivory, South American and Mexi-
can gold and featherwork, Syrian metalwork, shoes from Lapland, and
kayaks from Greenland, to name but a few. The question I want to raise is
a simple one: How did these varied and heterogeneous things make their
way to the Munichs or Pragues of the sixteenth century?
The answer is, of course, much more complicated than the question. In
this essay I will address one part of that answer, speaking to the crucial role
played by the extraordinarily wealthy and powerful merchant-banking fam-
ilies-such as the Fuggers, Welsers, or Medicis, but especially the first-in
bringing these sorts of objects and materials from their points of origin to
the courts and collections of Europe. In particular, I will discuss one mem-
ber of the Fugger family, Hans Jacob Fugger, 1516-75, and the part he
played in the material and conceptual formation of the Wunderkammer.
We know, for example, of several instances in which the Fuggers were
involved in the procurement of particular objects, as for example two ivory
caskets they secured from Ceylon via Lisbon, and provided for Albrecht V
in 1566.2 Ivory importation as a raw material was a substantial business in
itself, with the Fugger offices in Antwerp making a contract in 1548 to
exchange 6,750 hundredweight of brass rings (the family controlled enor-
mous copper mining resources) for large shipments of ivory from Benin, to
be crafted into fine objects for resale. 3 Max Fugger, in the 1560s, was very
active in procuring gemstones, finished necklaces, and other pieces of jew-
elry for Albrecht V, both for his own collections and as gifts to figures such
as King Philip II of Spain. 4 In the latter part of the century, via Antwerp and
through its connections in India, the House of Fugger imported monkeys,
parrots, peacocks, wildcats, and other live animals; orange trees, almond
trees, rosemary bushes, and other live botanicals; camphor, pearls, leopard
skins, indigo, gemstones, and similar natural by-products. 5 These examples
of naturalia, living or not, ended up in the Fuggers' own collections as well
as those of their patrons. Craftsman were employed to work on the raw
materials, transforming them into finished products to be sold or given to
wealthy and noble clients. In a similar vein, the Fuggers may have been the
means by which the merchant-scholar Philipp Hainhofer gained access to
South American objects for his own Kunstkammer, again through the Fug-
ger offices in Portugal. 6
The question of how the Fuggers, or other firms like theirs, contributed
to the procurement of exotica for these collections, and the implications
thereof, is not a trivial one. The objects collected in Wunderkammern, espe-
cially the exotica, flooded in from throughout the known world, and even at
times from beyond it. In an era before the establishment of disciplines such
as zoology or botany, ethnography or anthropology, the stories these objects
told derived in no small part from the biographies they acquired moving
from hand to hand. Their original contexts, uses, and narratives were fil-
tered through the numerous people involved at each stage of their journey.
The Fuggers and their representatives, and those of other families, were
critical participants in the life histories of these objects. Furthermore, the
diverse interests of the Fuggers themselves inevitably affected the procure-
ment process. Certainly business concerns were paramount, and an eye was
always kept on the bottom line. But the Fuggers, by the time of Hans Jacob,

Merchants and Marvels


were intimates of dukes, kings, and emperors, serving at times as courtiers,
advisors, financial consultants, and bankers. They were humanists and schol-
ars, trained at the finest universities in Europe. They were also avid collectors
of books, ancient coins, exotica, musical and mathematical instruments, fully
conversant with the thematic collecting interests of their clients. And perhaps
most important, they combined their intellectual and acquisitory pursuits
with the practical matters of running a business and communications empire.
The sheer vastness of their wealth, of their land holdings, and of their power
made them coequals in many ways with the nobility they served.
The Fugger's own collecting practices are central to understanding their
role in relation to princely collections. In this period, collecting by its very
nature was a communal activity. Princes, scholars, merchants, or apothecaries
assembled their collections through complex systems of exchange, gift giving,
commerce, patronage, and other forms of social and financial intercourse. To
some extent, the activity of collecting provided a social nexus, in which noble,
scholar, tradesman, and even craftsman could participate in the same realm. By
participating in this system as collectors as well as purveyors, the Fuggers and
other such families placed themselves within an intellectual and social milieu
that furthered much more than their business goals. As a result, at least in part,
the Fuggers gained the rank of minor nobility, status as legitimate scholars and
humanists, and a role as patrons of the arts, scholarship, and technology.
Lorraine Daston and Ken Arnold have both stressed the importance of
travel in relation to curiosity cabinets. As Daston has written, "Travel was
the alpha and omega of collecting, being both the source of the bulk of the
objects-voyages of exploration and subsequent trade with the newly dis-
covered lands created a steady flow of exotica-and the occasion for
inspecting them in Amsterdam, Oxford, Venice, Paris, Augsburg, Uppsala,
or wherever the curious and peripatetic tourist might land."7 Arnold dis-
cusses the close affinities between travel and collecting, noting that "the rela-
tionship between the two was precisely reciprocal: one traveled in order to
collect, but also one collected in order to travel."8 Indeed, as Arnold notes,
viewing a curiosity cabinet was itself a microcosmic form of travel, "through
a world brought back and reassembled in a cabinet."9 I would like to deviate
slightly from the general question of travel and instead consider the impor-
tance of trade in the formation of the cabinets. The two issues are closely
intertwined, but it is especially instructive here to consider the commercial
aspects of travel. Intellectual history often places trade, and the commercial
world in general, in the silent shadows when considering perceptions and
models of the cosmos. lO We can use the Fuggers and Wunderkammern to
bring to light some of the implications of their close relationship.
In particular, I will discuss the relationship of sixteenth-century commer-
cial networks to both the material and the conceptual formation of micro-

Mark A. Meadow
cosmic collections north of the Alps. In the course of the larger research pro-
ject here outlined, I want to situate this commercial world within the broader
framework of similar networks, such as the travel and epistolary networks of
the humanists; the closely related networks joining together universities and
their faculty and students; and the tightly woven fabric of Europe's courts,
stitched together through intermarriage and political alliances.
These various skeins of interrelationship are themselves interdependent
and interwoven. Hans Jacob Fugger stands as an excellent example of this,
bringing together in a single individual the worlds of the university, human-
ism, the courts, and commerce. As we will see in greater detail below, Fug-
ger plays not one, but several significant roles in the founding moments of
the Wunderkammer: he was himself a patron, scholar, and collector, but also,
following a financial reversal, he worked as a librarian and procurer of
books and objects for one of the very earliest of the collections under consid-
eration, that of Albrecht V, elector of Bavaria.
I have been pushing very hard at the metaphors of threads and cloth in
order to allow myself to bring in an enlightening fable from Italo Calvino.
In his Invisible Cities, he includes, under the rubric of "trading cities," a place
called Ersilia. In this city,

to establish the relationships that sustain the city's life, the inhabitants
stretch strings from the corners of the houses, white or black or gray or
black-and-white according to whether they mark a relationship of blood,
of trade, authority, agency. When the strings become so numerous that
you can no longer pass among them, the inhabitants leave: the houses are
dismantled; only the strings and their supports remain. ll

The web of many-colored strings that Calvino describes serves as a map of


relationships, and those relationships are themselves markers for what we
would now call data flow. Reconstructing such threads may therefore serve
as a critical analytical tool. We might well say that the houses of the sixteenth
century have been dismantled, and certainly the inhabitants of the period
have long since departed. But we can still make out the tapestry of social and
cultural life that remains in the various strands of relationship. While calling
attention to one color of string, the gold and silver strings of commerce, I
will also be touching on many of the others. Indeed, in the world of the
curiosity cabinet, as in any other place or period, the strands of blood, trade,
authority, and agency were all present, and were all interdependent. To
touch any of these strands is to set all the others in sympathetic vibration.
The essay that follows, I should note, is more a blueprint for an ongoing
research project than the finished results of one. Tracing anyone of the
social, intellectual, economic, or political networks of the period is itself a

Merchants and Marvels 185


daunting task. Examining all of these, and their points of intersection, is
Herculean. Nonetheless, if we are truly to understand the conceptual matrix
in which the Wunderkammer functioned, this line of research must be pur-
sued. The larger study that will develop from this report is a component of
a major research initiative undertaken within the University of California.
Called "Microcosms: Objects of Knowledge," this broader project is exam-
ining the history, functions, and future of material collection~ in the contem-
porary universityY As is appropriate for a place that calls itself a university,
these vast holdings are universal in scope, and when examined holistically
bear a surprisingly close resemblance to the range of objects found in a Wun-
derkammer. The Microcosms Project considers these collections, and the uni-
versity itself, from an historical perspective, turning especially to the
sixteenth century, and the Wunderkammer, as one point of origin for the
modern university and its collections. As we are all aware at the beginning
of the twenty-first century, the university is itself intimately bound into a
vast range of networks, including those of commerce. Nor is it any coinci-
dence that I have chosen to emphasize metaphors of threads, networks, and
skeins, considering how intimately bound the worlds of information and
trade have become in the digital community that we have come habitually to
call the "net" or the "web."

THE HOUSE OF FUGGER

Before turning to our particular Fugger, Hans Jacob, I will briefly sketch
some of the family history. The Fuggers arrived in Augsburg in I367 in the
person of one Hans Fugger, a clothmaker who appears to have founded the
family business by importing his own raw materials, rather than relying
upon local merchants for them. The firm developed slowly in the next gen-
eration, first through Hans's sons Andreas and Jacob I, and then through
Jacob's capable widow, Barbara Basinger, who ran the business until her
children were of age to assume control themselves. From the widow Fugger,
the business passed on first to her eldest and youngest sons, Ulrich and
Georg. Ulrich and Georg first established a Fugger presence in the German
merchants' trade building in Venice, the Fondaco dei Tedeschi (German busi-
ness house).
Finally in 1485, the middle child, Jacob II, who was later to become
known as Jacob the Rich, entered the management of the business by taking
charge of the Innsbruck office and aggressively pursuing mining opportuni-
ties in the Tyrol. Jacob II had trained in bookkeeping and other aspects of
business management at the Venetian Fondaco since I478. Ulrich, Georg,
and Jacob II formed a trade partnership in 1494, in which the business theo-

r86 Mark A. Meadow


retically was equally shared among them. Although he did not formally
become the director of the company until Ulrich's death in 15IO, it was Jacob
II who contributed most to creating the enormous fortune that catapulted
the Fuggers into the center of sixteenth-century European commerce and
political power. Jacob II, surviving his two brothers and desiring to extend
his control over the firm, later changed the partnership model of governance
to that of a single, autocratic director.
Jacob's basic strategy was a very simple but highly effective one. He lent
money to the ruling houses of Europe at very high rates of interest, with the
loans secured against the income-generating resources that such dukes,
princes, kings, popes, and emperors could provide: silver and copper mines,
agricultural communities and land, and so forth. A frequent formula
employed by Jacob II was to lend large sums of money on a long term basis,
with the stipulation that until the loan and its interest were repaid, the pro-
duction of a particular site, usually a mine, went directly to the Fuggers. In
certain instances, should the loan not be repaid in the allotted time, the
secured property reverted permanently to the Fuggers. In this way, massive
amounts of capital were quickly acquired, together with more stable sources
of future earnings in the form of real property. Jacob built especially upon
the relations his brother Ulrich had already established with the Habsburgs;
the Fuggers, for instance, were one of the main, perhaps even the single
most crucial, resource in assuring the election of Charles V as holy Roman
emperor in 1519. This same connection to the Habsburgs eventually led to
the fading of the Fugger family star, as improperly secured loans and even
private family capital went to sustain Charles and his son Philip II, king of
Spain, only to result in Habsburg defaults in 1557, 1574, 1575, and 1596. The
first of these financial crises was among the main reasons why Hans Jacob
Fugger was eventually forced out of the family business and into the service
of Albrecht of Bavaria and his collections, thus indirectly setting into motion
the story with which we are concerned.
From the time of Jacob Ion, as the business grew increasingly large and
complex, the Fuggers followed a strategic program of education for their
offspring to ensure that the successors to the firm were properly prepared in
languages, mathematics, law, and the humanist cultural background that
would allow them to converse with the ruling elite. At first this involved
sending them to the Fugger outposts in Europe, initially, as in the case of the
young Jacob II, to the Fondaco dei Tedeschi in Venice, but eventually young
Fuggers set out on a grand tour throughout the continent. A bit later, in the
time of Hans Jacob's youth, this period of educational preparation also
specifically included university education. Here is one of the key points
where, for the Fuggers and other such families, the worlds of commerce and
humanism intersected.

Merchants and Marvels


A business of the scale of the Fuggers', spread across all of Europe and
into the vast lands beyond, necessitated efficient lines of communication.
The Fuggers operated what are called "factories," that is, places of business
run by individuals authorized to conduct business for the family, the "fac-
tors," throughout Europe and beyond. Among the places in which the Fug-
gers had offices were Lisbon, Seville, Madrid, Saragossa, the Tyrol, Vienna,
Innsbruck, Munich, Leipzig, Nuremberg, Frankfurt, Cologne, Antwerp,
Amsterdam, Paris, Lyon, Strasbourg, London, Helsingor, Malmo, Danzig,
Riga, Narva, Poznan, Warsaw, Kracow, Ofen, Breslau, Pest, Venice, Rome,
Florence, and the Levant. Through these sites, and many, many others too
numerous to list, the Fuggers acted as brokers for trade flowing throughout
the known world. In fact, Fugger factors were located in virtually every com-
munity of any economic significance in Europe, which means down to locales
of only a few thousand residents, with (largely for political reasons) a some-
what more modest presence in France and Spain. But the Fugger mercantile
operations were by no means limited to Europe. The Fuggers were involved
in all areas of international trade, including between the New World and
Europe, where they had offices in Santo Domingo, the Yucatan peninsula,
Brazil, and elsewhereY In 1531, the same year that Pissaro reached Peru, the
Fuggers were granted a contract to colonize and economically exploit the
western coast of Latin America from the southernmost point of Pizzaro's
dominion to the tip of Tierra del Fuego. 14 They financed commercial ven-
tures in India and Ceylon through Lisbon, traded in goods and slaves from
Africa, and broke red merchandise from as far away as East Asia.
All of the Fugger outposts were expected to stay in constant touch with
the home office, and, where necessary, with each other. This led to the cre-
ation of a very efficient system of communication, with letters flowing con-
stantly in and out of the home office; paralleling a similarly efficient
mechanism for transporting goods in bulk. 15 Primarily from library pur-
chases, the facet of Fugger collecting that has been most thoroughly
researched, we know that acquisitions were made along these lines of com-
munication. 16 As business letters were sent out from Augsburg, they would
include requests to secure one or another book or object, which in turn
would accompany the reply, bringing the desired purchase safely home in
very short order.
If we think only of two of the major Fugger factories, those in Venice and
Antwerp, we can begin to understand something about the position the
Fuggers were in to tap into the fullest range of natural objects and human
artifacts as they traveled along the veins and arteries of the early modern
commercial world. Venice remained the single most important port in
southern Europe, with particular connections to Africa, the Middle East,
and on into Asia. Antwerp played a similar role in the North, being the cen-

188 Mark A. Meadow


ter of commerce with Scandinavia, the Baltic, England, the Iberian penin-
sula, and various Spanish and Portuguese territories in Africa and the New
World. The Fuggers quickly came to dominant positions in both of these
markets, and between the two had access to virtually any materials or goods
that could be commercially transacted. This is the beginning of the answer
to the question I posed at the start of this chapter; important research
remains to be done on the specific relationship between the formation of any
given princely collection and the financial/mercantile web with which that
court was linked. If we take a Kunst- or Wunderkammer to be a representa-
tion of the world, a microcosm, then the accessibility of particular markets,
trade routes, and therefore particular objects will have a direct bearing on
the model of the world thus created.

HANS JACOB FUGGER

As I turn to Hans Jacob Fugger, I also shift to a different aspect of the ques-
tion. That is, however important the biographies of objects are to their sig-
nificance within a collection, and however central the Fuggers are to the
formation of those biographies, in the person of Hans Jacob we come to a
figure who had a direct bearing upon the activity of collecting as a necessary
part of sixteenth-century statecraft and also upon the ordering systems
employed in these collections.
Hans Jacob was born in December 1516, the son of Raimund Fugger and
nephew of Anton, who was then in charge of the company. I ? Anton and
Raimund were the children of Georg Fugger, the brother of Jacob II, who
had died childless. Raimund was a renowned collector of antiquities and
encouraged his son in similar directions. Hans Jacob had an unusually thor-
ough education, with studies in Germany, Italy, France, Spain, and the
Netherlands. His education was primarily classical and linguistic-he was
apparently fluent in Latin, Italian, French, Spanish, Polish, Hungarian, and,
in all likelihood, Dutch as well. I8 Among those with whom he studied were
Wolfgang Bosch, later tutor to Albrecht V of Bavaria, and Johannes Secun-
dus, later court humanist to Margaret of Austria and Philip of Burgundy. 19
His advanced studies, first with Viglius Zwichem van Aytta, were oriented
toward law. 20 He followed Viglius from Dole to Bourges. While in Bourges he
also studied with Andreas Alciati, and appears already to have become a bib-
liophile, lending Alciati his own copy of Titus Livius. 21 From Bourges he
moved to the university in Padua, and then to Bologna by 1534, where he was
named syndic of the German Trading Nation while still a student.
It is worth briefly looking at some of Hans Jacob's fellow students, who
make a very impressive list. These include such humanist scholars as

Merchants and Marvels


Hieronymus Wolf, translator of Demosthenes; Sigmund Gelenius, transla-
tor of Josephus; and Roger Ascham, tutor to Elizabeth I of England. 22 In
Bologna, Hans Jacob counted among his fellow students Alessandro Far-
nese, duke of Parma, governor general of the Netherlands, cardinal, and
patron of the arts, and someone who was later to name Fugger his own
"patrone"; Christopher Madruzzo, later bishop of Trient and Brixen, cardi-
nal, and governor of Milan; Stanislaus Hosius, later cardinal and bishop of
Augsburg; Otto Truchsess von Waldberg, also a future bishop of Augsburg;
and Wiguleius Hund, later chancellor to the court of Bavaria.23
Hans Jacob Fugger was an author in his own right, drafting a history of
his own family in 1541-45 and a history of the Habsburgs in 1555, which
itself was a sort of collection, containing scores of portraits, images of places,
genealogies, images of monuments, insignia, and thousands of coats of
arms.24 This assemblage of images and charts, in fact, corresponds very
closely to the first of five classes in Samuel Quiccheberg's Inscriptiones vel tit-
uli Theatri amplissimi (Inscriptions or headings of the most complete the-
ater), a treatise on collecting and an organizational plan for Albrecht V's
collections, and the earliest known treatise on museums. 25 Hans Jacob was
furthermore a prolific patron of scholarship: Maasen lists more than sixty
works dedicated to him, including Sigismund Gelenius's work on Flavius
Josephus, Conrad Gessner's work on libraries and ordering systems, Jacopo
Strada's scholarship on antiquities, Panvinius's history of the church and
four works of Hieronymus Wolf.26
In 1535, Hans Jacob's father died, and he was summoned back to Augs-
burg by his uncle Anton to assume his position in the firm. While Hans
Jacob was contractually given the equivalent of second-in-command of the
family firm, Anton decided he was still not quite ready to take up the reins.
So Hans Jacob was sent on a second tour, now specifically oriented toward
familiarizing him with the business. Beginning with an extended stay in
Antwerp, he traveled very widely among the Fugger factories in Europe.
Before his return to Augsburg, Hans Jacob entered service in the court of
the Habsburg Ferdinand I, then king of Bohemia, brother to Holy Roman
Emperor Charles V, and from 1556 emperor in his own right. While at Fer-
dinand's court, Hans Jacob served as tutor to Ferdinand's children, including
Maximilian, holy Roman emperor following his father's death in 1564, and
Archduke Charles. These figures loom large in the history of the Wun-
derkammer. Ferdinand is traditionally credited with founding the Habsburg
Kunstkammer, with a particular interest in mechanical devices such as clocks
and in antiquities and coins. Maximilian II continued his father's interests in
collecting, especially antiquities and coins, but also artifacts of natural history.
He later hired Jacopo Strada as architect of his collections, both conceptually
and literally, as Strada designed the first Habsburg buildings devoted to the

Mark A. Meadow
family collections. The more famous collections of Maximilian's son Rudolf
II built upon these foundations. The young Hans Jacob was introduced to
more than just princes and princely collections while in Ferdinand's service;
he also met his first wife, Ursula von Harrach. On 2I June I540 the couple
was wed, with the chief steward of Charles V's court in attendanceY
Fugger's return to Augsburg in I540 heralded a lengthy period of involve-
ment in politics and government, which included membership in different
parts of the Augsburg city councils, and a stint as a mayor. His connections
with the Habsburgs were exploited by the city during the religious troubles,
with Hans Jacob being sent out on more than one occasion to mollify Cardi-
nal Granvelle or the emperor himself.28 Cardinal Granvelle and Emperor
Charles V both were later houseguests of Hans Jacob while visiting Augs-
burg. After having attended at least two Reichstage, Hans Jacob reached the
pinnacle of his political career in I549, when he was named imperial coun-
cilor by Charles V. At some point during this period, Hans Jacob also devel-
oped very close ties to Albrecht of Bavaria, indicated by records of Albrecht
serving as godfather to some of Hans Jacob's twenty-one children.
In 1560, Anton died, leaving the business in the hands of Hans Jacob and
his brothers. Anton, largely through circumstances beyond his control-the
Habsburg financial crises mentioned earlier-left the business in compara-
tively bad shape. Things were not to improve under Hans Jacob, who by
1565 found himself personally bankrupt and the family fortune in not much
better shape. Hans Jacob did not even have enough funds to cover his own
tax debts, despite selling off most of own possessions. His friendship with
Albrecht here paid off, when the latter personally extended him the money
required to stave off complete disaster. Albrecht mediated negotiations
between Hans Jacob and the rest of the family, which resulted in a return of
the business to Anton's own children and Hans Jacob's permanent removal
from the firm. One of the terms of the agreement appears to have been that
Hans Jacob enter the service of Albrecht as court librarian, here returning to
his earlier and perhaps more temperamentally suitable profession as human-
ist and scholar. 29
Serving as a librarian must have suited Hans Jacob well. Even before his
time, the Fugger family library was quite famous, but he had turned it into
a collection nearly without equal. The Fugger library ranged from the latest
vernacular books off the presses of all Europe, and a very comprehensive set
of classical texts, to medieval, Byzantine, and even Syrian manuscripts. 30 In
his heyday, Hans Jacob had hired as librarians and curators such individuals
as Hieronymus Wolf; 31 Jacopo Strada, who may have had his first signifi-
cant employment from Fugger;32 and Samuel Quiccheberg, who would
move into Albrecht's service at the same time that his former employer did
and would write the Inscriptiones vel tituli there in I565.33

Merchants and Marvels


Given the later activities and careers of Strada and Quiccheberg, we must
seriously consider their common link to Fugger. The Fugger collections
were not by any means limited to books. We have already noted Raimund's
interest in antiquities, which Hans Jacob continued. 34 Raimund also had a
considerable interest in musical instruments, with a passion above all for the
lute, and amassed a collection of instruments so vast it is difficult to imagine
how it could be stored." The Fuggers were active patrons of the arts who
commissioned sculpture, architecture, and large numbers of paintings from
artists such as Hans Maler and Titian. 36 Hans Jacob, working with Strada,
acquired a very significant collection of antique coins, which put them in an
excellent position to produce jointly a thirty-volume catalog of drawings of
ancient coins, an enormous undertaking. l 7 Another Fugger, Marx, had an
avid interest in mathematics and collected mathematical instruments. We
know much less about the family's interest in naturalia, this being virtually
unexplored territory in Fugger studies. The Fuggers collected gemstones
and jewelry, including four pieces originally from the Burgundian treasury,
and not infrequently resold them to such clients as the Habsburg emperors,
the Medicis, and the sultan of Turkey.ls We know of individual purchases of
coral objects and so forth, but much work remains to be done on the full
extent of Fugger collecting. 39 Antiquities, coins, gems, and scientific instru-
ments are among the objects that formed the core of any humanistically ori-
ented collection, that is, the type of collection we find among the nobility of
Europe north of the Alps.
While in Hans Jacob's service, Quiccheberg apparently devised a thematic
ordering system for the Fugger library, perhaps in direct collaboration with
Fugger himself, based upon the work of Conrad Gesner.40 Once Quiccheberg
and Fugger moved to the court in Munich, they put into place a similar cata-
loguing system for Albrecht's library. This system certainly formed the con-
ceptual basis for the organizational scheme Quiccheberg developed for the
Wittelsbach collections and presented in his Inscriptiones vel tituli.
The exact circumstances of Strada's and Quiccheberg's employment and
activities under Hans Jacob Fugger remain to be researched, as does a
detailed study of the collecting activities of Fugger under Albrecht V. But it
should already be clear that Hans Jacob played a key role in the formation of
the Kunst- or Wunderkammer in the Germanic territories. Albrecht V's col-
lections, while predating the arrival of Fugger, underwent a massive
increase in scale and a change in nature just as his association with them
begins. And the Bavarian collections, while not the very earliest in the
region, Ambras in particular predating them, are the first to lay claim
through their variety and their mode of display to being a site for the study
and accumulation of universal knowledge, and to the practical application of
that knowledge in the governance of the state. The more famous collections

Mark A. Meadow
of Maximilian in Vienna and Rudolf II in Prague are conceptually related to
those of Munich, which should hardly be surprising given the role of inter-
mediary played by Jacopo Strada. It is tempting to think, and may well be
the case, that two qualities of the Munich collection relate directly to the con-
nection with Fugger: here I speak of the collection as systematically
arranged, and of the clear imperative to put the collection to direct practical
use. The Fugger's own collections shaded imperceptibly into the conduct of
their business, with the library being the clearest indication: the Fuggers
effectively compiled two different sorts of library. One was a magnificent
example of a humanist and antiquarian collection, primarily of intellectual
and aesthetic interest, with precious manuscripts, a remarkably complete set
of standard and obscure works of the Greeks, the Romans, Patristic texts,
and works of humanist scholarship. The other library was a more immedi-
ately pragmatic reference source for business, that would have contained
atlases and travel literature, treatises on accounting, mining, law, and so
forth. On a more conceptual level, we would do well to recognize that the
vast amount of data necessary to the running of the Fugger firm, in the form
of business records, newsletters, and other accounts of current events, inven-
tories, and even the commercial goods themselves, presented a powerful
challenge in terms of efficient storage and retrieval. Hans Jacob Fugger's
interest in ordering systems had a practical origin as well as a humanist one.

COMMERCE IN THE CABINET OF CURIOSITIES

In closing, I would like to think a bit about commerce in the Wunderkammer.


Certainly the financial basis of the collections of the nobility is clear. The ear-
liest of these assemblages were neither collections of art nor of curiosities,
although they may well have contained such objects, but rather were trea-
suries. Gold and silver work, gems and jewelry, even reliquaries were assets
that could be and were sold off to raise ready cash. In the case of Munich, in
1565 (not coincidentally the year Quiccheberg's treatise was published, and the
year Hans Jacob Fugger entered Albrecht's service) Albrecht V was the first
Wittelsbach ruler to declare certain objects, including two narwhal tusks, the
inalienable property of the dynasty, to be passed on to later generations with
the stipulation that they never be sold. 41 By and large, collections of the Wun-
derkammer type continued to emphasize monetarily valuable objects, most of
them acquired through mercantile houses like those of the Fuggers. These
objects moved in and out of the collections in an economy of their own, both
to generate cash and to cement relationships with other princes and scholars.
The Wunderkammern served also as repositories for intellectual capital,
functioning in a not dissimilar way to universities and their collections

Merchants and Marvels 193


today. Antiquities and numismatic collections were fundamental resources
for study of the histories, languages, and cultures of the ancient world. 42
Products of metalsmiths, turners, jewelers, and armorers served to stimulate
technical developments both by injecting capital into the higher ends of pro-
duction and by acquiring examples of nonlocal techniques. Wunderkammern
and other such collections similarly encouraged the development of scien-
tific instruments, and therefore of scientific endeavor. We can even think of
the Wunderkammern as assemblages of examples of local and exotic raw
materials and processes; there is for instance a consistent interest in mining
to be found in these collections - think back to the origins of Fugger
wealth, and noble wealth, in mines for precious ores.
Objects were not the only items collected in Wunderkammern. Scholars,
craftsmen, and other specialists were just as eagerly acquired. Jacopo Strada,
for instance, begins working with Fugger, is brought into Albrecht's service,
and ends up with Maximilian. Quiccheberg, a physician by training, moves
from a university setting at Ingoldstadt to work for Fugger and then
Albrecht. Hans Jacob Fugger himself was acquired by Albrecht V in what
was essentially a financial transaction between Albrecht and the house of
Fugger. These were the elite of the collections personnel, but there were also
printers and bookbinders, turners and jewelers, armorers, equerries, and
others attached to the service side of the collections.
This brings us to another very interesting point that emerges from look-
ing closely at the infrastructure of Albrecht's collections, using Quic-
cheberg's text as a lens to help us focus . Almost all attention to these
collections has gone into reconstructing first their contents and then their
arrangement. Scholars have looked at Quiccheberg, deservedly so, primarily
for the intriguing evidence he presents about what was collected and about
how it was ordered. But the system of classes and inscriptions for objects is
only a part of his treatise. He argues also for conceiving of the collections as
one part of a larger complex of workshops and ateliers, including a printing
room, a mint, and a pharmacy.43 In turn, this brings us to a question of
access: we expect the collections to be open to the duke and his family, to dis-
tinguished guests of the house and to resident and visiting scholars of high
repute. But it is also clear that craftsmen and artisans could make use of it as
well, for the good of regional technology and the economy.
Let me end with a quote from Quiccheberg:

For I sense that it cannot be expressed by any person's eloquence how


much wisdom and how much use for administering the state- in the
civil and military spheres and the ecclesiastical and literary-can be
gained from examination and study of the images and objects that I have
described. 44

194 Mark A. Meadow


This passage is found in the section of Quiccheberg's treatise in which he
amplifies on the purposes and structure of the collection. In this passage and
the one immediately preceding, he compares and contrasts the accumulation
of objects with the training suggested by Cicero (the epitome of eloquence for
the Renaissance) for the ideal orator. That person should be able to enumerate
and learn about all things because all in the world that pertains to mankind is
the natural domain of the orator. The collecting of objects, as enumerated by
Quiccheberg, is an equivalent to the Ciceronian collecting of knowledge and
ideas. But the equivalence is not exact, because however eloquent an orator
may be, be it Cicero himself, the pragmatic value of the collection could not be
conveyed as well by him as it could by the objects themselves.
And it is with the question of the practical use of the princely Wun-
derkammern that we reach the heart of the matter: these collections aided in
the "administering of the state," the business of governance, and did it in
very pragmatic ways. Certainly they were means of projecting images of
princely wealth, power, erudition, and identity, but they were also very prac-
tical repositories of practical knowledge. As I have argued here, we must see
this insistence on practicality as having origins in the dual concerns of mer-
chants such as the Fuggers. It was they who dealt with and dealt in all the
material things of the world, they who acquired these objects for themselves
and for their clients, they who had the necessity and experience of ordering
the world for both business and scholarly ends, they who best understood
how these were the warp and woof of collecting, and how the two combined
to create a representation, a self-portrait, of the collector. Certainly com-
merce and the merchants who practiced it largely determined which objects
ended up in the Wunderkammer, and helped create and convey their life his-
tories as well, points that deserve more scholarly attention. But we must also
recognize that they played a strong hand in founding and shaping the Wun-
derkammer itself, bringing the threads of their long-established habits of
organizing repositories into the tapestry. In this regard I think it is telling
that Quiccheberg and Strada both got their start working with Hans Jacob
in the Fugger collections, where business came first, and that Albrecht V's
collections only started to take their full form as a Wunderkammer with the
arrival of Fugger. Hans Jacob Fugger, rather like the Zelig of sixteenth-
century collecting, turns up surprisingly often in the picture of the founding
moments of the Wunderkammer.

Merchants and Marvels


Many of the conceptual underpinnings of this paper derive from discussion and debate with
the members of the Microcosms Residential Research Group, which convened at the Univer-
sity of California Humanities Research Institute from January to June 1999: Ken Arnold,
Rosemary Joyce, Rebecca Lemov, Sonnet Retman, and Bruce Robertson. lowe all of these
colleagues and friends a great debt. As a visitor to the UCHRI Residency, Dirk Jansen was
instrumental in bringing the significance of Hans Jacob Fugger for the Wunderkammer to my
attention, and has since been an invaluable resource. The writing of this chapter was aided
enormously by my indefatigable research assistants, Emily Peters and Amy Buono. This
research was made possible by a Getty Grant Program Senior Collaborative Research Grant
and funding from the University of California Office of the President, the Interdisciplinary
Humanities Center of the University of California Santa Barbara, and the UCSB Committee
on Research.

I. I introduce these terms from the beginning as a gesture toward the pioneering research

on these early collections written by Julius von Schlosser. It is important to note that collec-
tions of the period were quite diverse in form and function, and we should be cautious about
speaking about them as a unified phenomenon. For the sake of convenience, I shall use the
term Wunderkammer for the remainder of this essay, recognizing that its use was infrequent
in the period under consideration. For the purposes of this paper, I use the term to refer to
heterogeneous collections that aspired to produce, store, and represent universal knowledge.
Julius von Schlosser, Die Kunst- und Wunderkammern der Spiitrenaissance, Vol. I I of Mono-
graphien des Kunstgewerbes (Leipzig: Klinkhardt & Biermann, 1908).
The literature on curiosity cabinets, Kunstkammern, Wunderkammern, and the many other
equivalent forms of early modern collecting is too vast to be given here. An excellent intro-
duction to the topic and overview of many of the most important collections can be found in
Oliver Impey and Arthur MacGregor, eds., The Origins of Museums: The Cabinet of Curiosities
in Sixteenth- and Seventeenth-Century Europe (Oxford: Clarendon Press, 1985). For Albrecht
V's collections, see especially Lorenz Seelig, "The Munich Kunstkammer,"1565-1807, in the
volume just mentioned; Jacob Stockbauer, Die Kunstbestrebungen am Bayerischen Hofe unter
Herzog Albrecht V. und seinem Nachfolger Wilhelm v., vol. 8 of Quellenschriften fiir Kunst-
geschichte und Kunsttechnik des Mittelalters und del" Renaissance (Vienna: Wilhem Braumiiler
Universitats-Verlagbuchhandlung, 1874); and Herbert Brunner, Die Kunstschiitze del" Miinch-
ner Residenz (Munich: Siiddeutscher Verlag, 1977). For Rudolf II's collections, see Robert
John Weston Evans, Rudolf II and His World: A Study in Intellectual History: 1576-1612
(Oxford: Oxford University Press, 1973); E. Fucikova, Die Kunst am Hofe Rudolfs II (Prague:
Artia Verlag, 1988); and Thomas DaCosta Kaufmann, The School of Prague: Painting at the
Court of Rudolf II (Chicago: University of Chicago Press, 1988).
2. See Seelig, "The Munich Kunstkammer," 83.

3. Michael Gorgas, "Animal Trade between India and Western Eurasia in the Sixteenth
Century -The Role of the Fuggers in Animal Trading," in Indo-Portuguese Trade and the
Fuggers ofthe Sixteenth Century, ed. Kuzhippalli Skaria Mathew (New Delhi: Manohar, 1977),
195-225, esp. 218-222.
4. Stock bauer, Die Kunstbestrebungen am Bayerischen Hofe, 91 - 107 passim.
5. Gorgas, "Animal Trade," 218-222.
6. See Hans-Olaf Bostrom, "Philipp Hainhofer and Gustavus Adolphus's Kunstschrank in
Uppsala," in Origins ofMuseums, 90- 101, here 91. Hainhofer's cabinet, here more literally a piece
of furniture, was later purchased and presented to King Gustavus Adolphus, thus illustrating the
migration of objects from source to merchant-trader to merchant-collector to royal collection.

Mark A. Meadow
7· Lorraine Daston, "The Factual Sensibilitys" Isis 79 (1988), 455.
8. Ken Arnold, Cabinets for the Curious: Practicing Science in Early Modern English Muse-
ums, Ph.D. diss., Princeton University, 1991 (UMI Dissertation Services, 1992), 139.
9. Ibid.
10. An admirable exception is Lisa Jardine's recent Worldly Goods: A New Hist01Y of the
Renaissance (London and New York: W. W. Norton, 1996). Her insights into the relation of com-
merce and intellectual life have helped shape the more specialized argument here presented.
11. Italo Calvino, Invisible Cities, trans. William Weaver (New York: Harcourt Brace

Javonovich, 1974),76.
12. The Microcosms Project is described in Mark Meadow and Bruce Robertson, "Micro-
cosms: Objects of Knowledge," AI & Society 14 (2000), 223-229.
13. For Fugger trade with the New World, see Karl Heinz Pan horst, Deutschland und
Amerika; ein Riickblick auf das Zeitalter der Entdeckungen und die ersten deutsch-amerikanischen
Verbindungen unter; besonderer Beachtung der Unternehmungen der Fugger und Welser
(Munich: E. Reinhard, 1928), Konrad Habler, Die Geschichte der Fugger'schen Handlung in
Spanien, Vol. I of Socialgeschichtliche Forschungen, Erganzungshefte zur Zeitschrift ftir
Social- und Wirthschaftsgeschichte (Weimer: E. Felber, 1897); Hermann Kellenbenz, Die
Fugger in Spanien und Portugal bis 1560: Ein Groj3unternehmen des 16. Jahrhunderts (Schriften
der Philosophischen Fakultiiten der Universitat Augsburg, 33:I), 3 vols. (Munich: Verlag
Ernst Vogel, 1990).
14. Panhorst, Deutschland und Amerika, passim, esp. 278-283. A shift in Iberian politics
effectively ended this venture before it began.
15. The Fugger newsletters have never been published in their entirety. Nonetheless, sev-
eral extensive sets of selected examples give a good sense of their contents and breadth of cov-
erage. See, for example, Victor Klarwill, Fugger-Zeitungen: Ungedruckte Briefe an das Haus
Fugger aus den Jahren 1568-1605 (Vienna: Rikola Verlag, 1923); Victor von Klarwill, ed., The
Fugger News-Letters: Being a Selection of unpublished letters from the Correspondents of the
House of Fugger during the years 1568-1605 (London: John Lane The Bodley Head, 1924); V.
von Klarwill, ed., The Fugger News-Letters, Second Series: Being a further Selection from the
Fugger papers specially referring to Queen Elizabeth and matters relating to England during the
years 1568-1605 (London: John Lane The Bodley Head, 1924); and George T. Matthews, ed.,
News and Rumor in Renaissance Europe (The Fugger Newsletters) (New York: G. Putnam's
Sons, 1959).
16. For the Fugger libraries, the standard references remain Otto Hartig, Die Griindung
der Miinchner Hojbibliothek durch Albrecht V. und Johann Jacob Fugger; (Munich: Verlag der
Koniglich Bayerischen Akademie der Wissenschaften, I9I7); and Paul Lehmann, Eine
Geschichte der alten Fuggerbibliotheken, 2 vols., Studien zur Fuggergeschichte, 12 (Ttibingen:
J. C. B. Mohr, 1956-1960).
17. For Hans Jacob Fugger, see Wilhelm Maasen, Hans Jakob Fugger (1516-1575): Ein
Beitrag zur Geschichte des XVI. Jahrhunderts, Vol. 5 of Historische Forschungen und Quellen
(Munich: Datterer, I922); Lehmann, Eine Geschichte der allen Fuggerbibliotheken, Vol. I, esp.
41 -73; and Hartig, Die Griindung der Miinchner Hojbibliothek, 193-223. For Anton Fugger,
see Herman Kellenbenz, Anton Fugger (1493-1560), Weissenhorn (1993) and Johannes
Burkhardt, ed., Anton Fugger (1493-1560): Vortrage und Dokumentation zumfiinfhundertjdhri-
gen Jubildum, Vol. 36 of (Studien zur Fuggergeschichte) (Weissenhorn, A. H. Konrad, 1994).
18. For Hans Jacob Fugger's education, see Lehmann, Eine Geschichte der alten Fuggerbib-
liotheken, Vol. 1,42- 44; Hartig, Die Griindung der Miinchner Hojbibliothek, 194-196; and
Maasen, Hans Jakob Fugger, 3-12.
19. See Maasen, Hans Jakob Fugger, 6. For Johannes Secundus, see Dougall Crane,
Johannes Secundus, His Life, Work- and Influence on English Literature (Beitriige zur englischen

Merchants and Marvels 197


Philologie, 16) (Leipzig: B. Tauchnitz, I93l), and Clifford Endres, Johannes Secundus: The
Latin Love Elegy in the Renaissance, (Hamden, Conn.: Archon Books, 1981).
20. For Viglius, see E. H. Waterbolk and Th. S. H. Bos, eds., Viglzana : bronnen, brieven en
rekeningen betref!ende Viglius van Aytta (Estrikken, So) (Groningen: Frysk Ynstitut en His-
torisch Instituut R. U. Grins, 1975); and Folkert Postma, Viglius van Aytta als humanist en
dzplomaat (1507-1549), (Zutphen: Walburg Pers., ca. 1983).
21. This anecdote is recounted by Hans Jacob in Cod. Vat. Lat. 64I2, fol. lOs. For Alciati,
see Ernst von Moeller, Andreas Alciati (1492-1550): Ein Beitrag zur Entstehungsgeschichte der
modern en Jurisprudenz ( Studien zur Erlauterung des Burgerlichen Rechts, 25) (Breslau: M. &
H. Marcus, 1907); Frederik Willem Gerard Leeman, Alciatus' Emblemata: denkbeelden en
voorbeelden (Groningen: Bouma's Boekhuis, 1984); and Johannes Kohler. Der "Emblematum
liber" von Andreas Alciatus (1492-1550): Eine Untersuchung zur Entstehung, FO/mung antiker
Quellen und padagogischen Wirkung im 16. Jahrhundert, (Beitdige zur historischen Bildungs-
forschung, 3) (Hildesheim: A. Lax, 1986).
22. For Hieronymus Wolf, see Hieronymus Wolf, Der Vater der deutschen Byzantinistik:
das Leben des Hieronymus Wolf von ihm selbst erzahlt, trans. Hans-Georg Beck (Miscellanea
Byzantina Monacensia, 29) (Munich: Institut fur Byzantinistik und neugriechische Philologie
der Universitat, 1984). For Ascham, see Alfred Katterfeld, Roger Ascham: Sein Leben und seine
Werke (Strasbourg: K. J. Trubner, 1879), esp. 140 - 41, and Lawrence Ryan, Roger Ascham.
(Stanford, Calif.: Stanford University Press, 1963). In I55I, Ascham was a houseguest of
Hans Jacob Fugger.
23. For Farnese, see Leon van der Essen, Alexandre Farnese, prince de Parme, gouverneur
general des Pays-Bas (1545-1592), (Bibliotheque du seizieme siecle) (Brussels: Librairie
nationale d'art et d'historie, 1933); Antonio Bezzi, Alessandro Farnese: una vita per un idea Ie
(Collana di storia, arti figurative e architettura, 12) (Parma: L. Battei, 1977); and Alessandro
Pietromarchi, Alessandro Farnese: l'eroe italiano delle Fiandre (Le Storie della Storia, I3)
(Rome: Gangemi, 1998). For Madruzzo, see Antonio Monti, Filippo II e it card. Cristofaro
Madruzzo, gobernatore di Milano (1556- 1557) (Milan: Societil editrice Dante Alighi eri di
Albrighi, Segati & Co., 1924). For Hosius, see Joseph Lortz, Kardinat Stanislaus Hosius:
Beitriige zur Erkenntnis der Personlichkeit und des Wakes (Braunsberg: Herder, I93I). For
Truchsess, see Bernhard Schwarz, Kardinal Otto Truchsess von Waldburg, FiirstbischoJ von
Augsburg; sein Leben und Wirken bis zur Wahl als FiirstbischoJ von Augsburg (1514-1543)
(Geschichtliche Darstellungen und Quellen, 5) (Hildscheim: F. Borgmeyer, 1923).
24. Hans Jacob Fugger, Geheimen Ehrenbuch Mannsstammens und Namens der Eerlichen
und altloblichen Fuggerischen Geschlechts, 1541 - 1545. Copies of this manuscript may be found
in the Nuremberg Germanisches Museum and in the Fugger Museum in Augsburg. Hans
Jacob Fugger, Warhaf!tige Beschreibung Zwaier Inn ainem Der alter Edlesten vralten vnd
hochloblichisten Geschlechten der Christenhait des Habspurgischen vnnd Osterreichischen gebliiets,
sampt derselbigen lobwurdigen herkommen, Geburten, leben, Regiment vnnd Ritterlichen geth -
aten, Von dem anfanng biss auf! die Vnuberwindtlichisten Grossmechtigisten Fursten vnd herren,
herrn Carolum, den Junffien vnd Ferdinandum, der ersten, Romischen Kaiser vnd Konige, auch
recht ordenliche Erwolte vnd gekronte, Obriste haupter der Christenhait, 1555. Maasen doubts
that Fugger was in fact the author of this compendious work, suggesting instead that he may
have served merely as editor and that the writing itself was by the Augsburg cobbler and his-
torian Clemens Jager. See Maasen, Hans Jakob Fugger, 59-73, esp. 67-68. Copies of this work
reside in the Staatsbibliotheken in Munich and Vienna. Bibliographic refs. to Fugger and
Habsburg histories.
25. Quiccheberg, Inscriptiones vel tituli Theatri amplissimi complectentis rerum vniuersitatis
singulas materias et imagines eximias, ut idem reete quoque dici possit: Promptuarium artifi-
ciosarum miraculosarumque rerum, ac omnis rari thesauri et pretiosre supellectilis, structurre atque

Mark A. Meadow
pictura:, qua: hie simul in theatro conuqiri consuluntur, ut eorum frequenti inspectione tracta-
tionecque, singularis aliqua rerum cognitio et prudentia admiranda, cito, facilie ac tuto comparari
possit. (Munich: Adam Berg, 1565).
26. See Maasen, Hans Jakob Fugger, 74-90, on Hans Jacob Fugger's role as a patron of
scholarship and the books dedicated to him as a result.
27. For Hans Jacob Fugger at the court of Fer din and, see Maasen, Hans Jakob Fugger, 8-10.
28. For Hans Jacob Fugger's role in Augsburg politics, see Maasen, Hans Jakob Fugger, 12-30.
29· For Hans Jacob Fugger at the court in Munich, see Maasen, Hans Jakob Fugger, 45-58.
30. For the Byzantine holdings, see B. Mondrain, "Copistes et collectionneurs de manu-
scrts grecs au milieu du XVIe Siecle," Byzantinische ZeitschriJi 84 (199 1/92): 354-390.
31. Wolf began his employment as librarian to the Fuggers in 1551. The great librarian
Conrad Gessner had been offered the position in 1545, but the arrangements never came to
fruition. Wolf remained in this post until 1557. See Lehmann, Eine Ceschichte del' alten Fug-
gerbibiliotheken, Vol. I: 50-57.
32. Jacopo Strada began working for Hans Jacob Fugger while living in Nuremberg in 1544.
At Hans Jacob's request, and with his financial support, Strada prepared a thirty-folio volume set
of numismatic drawings to serve as a standard reference for Fugger's collections. He continued
to serve Fugger as an intermediary in the purchase of antiquities and coins in Rome. Strada pro-
duced architectural drawings for Albrecht V's Antiquarium in Munich, as well as designing his
own house in Vienna. He entered the service of the Habsburgs in 1558, and continued in their
employ until 1579. For Strada's work in the German-speaking world, see Renate von Busch, Stu-
dien zu deutschen Antikensammlungen des ]6. Jahrhunderts (Diss. Tiibingen: 1973). For other
aspects of Strada's career, see Dirk Jansen, "Jacopo Strada (1515-1588): Antiquario della Sacra
Cesarea Maesta," Letds Kunsthistorisch Jaarboek, I (Leiden: 1982),57-69; E. Fudkov<i, "Einige
Erwagungen zum Werk des Jacopo und Ottavio Strada," Leids Kunsthistorisch Jaarboek, I (Lei-
den: 1982),339 - 353; Dirk Jansen, "Jacopo Strada et Ie commerce d'art," Revue de l'art 77 (1987),
11-21; idem, "Gli strumenti del mecenatismo: Jacopo Strada alia corte de Massimiliano II," in
"Familia" del Principe e famiglia aristocratica (Biblioteca del Cinquecento, 41) ed. Cesare Moz-
zarelli (Rome: Bulzoni, ca. 1988), 681-715; idem, "Example and examples: The potential influ-
ence of Jacopo Strada on the development of Rudolphine art," in Prag urn 1600: Beitrage zur
Kunst und Kultur am Hofe Rudolfi II, (FrerenlEmsland: Luca, 1988), pp. 132-146; idem, "Der
Mantuaner Antiquarius Jacopo Strada," in Fiirstenhofe der Renaissance: Ciulio Romano und die
klassische Tradition (Vienna: Kunsthistorisches Museum, 1989),308-323; idem, "Jacopo Strada's
Antiquarian Interests: A Survey of his Musaeum and its Purpose," Xenia: Semestrale di Antichitd,
21 (1991),59-76; and idem, "The Instruments of Patronage: Jacopo Strada at the Court of Max-
imilian II: A Case Study," in Kaiser Maximilian II: Kultur und Politik im ]6. Jahrhundert (Wiener
Betrage zur Geschichte Neuzeit, 19), ed. Friedrich Edelmayer and Alfred Kohler (Vienna: Ver-
lag fiir Geschichte und Politik; Munich: Oldenbourg, 1992), 182-202.
33. Quiccheberg came into the employ of the Fuggers in 1555, first as physician to Anton
Fugger "dero leib zu warten und zu artzneyen," but was working as librarian at least as early
as 1559. For Quiccheberg, see esp. Harriet Roth, Der Anfang der Museumslehre in Deutschland:
das Traktat "Inscriptiones vel tituli theatri amplissimi" von Samuel Quiccheberg; lateinisch-
deutsch, Berlin, 2000; and Patrice Falguieres, "Fondation du Theatre ou Methode de l'exposi-
tion universelle: les Inscriptions de Samuel Quicchelberg (1565)," Les Cahiers du Musee
National d'Art Moderne 40 (1992), 91-109. For Quiccheberg at Munich, see Otto Hartig, "Der
Arzt Samuel Quicchelberg, der erste Museologe Deutschlands, am Hofe Albrechts V. in
Miinchen," Bayerland 44 (1933), 630- 633.
34. For Raimund Fugger's collections of antiquities, see Norbert Lieb, Die Fugger und die
Kunst, Vol. 2 of Studien der Fuggergeschichte (Munich: Verlag Schnell und Steiner,
1952-58),46-5 1,349-351.

Merchants and Marvels 199


35. Raimund Fugger's astonishing collection of lutes and harpsichords was cataloged in
1566, at the time they were acquired en masse by Albrecht V. See Stockbauer, Die Kunstbe-
strebungen am Bayerischen Hofe, 8I -84-
36. The best work on the art patronage of the Fuggers remains Norbert Lieb, Die Fugger
und die Kunst, see esp. Vol. 2,303-305 for Titian.
37. ]acopo Strada, Antiquorum numismatum. Later an excerpted version was published as
Epitome Thesavri antiqvitatvm, hoc est, impp. Rom. Orientalivm et Occidentalivm Iconvm ex
antiquis Numimatibus quam fidelissimie deliniatarum (Lyon: 1553). For Albrecht V's numis-
matical collections, see Stockbauer, Die Kunstbestrebungen am Bayerischen Hofe, 70-72.
38. For the Fugger collections and trade in jewels, see Lieb, Die Fugger und die Kunst, Vol.
2, I33-I38, and esp. 137-138 concerning the Burgundian gems and their eventual resale.
39. For the purchase of coral, see Lieb, Die Fugger und die Kunst, Vol. 2, 140.
40. See Lehmann, Eine Geschichte der alten Fuggerbibiliotheken, Vol. 1,57.
41. See Seelig, "The Munich Kunstkammer," 76. Albrecht V was preceded in this by the
Habsburgs Maximilian II and Ferdinand II, who declared certain artifacts inalienable prop-
erty only a year earlier in 1564. These included an agate bowl and a "unicorn" horn, another
narwhal tusk. See Elisabeth Scheicher, "The Collection of Archduke Ferdinand II at Schloss
Ambras," in The Origins of Museums, 29-38, here 30; and Rudolf Distelberger, "The Habs-
burg Collections in Vienna during the Seventeenth Century," in The Origins of Museums,
39-4 6, here 43.
42. For the intellectual importance of antique collections, see esp. Horst Bredekamp,
"Antikensehnsucht und Maschinenglauben," in Forschungen zur Villa Albani: Antike Kunst
und die Epoche der Aufkliirung (Frankfurter Forschungen zur Kunst, IO), ed. Herbert Beck
and Peter Bol (Berlin: Gebr. Mann, 1982). For the conceptual importance of ancient coins to
the intellectual premises of the curiosity cabinet, see Arnold, Cabinets for the Curious, 42-87.
43. Quiccheberg, Inscriptiones vel tituli, Civ r-Di r.
44. Ibid ., Di v-Dii r.

200 Mark A. Meadow


Practical Alchemy and
Commercial Exchange in the
Holy Roman Empire
TARA E. NUMMEDAL

B y the mid-sixteenth century, alchemy was of widespread interest in


the Holy Roman Empire. No longer the preserve of learned natural
philosophers and initiates alone, the alchemical arts engaged princes,
pastors, and craftspeople, both male and female. This diverse group of
enthusiasts devoured alchemical literature as publishers ushered ancient
and modern authors into print; they also traded techniques with fellow stu-
dents of nature and bought recipes from peddlers of alchemical secrets. Not
surprisingly, given alchemy's wide purview of the theoretical and the prac-
tical as well as the mystical and the material, alchemical practitioners dif-
fered about how precisely to define their art, how to master it, and what to
do with it. By the end of the century, practitioners increasingly disagreed:
what exactly was alchemy, and, as it gained publicity and the support of
political leaders, what were its goals to be?
In this paper, I argue that in the late sixteenth century, alchemists offered
at least two different answers to these questions. The first view, rather tra-
ditional, saw alchemy as a natural philosophy which sought to understand
God through his greatest revelation: nature. From this perspective, even the
practice of alchemy was pious. In using his art to heal natural bodies
(whether human or metallic), the philosophical alchemist sought nothing
less than the regeneration of the world by cleansing it of impurities result-
ing from the Fall of Adam and Eve. A second strain in early modern
alchemy took a much more pragmatic perspective, emphasizing instead
alchemy's utility and productivity in the world of things. This practical
alchemy was markedly commercial in the sense that it was both accessible
through and supported by a growing market in alchemical goods and ser-
vices. Though these two threads in early modern alchemy were not identified
exclusively with particular individuals or groups (indeed, a single individual
could exhibit both tendencies), nonetheless, they were increasingly in conflict

201
in this period. Both had potential to triumph and determine whether
alchemy was ultimately to be about understanding God and learning or
profit and the production of things.
Over the past few decades, historians have done much excellent work on
the philosophical and spiritual aspects of alchemy. They have shown that
early modern European scholars and political elites viewed alchemy not only
as possible, but often as central to their intellectual, religious, and political
activities.' In focusing on alchemy as an idea, philosophy, or metaphor, and
explaining why it made sense to early modern Europeans, these historians
have been pivotal in reevaluating alchemy's marginality and demonstrating
its importance and legitimacy both before and during the formulation of the
new science. It has been much more difficult, however, to appreciate
alchemical practitioners who claimed to be able to do alchemy successfully,
to actually transmute metals or create the philosophers' stone, largely
because it seems so obvious to us in the twenty-first century that it is not pos-
sible to create gold out of iron. 2 And yet alchemy as a practice and the
alchemical production of things was just as important to early modern
patrons and practitioners as it was as an idea. As alchemists' contracts, pro-
posals, and laboratory reports reveal, a vital community of practitioners was
at work in the Holy Roman Empire on a range of alchemical projects. If we
wish to understand the relationship between alchemy and commerce, we
must first understand the practice of alchemy and its relationship to emerg-
ing markets and the world of goods.

ALCHEMY AS COMMODITY

Interested patrons and practitioners could find a great deal of alchemical


knowledge for sale in the sixteenth century. Novices who wished to pursue
their interest in alchemy would have found a variety of literature in the book
stalls, where ancient and medieval Islamic and Christian authors joined
more modern authorities such as Paracelsus (ca. 1493-1541).3 Practical
books in the vernacular sold particularly well and did much to expand the
audience for alchemy. William Eamon has noted how printers often
amended these "how-to books" to make them more accessible for audiences
from the middling classes, inserting indexes, prefaces, and translations of
difficult or technical terms. 4 The 1570 version of the Kunstbiichlein (or skills
booklet) titled Alchimia, for example, began with an "Explanation of Some
Latin Words" that translated sol to gold, corpus to "any metal or material,"
and so on. 5 With these sorts of additions, Eamon notes, "philosophical tradi-
tions such as alchemy were given a new relevance by being placed within the
reach of general readers."6 Furthermore, students of alchemy who did not

202 Tara E. Nummedal


wish to (or could not) buy an entire book might come across a fellow enthu-
siast willing to share a recipe for a small fee or a skilled practitioner willing
to instruct them in a particular process in exchange for pay. Alchemy had
become much more accessible and widely dispersed than it had been to its
medieval devotees.
The career of the Wolfenbiittel alchemist Philipp Sommering (ca. 1535-
75) provides a rare glimpse of a practical alchemist negotiating this market
of alchemical expertise. The son of a pastor in Thombach, Sommering
attended school before taking up a series of positions in the Lutheran
Church. When war and a dispute disrupted his position as a pastor in 1555,
Sommering set off to wander, by his own account, two hundred miles
throughout the Holy Roman Empire. On his journeys, Sommering met two
men from whom he obtained his first alchemical book. Sommering pro-
cured from a fellow pastor a second book, in which he reportedly read about
certain distilling techniques. With only these two books, Sommering later
reported, he began to try his hand at alchemy. He furthered his studies by
paying a woodcarver in Erfurt 5 thaler to teach him distillation and subli-
mation, then copied down the varieties of plants and bought I I thaler worth
of herbs from an apothecary.7
Still in Erfurt, Sommering discussed the art of alchemy with someone he
identified as a "Philosopher" and bought two more books: the Book of Isaac
("in which there are many good things") and the Hexameron of Bernardus. 8
Together these two books cost Sommering the considerable sum of 400
thaler, an expense he shared with Abel Scherdinger, confessor to the Count
of Hennenberg. Scherdinger and Sommering clearly viewed these books as
an investment, for they made an agreement to pursue alchemy together and
to split their profits equally. Sommering had the further good fortune of
learning "a highly secret art, namely the regulation of the fire" from an
alchemist named Martin Gurlach. This cost Sommering nothing more than
perhaps the price of a few beers, inasmuch as he reportedly got the informa-
tion from Gurlach "while drinking." Having purchased books, recipes, and
skills from a variety of people, Sommering was thus ready to work as an
alchemist. In 1566 he and Scherdinger signed a contract to produce the
philosophers' stone for Duke Johann Friedrich of Sachsen-Gotha. 9
Several aspects of this story are remarkable. The impressive array of peo-
ple from whom Sommering bought his knowledge-including pastors,
another alchemist, an apothecary, and a philosopher-is an indication of
how widely dispersed alchemical knowledge had become. Similarly, that
knowledge came in a wide variety of forms. Sommering not only bought
books, but gleaned alchemical secrets during official lessons (from the
woodcarver) and casual conversation (from the alchemist). Most strikingly,
however, Sommering's tale illustrates the extent to which alchemy was for

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 203
sale by the mid-sixteenth century. Alchemical knowledge itself had become
a commodity.
Alchemy's entrance into the marketplace created new forms of alchemi-
cal knowledge and new standards for measuring it. Finding it difficult to
place lifetimes of learning and vast philosophical systems in compact (and
marketable) books, sellers often repackaged this knowledge in the form of
recipes and processes, eliminating (or at least deemphasizing) larger theoret-
ical frameworks. 1O At the same time, the market forced buyers and sellers to
place a monetary value on alchemical knowledge. As a result, the kind of
alchemy which enthusiasts bought and sold at the end of the sixteenth cen-
tury increasingly came to emphasize qualities that promised immediate
returns and highlighted profit and utility.
Nowhere was this new emphasis on alchemy's productive potential clearer
than in the manner its practitioners promoted themselves to likely patrons.
One strategy alchemists used to peddle their wares was to highlight the
potential profits their processes could offer. When the metallurgist, mint offi-
cial, and technical author Lazar Ercker (ca. 1530-94) wrote to Duke Julius of
Braunschweig-Wolfenbiittel (1528-89), he underscored the money, quite lit-
erally, that he could help generate. I I In this 1585 letter, Ercker described a
process by which "using a powder, I can bring Rheinisch, or other low qual-
ity gold, in a few days to proper Ducat-quality gold," which was worth twice
as much. Ercker claimed that he could transmute 100 marks (about 233.85g)
of Rheinisch Goldgulden a week, with an extra cost of 10 thaler "for the coals
and all the Instruments." Promoting the efficiency of his process, Ercker
noted that he could make use of the by-products as well. "The silver which
the Rheinisch Goldgulden have in them will be melted out of the powder
again," he boasted, "and the gold which the powder has also absorbed, of
which there is little, will be separated out and used as is useful."1 2
Ercker emphasized the profits his process would generate. "I am of the
humble opinion," he wrote, "that for every hundred Marks of Rheinisch
Goldgulden, given the initial costs, there should be a surplus and financial
profit of at least seventy or eighty Thaler."u As support for his claims,
Ercker cited his own results using the technique to mint coins for a mer-
chant from Nuremberg. The merchant profited handsomely (according to
Ercker) producing as much as 2,000 thaler in a year. In Duke Julius's case,
Ercker pledged, the profits promised to be even greater. "In my opinion," he
wrote to Julius, "it would be much more lucrative and useful to Your
Princely Grace because Your Princely Grace can invest [verlegen] much
more than a merchant-which, however, can not happen at all without this
invented art of mine."14
As Ercker's pledge suggests, practical alchemists often marketed their
skills not only by underscoring productivity, but also efficiency. Their claims

Tara E. Nummedal
were usually very specific, outlining the exact ingredients, in precise quanti-
ties, and the resulting amount of precious metal. Petr Hlavsa of Liboslav, the
manager of Bohemian magnate Vilem-Rozmberk's (1535-92) Prague
alchemical laboratories, described one alchemist's technique in just such
detail in 1574. 15 For this process, which according to Hlavsa was "truly in
accordance with the alchemical art," the alchemist Cristoff von Hirschen-
berg started with 8 Loth gold and 8 Loth silver.16 Using "the accompanying
powders and materials," Hirschenberg increased the proportion of gold,
producing 5 Loth silver and I I Loth gold "which should pass any tests and
should remain fixed."I? The alchemist Michael Polhaimer (1566/67-98)
signed a contract in 1595 with Count Wolfgang II von Hohenlohe (1546-
1610) to perform a slightly different process: an "augmentation" that would
transmute 2 pounds (or 64 Loth) of mercury into 10 Loth of "fine silver."18
Such processes were typical among practical alchemists in their specificity
and accuracy; they also reflect an awareness of the patrons' desire to know
exactly how much money they would have to invest in this type of work and
what kind of rewards it could yield.
When practical alchemists did win the support of patrons, they typically
set down the terms of employ in a very businesslike contract. These con-
tracts transferred the specific details of the proposals into a legally binding
document, stipulating in detail the type of processes the alchemist was to
carry out and the deadline for completion, as well as the patron's duties in
terms of payment, facilities, and materials. The contract that Philipp Som-
mering and Abel Scherdinger entered into in 1566 shortly after they
acquired their two prized books was typical. After demonstrating their art
at the court of Duke Johann Friedrich of Sachsen-Gotha, the alchemists
agreed to give the duke 10 percent of the proceeds from their philosophers'
stone in exchange for an advance of 760 thaler, raw materials, and equip-
ment. 19 This kind of arrangement differed from a more general patronage
relationship in its specificity. Whereas many philosophical alchemists
counted princes as their patrons, they never signed contracts of this type.
Instead, they were typically hired as court physicians, expected to perform a
variety of duties associated with their position. The contracts that practical
alchemists signed, on the other hand, were ordinarily limited to the perfor-
mance of a single, specific process. 20

USEFUL ALCHEMY AND THE WORLD OF THINGS

The range of skills that practical alchemists claimed to possess, however,


could encompass anything from medicine to metallurgy. The entire laundry
list appeared in a 1597 text written by Alexander Lauterwald in praise of the

Practical Alchemy and Commercial Exhange in the Holy Roman Empire


alchemical ans. n Lauterwald's interlocutor in the treatise and the embodi-
ment of alchemy, Chimia, proclaims proudly that "no one can do without
me" before outlining the full range of activities-from cooking to the fabri-
cation of precious stones-in which she can be of use. Chimia reserves her
strongest claims of utility, however, for mining and medicine.

To that I must add even more / The minerals must choose me as well /
when they want to separate themselves from others, / from the ore, in which
they languish, / despairing that they are not bright. All of this I do without
danger. / Indeed, I make all the metals right. / [Without me] they can not
come clean, / nor can they please people; / Since they aren't properly
worked, / They crumble under use. / Nor can the ore be used / If it isn't first
cleansed through my breath, / [madel pure and clean / Such things know
my children alone. / They are the goldsmiths and assayers, / The mint mas-
ters and jewelers / I can bring forth gemstones as well/Make glass that can
bend light / Many lovely distillations / Are used for medicine / In which the
great secret is buried / He who achieves this, need not worry,zz

Although today we tend to think of medicine, metallurgy, and the produc-


tion of jewels as separate activities, Lauterwald's poem reminds us that in
the minds of early modern Europeans, they were closely related activities, all
of which could fall within the provenance of the alchemist.
Practical alchemy was, therefore, by no means limited to the production
of noble metals. Alchemists frequently carried out their metallurgical pro-
jects alongside medicinal ones, such as the "theophrastian universal medi-
cine" that alchemist Michael Heinrich Wagenmann vom Hoff contracted
with Duke Friedrich of Wiirttemberg (1557-1608) to make in December
1598.23 Some of these medicines were more obviously connected to precious
metals, such as the olium solis et lunae (oil of sun [gold] and moon [silver])
which Melchior Hornug prepared with "a little of our gold" from his
patron's mine in Reichenstein, Silesia, or the seemingly omnipresent medic-
inal golden liquid, potable gold. 24 The close connection that Paracelsus and
others drew between the new chemical medicine and alchemy, in addition to
the overlap in distilling skills involved in both, ensured that most practical
alchemists might turn their attentions to medicine as easily as to metals. The
proper powder or liquor might just as easily "heal" or purify a human body
as a metallic body.
Practical alchemy could be useful in other ways as well. A female
alchemist named Anna Zieglerin (ca. 1556-75) extended the alchemical con-
cern with generation beyond metals to animal and vegetable material. In an
unpublished booklet written in 1573 for her patron, she described a method
"for when one wants to have cherries, grapes or other good, ripe fruit early

206 Tara E. Nummedal


in winter."25 Most remarkably, Zieglerin shared with Duke Julius of Braun-
schweig-Wolfenbiittel her unique understanding of the homunculus,
imparting a method by which the tincture used for transmuting metals
might also be used to engender children. Zieglerin recommended that
women having difficulty getting pregnant drink the alchemical tincture
daily. When the pregnancy succeeded and the baby was born, the mother
should "let the baby taste no mother's milk and give it nothing to eat or
drink ... [but] three times a day let it have three drops [of the tincture] in its
mouth."26 Here Zieglerin demonstrated yet again just how interconnected
minerals, plants, and animals could be in view of the alchemist. Just as met-
als like gold could be used to cure humans, so too could babies thrive on the
same tincture that brought noble metals out of base.
Despite all of these other activities, the transmutation or multiplication of
metals remained the heart of the practical alchemical enterprise and the skill
for which patrons seemed most to value its practitioners. Proposals and
claims varied widely, but practical alchemists ordinarily had either a tincture
or a "process" for this purpose. Although tinctures also had medicinal uses,
in the context of metallic transmutation or multiplication the term desig-
nated a liquid or powder that could "tinge" or transmute metals.27 Typically
alchemists already possessed the tincture and asked only for the chance to
demonstrate its potential. The Cypriot alchemist Marco Bragadino (ca.
1545-91), for example, arrived in the Veneto in 1589 with his "medicine"
and successfully demonstrated it by transmuting 1 pound of quicksilver
under the critical eyes of two officials from the Venetian Mint and an assem-
blage of local nobility.28 Occasionally practitioners claimed only to have a
sure-fire recipe for the tincture rather than the substance itself, but they
swore that they could produce it given the necessary materials and equip-
ment. 29 A "process" or Kunstwerk (work of artistry or skill) most often
described a recipe or method for multiplying metals, and characteristically
required an initial investment from a patron to perform.
When practical alchemists successfully secured positions at the princely
courts of the Holy Roman Empire, they frequently found themselves work-
ing alongside others in large alchemical laboratories. Such laboratories
sprouted up in Saxony, Bavaria, Bohemia, Braunschweig-Wolfenbiittel, and
Wiirttemberg in the late sixteenth century, mini-alchemical workshops dot-
ting the map of the Holy Roman Empire. The activities in these laboratories
illuminate the extent to which practical alchemy was productive work.
Although the types and extent of alchemical activity in each of these labora-
tories certainly varied, we can get a sense of the kind of work it was from a
1608 inventory of the laboratories that Duke Friedrich I set up in his south-
ern German territory of Wiirttemberg. This inventory, taken upon the
Duke's death, provides a snapshot of the kind of alchemical activity that

Practical Alchemy and Commercial Exhange in the Holy Roman Empire


went on at noble courts in the Holy Roman Empire. Friedrich's court physi-
cians Abraham Schopf and Ulrich Porta, together with the manager of the
alchemical laboratories Chrystoff Wagner, drew up a list of "all the trans-
mutational and medicinal processes (Stiick), in addition to raw and prepared
materials, instruments, ovens, glasses, crucibles, cappels, and other tools
required for work" found in the ducal laboratories. According to the list,
Wagner himself was in the middle of eight different projects when the
inventory was made. lo In addition, the inventory listed five others at work in
the laboratories. Andreas "the chamberboy," for example, had already fin-
ished two vials of a tincture, soon to be demonstrated in a projection, and
was in the midst of a process which was to produce gold from mercury com-
bined with a certain elemental substance. Daniel Keller, another assistant,
was working on both a tingeing process and the multiplication of a tincture,
while Johan GeiJ3ler was also finishing a tincture which he was shortly to
demonstrate. Georgius Butina was earning four gulden weekly working on
an unnamed process for the duke and simultaneously working on a process
of his own involving gold. The last assistant, Adam Wiera, had just finished
a process that produced an entire pound of gold and would double in
another six weeks (thus producing 2 pounds). There were apparently others
at work in the laboratory as well, as the inventory noted that "what the other
workers are carrying out ... can also be inquired about." II
Taken together, these laboratories worked as a sort of alchemical manu-
factory-cum-workshop. Under the direction of Duke Friedrich's manager
Chrystoff Wagner, each of these assistants annually earned from 52 to 208
gulden (plus, in several cases, two new outfits yearly) to work on various
processes. l 2 Interestingly, few of the assistants were working on processes
that they had proposed to the duke; rather, Friedrich handed out processes
he collected elsewhere, assigning them to individual assistants. We learn
from the inventory, for example, that Daniel Keller's task was "a work of
great importance at the command of His Princely Grace." Similarly,
Andreas labored at "a work, which was mentioned to His Princely Grace by
Gerbelium of Strasbourg." The exception was Georgius Butina, who was
lucky enough to have "arranged with His Princely Grace license [to work
on] his own invention of a process." Butina, however, was unusual. For the
most part these assistants were hired hands, paid a yearly salary for produc-
ing alchemical goods on their sovereign's behalf.33
The activities in Duke Friedrich's laboratory demonstrate the extent to
which early modern alchemy was involved in the production of things. This
was productive knowledge, and its practitioners put it to work in creating a
variety of useful items. Indeed, this kind of alchemy seems strikingly mun-
dane. Because we imagine alchemical laboratories to be dark, smoky solitary
rooms and the work there to be highly secretive, it is tempting to say that the

208 Tara E. Nummedal .


kind of alchemical practice in Friedrich's laboratories was actually some-
thing else, metallurgy perhaps, or medicine. The practitioners, however,
believed that they were alchemists and that what they were doing was
alchemy. The market had given them access to the art, and their skills were
appreciated by patrons quite willing to pay for them. Their activities were
just as much a part of alchemy as were the mystical meditations of their
more spiritual colleagues.

PATRONS AND THE USES OF ALCHEMY

Why did princes like Friedrich devote such substantial resources to practical
alchemical projects? Historians such as R. J. W. Evans and Bruce Moran
have argued convincingly that alchemy could offer a solution to the political
and religious problems plaguing central Europe. As Evans noted decades
ago in his exploration of Emperor Rudolf II's well-known occult pursuits,
the alchemical view of nature posited a single divine order that underlay and
connected the natural and the human worlds. As such, the alchemist's work
in the laboratory was also work on the world, and "alchemists sought not
only the regeneration of metals through the [philosophers'] stone, but also
the moral and spiritual rebirth of mankind."34 This idea held particular
promise in a fractured Holy Roman Empire still reeling from the religious
wars following the Reformation. 35 Bruce Moran emphasized the political
side of this same coin. "The occult vision of unity and universality," he
noted, "offered an intellectual balsam for religious and political confusion.
As such, it became a surrogate reality, and it is in this sense that its patron-
age, as much at Hessen-Kassel as at other German courts, became finally a
patronage of despair."36
Certainly alchemy appealed to some princes on these abstract (yet very
real) levels. The increased appeal of alchemy just as the political and reli-
gious structure of the empire seemed to be falling apart is an important con-
nection, and does much to explain the power of alchemical ideas. Princes
also had more practical concerns, however, and alchemy addressed these as
well. We can well understand why it would have appealed to princes con-
cerned about their health, for example. Whether practitioners sought to cre-
ate a panacea in the philosophers' stone or simply the newly fashionable
chemical drugs vaunted by Paracelsus and his followers, alchemy certainly
promised medical marvels. And few Renaissance princes would have turned
down the pearls and gemstones some alchemists offered, let alone Anna
Zieglerin's wintertime fruit. Renaissance princes' delight in such wonders is
well-known and would have disposed them to appreciate alchemy's more
opulent productions.

Practical Alchemy and Commercial Exhange in the Holy Roman Empire


A number of central European princes saw even more potential in prac-
tical alchemy and understood it as a solution to the financial and mining
crises afflicting their territories in the second half of the sixteenth century.
Nearly constant warfare combined with Renaissance building projects and
courtly splendor drained princely coffers over the course of the sixteenth
century. At the same time, the rich central European mines, a fruitful source
of income through the mid-sixteenth century, began to stagnate. 3 ? Whereas
new technologies and financial investments had increased silver production
in central Europe fivefold between 1460 and 1550, this growth leveled off in
the second half of the sixteenth century as the balance shifted definitively
toward imported American silver. Silver imports to Europe increased more
than thirty-fold in sixty years, in fact, surging from 86 metric tons in the
1530s, to 1,118 in the 1570S and 2,707 in the 1590S.38 By 1600 the golden age
of central European mining had clearly come to a close.
Despite this slump, and perhaps in reaction to it, German and Bohemian
princes continued to take an active interest in mining, and several territorial
rulers quite actively pursued projects designed to exploit their territories'
natural resources. Duke Julius of Braunschweig-Wolfenbuttel, for example,
commissioned a report in 1572 on "all kinds of mountains, metals and other
uses which are found in [the mining regions of] the Harz and Rammels-
berg" and hired mining expert (Bergmeister) Hans Fischer to search the
ducal territory for natural resources. 39 In addition to these kinds of
exploratory projects, Julius invested heavily in his various mines, spending
roughly a third of his budget on them in 1579-80, a sum justified by the fact
that his mining enterprises were his largest source of income. 4o Other princes
took similar measures to develop their mining industries. 41
For these princes, practical alchemy was intimately related to the pursuit
of profits through mining. Alchemical expertise, particularly that of Schei-
dekunst, or smelting, could be extremely useful in mines where difficulties in
extracting precious metals from ore had caused a decline in productivity.
Recall Chimia's claim in Lauterwald's text: "The minerals must choose me
as well! When they want to separate themselves from others." This seems to
have been the case with the mine that Bohemian magnate Vilem Rozmberk
bought in Reichenstein (Lower Silesia), where the gold ore was particularly
thinly dispersed and difficult to smelt. The alchemical laboratory Vilem
established there immediately after he purchased the mine may well have
been intended to solve this problem. 42 Two alchemists proposed similar
processes to Duke Julius of Braunschweig-Wolfenbuttel in the 1570S. Caspar
Uden offered a process "by which copper and silver may be separated," and
Theophil T6pfer proposed a somewhat vague process for separating metals
"resulting from an alchemical technique."41 Although Uden and T6pfer
were not successful in their proposals, Duke Julius clearly valued alchemy's

2IO Tara E. Nummedal


contribution to his mining enterprises and credited it with their renewal.
Responding favorably to another proposal in 1576, Julius commented, "Like
our beloved Lord and Father, we have been so involved with alchemy that
we have paid dearly with thousands of thaler. Nevertheless, it has also taken
us so far that for one thing, we have improved our mines during our reign,
such that we now enjoy from our various mountains 480,000 gulden coins
more yearly."44
Not only could alchemists assist princes whose ore was difficult to smelt,
but they also promised either to multiply existing precious metals or to turn
metals oflesser quality into gold or silver. The alchemist Georg Honauer (d.
1597) held out this possibility to Duke Friedrich I ofWiirttemberg when he
claimed to possess a process with which two men could produce one zentner
(roo pounds) of gold weekly from the iron in Friedrich's iron-rich territory
of Mompelgard. ("One could also organize it like a large mine," Honauer
added, "with a thousand men.")45 Honauer arrived just as Duke Friedrich
had demonstrated his commitment to developing his mining industries by
announcing a reward for the location of new ore deposits. Friedrich did not
hesitate to hire Honauer; he brought him to Stuttgart in 1596 under ducal
protection, converted the old garden house into an alchemical laboratory,
and provided Honauer with thirteen assistants. After the alchemist proved
his skill in several small trials, Friedrich got down to business: he imported
25 zentner (2,500 pounds) of iron from Mompelgard and charged Honauer
to get to work. 46
Duke Friedrich I and Georg Honauer may have hoped for unusually
spectacular results, but their basic understanding of alchemy as an extremely
productive and versatile art was common among central European patrons
and practitioners in the decades before the Thirty Years' War. Above all,
these practitioners and patrons viewed practical alchemy as a means to gen-
erate profits, whether through sales of books, recipes, processes, or the appli-
cation of those processes to large-scale mining enterprises. This was
primarily a utilitarian use of alchemy, aimed ultimately less at the produc-
tion of broad hypotheses about the natural order than at understanding how
to manipulate nature in order to make it more prolific.
At the end of the seventeenth century, Johann Joachim Becher (1635-82)
would develop a much more sophisticated formulation of the relationship
between alchemy and commerce that linked them metaphorically through
the production and consumption common to them both. By equating the
two, Becher hoped to translate his patrons' interest in alchemy into support
for commercial projects in the German lands. 47 At the end of the sixteenth
century, however, neither the projects nor their promoters articulated such a
comprehensive view. Practical alchemists around 1600 focused their propos-
als on specific processes and their immediate yields rather than on large-

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 2II
scale economic projects. For their part, princely alchemical schemes did not
situate alchemy within a broader reconceptualization of their economies,
nor did they engage yet in the global economies we associate with seven-
teenth- and eighteenth-century commerce.
Instead, these alchemical projects sought to employ new means in order
to maintain traditional ways of generating money (such as mining). The
goal- to increase the production of precious metals in a territory - was still
fairly traditional, even if the means were somewhat more innovative. In this
context, Ercker's comment that Duke Julius could produce "much more
than a merchant" with his alchemical process may be even more revealing.
Unlike Becher's equation of commerce and alchemy, which sought to
engage the emperor in the commercial activities of the merchant, the practi-
cal alchemy that flourished a century earlier may have been a way to do just
the opposite: find a way to improve on traditional methods of making
money precisely without involving the empire's princes in the suspect world
of the merchant. As such, the princely amalgam of alchemy and mining was
a curious blend of tradition and innovation, indicative of the halting emer-
gence of early modern commerce.

CRITIQUES OF THE NEW ALCHEMICAL COMMERCIALISM

Alchemists' participation in the world of commerce did not go unnoticed.


Critics of alchemy as a whole expressed doubt about whether alchemy actu-
ally could achieve the creation of wealth it promised. At the same time, those
who believed in alchemy but pursued it with more philosophical or spiritual
priorities in mind attacked profit-seeking practitioners in print. Drawing on
older conflicts about whether alchemists should use the art solely to make
gold and silver, these critics raised a new objection about the buying and sell-
ing of alchemical secrets. In fact, they denied that alchemists with commer-
cial tendencies were "true" alchemists at all and dismissed them as impostors
and frauds. As the sixteenth century came to a close, the alchemical commu-
nity seemed divided between those who believed that alchemy's objective
should be the production of profit, and those who pursued alchemy as a spir-
itual act with the potential for the regeneration of the world.
The Protestant clergyman and rector Johannes Clajus (1532-92) was
among those who were highly skeptical of alchemy altogether. In his satiri-
cal treatise, Altkumistica, Das ist die ware Goldkunst ... aus Mist gut Gold zu
machen (Old-cow-manure, or, the true golden art . .. of making good gold out of
manure, 1586), Clajus expressed traditional doubts about the use of alchemy
to create wealth. He used a clever play on words to contrast the practice of
alchemy (Alchemisterey) with the traditional agricultural method of making

212 Tara E. Nummedal


a living by fertilizing fields with cow manure (Altkuhmisterey, or "old-cow-
manure-istry").49 Clajus listed an abundance of products that could ulti-
mately result from such a well-fertilized field: eggs, meat, milk, wool, pelts,
leather, hemp oil, and flax, all of which had commercial value. In this way,
he argued, the practitioners of traditional farming, or Altkuhmisterey, could
turn manure into gold.49
Clajus's treatise juxtaposed this peasant Altkuhmist with the alchemist,
setting up an opposition between the traditional agricultural livelihood and
what he clearly saw as a new method: manufacturing it alchemically. Clajus
rejected the possibility of alchemically manufacturing gold and feared that
with so many fixated on an impossible alchemical dream, society would ruin
itself. "Because just now all over this land," he wrote,

Alchemy is growing rampant / And is wreaking havoc more and more /


Many apply themselves diligently to making gold / But end up only falsi-
fying metal/Scattering false coins all over the place. / The fact that many
are seduced / as one can easily establish with examples / makes a mockery
of alchemy / which is nothing but fraud. 50

Cia jus felt that alchemists were most dangerous to themselves. They wasted
their money pursuing a hopeless fantasy, perhaps, like S6mmering, spending
a fortune on books and materials. (Practitioners of Altkuhmisterey, on the
other hand, would find that their fields always provided plenty.) Clajus also
noted that the unlucky could face pitfalls even worse than poverty. "Many
lose eyes and hands, many are beheaded, many burned," Clajus warned,
hinting at several prominent alchemists who went to the gallows for failing
to produce gold for their patrons. 51 Given alchemy's capacity to both impov-
erish and incriminate its practitioners, Clajus concluded, alchemists ought to
stick to more traditional (and certain) ways of making a living, such as farm-
ing. He ended with an admonition: "This is why I recommend Altkuhmis-
terey . .. / With God it is certain and secure / It bears gold out of manure / it
is to be tried. "52
Proponents of the alchemical arts disagreed with Clajus's denunciation,
of course, and wrote treatises praising alchemy's virtues. Many of alchemy's
advocates found themselves walking a fine line, however, with regard to
criticisms like Clajus's. As much as they rejected arguments about alchemy's
futility, a number of alchemists found themselves sympathetic to concerns
about alchemists' participation in the marketplace. Rejecting both the use of
alchemy solely to create gold and silver and the practice of selling alchemical
knowledge for money, critical alchemists increasingly argued that by defini-
tion, practitioners engaged in alchemy's commercial dimension were not
"true" alchemists at all.

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 21 3


The notion that alchemy was about much more than gold and silver had
a long history; for centuries some practitioners of the art had struggled to
distance themselves from what they saw as the corruption of alchemy. In the
early modern period, Paracelsus insistently reminded his readers that he
used the term "alchemy" to mean a technique, usually medicinal, not the
preparation of gold or silver. 53 The Paracelsian Alexander Lauterwald
expressed these sentiments as well in two treatises which responded directly
to Clajus: Widerlegung del' Altkuhmisterey (Refutation ofA ltkuh misterey , 1597)
and Colloquium Philosophicum (Philosophical Colloquium, 1597)' In the Col-
loquium Philosophicum, a dialogue between the three sisters Chimia, Sapien-
tia, and Natura and a young novice in the alchemical arts, Lauterwald's
interlocutor Chimia warned the boy to beware:

If you hear someone openly say / ... he wants to make silver and gold /
you mustn't give him any money, / for what does such a man need? / He
should earn his own living. / You mustn't seek gold; / people will easily
judge you. / Those lads [who do] are the bad seeds / whom I named
before. / Many honorable men have been cheated by them / parted from
their things / finally separated from their goods and possessions; / Only
then have they realized / that such lads deal in tricks. / And so seek the
truth / wherein you will find me / you find me also in my dear children /
they are the true philosophers. 54

Lauterwald's Chimia challenged the novice with a higher calling instead.


She explained that the alchemist's true purpose was to use the philosophers'
stone to heal all bodies, "human, animal and metallic," of the worldly cor-
ruption that followed the Fall of Adam and Eve. In this sense, alchemy was
spiritual work, aimed at the regeneration and ennobling of a corrupt and
fallen world. 55
Although the vociferousness with which spiritual and philosophical
alchemists tried to distance themselves from gold-making reached a
crescendo around 1600, their objections were hardly unique to the six-
teenth century. 56 Beginning in the I590s, however, critics began to react to
the burgeoning market in alchemical goods and princes' increasing inter-
est in alchemy's commercial application. In response, critical tracts began
to focus on a new issue: the sale of alchemical recipes and processes for
profit. Lauterwald's treatise reflects this trend as well. He denounced mar-
ket-oriented alchemists (or "process-sellers," as another observer called
them 57) as frauds who seduced others with alchemy's promise of riches
and eternal life only to trick them out of their money. Again speaking
through Chi mia, Lauterwald issued a stern warning to stay away from
such impostors.

214 Tara E. Nummedal


If such a visitor comes to you / Of whom you have heard before and / Who
claims that he can make gold / Ask him what kind of pay he wants. / He
will come finely dressed, / Finished off with a golden visage. / He will
require three hundred ducats / For this he will counsel / A tincture that
spews a thousand. / Stay away from such a wicked type ... / This is what
I want to say to you / You mustn't support processes. 58

Lauterwald simply assumed that anyone who sold alchemy's secrets was a
fraud because the "true" alchemist would never do such a thing. Chimia's
words reflect a deep suspicion that most practitioners simply wanted to cash
in on alchemy's appeal to wealthy princes.
The Leipzig-born physician and mystical alchemist Heinrich Khunrath
(1560- 1605), best known for the engraving of the alchemist in his Labora-
tory-Oratory that appeared in his Amphitheatrum Sapientiae, took a slightly
milder approach.59 In a nineteen-page "Heartfelt warning and admonition
by a faithful devotee of the truth to all true devotees of the natural transmu-
tory alchemy, which one need keep an eye on because of the villainous grip
of the fraudulent malicious chymists," Khunrath merely issued a caveat
emptor. 60 "If a gold beetle flies up to you and says that he can make silver and
gold and wants to teach you how," Khunrath warned, "do not believe him
quickly and easily; because it is not as mean an art as many let themselves
dream it to be."61 Khunrath certainly did not spare harsh words .for "the
gold beetle guild of villans and ill-intentioned and fraudulent alchemists,"
but he did put the burden not to be tricked on the buyers of alchemical
secrets. 62 Khunrath's remedy was to expose the sleights of hand and tricks he
believed the impostors used to dupe potential backers. A well-informed
buyer, he evidently felt, would make sounder decisions in the alchemical
marketplace and stay away from common alchemists.
The physician and occult philosopher Michael Maier (1569-1622)
offered a much deeper and more nuanced critique of commercial alchemy.
Like Lauterwald and Khunrath, Maier believed that alchemy was funda-
mentally a spiritual art because it dealt with God's greatest secrets, and he
disapproved of those with lesser interests.63 In his 1616 Examen fucorum
pseudo-chymicorum (Swarm of drones, or a critical examination of the
pseudo-chymists),64 Maier publicized his vitriolic attack on those he consid-
ered "pseudo-chymists." His "four marks of the false alchemists" wove
together moral, intellectual, and commercial arguments into a damning
denunciation of practical alchemists, concluding that "such men are very
harmful both for the state and for Chymica."65
The second of Maier's four marks of the impostor explicitly took up the
issue of commercial exchange, imparting his disapproval of practical
alchemists who traded in the marketplace. Like Lauterwald, Maier registered

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 21 5


this censure by defining those who sold alchemical knowledge as false
alchemists. "It is an unmistakable sign of the pseudo-chymicus that he wants
to sell gold for gold, something uncertain as a fact and something priceless
for very little," Maier reasoned. In particular, he questioned the very idea of
alchemy as a commodity by focusing on the issue of prices:

It goes against all reason that someone who really had really mastered this
great art, tested over and over again in experiments, would want to sell
this knowledge to another for a piece of bread or a bit of gold. If he really
doesn't possess it, then it is as if he had sold wind and empty words for
money. If the latter is the case, then the scoundrel receives too much
money for the wind, and the buyer is cheated. If the former is the case,
then the seller is cheated."66

If a practitioner truly possessed the secrets of alchemy, in other words, why


would he or she sell it for a bit of gold, for surely it would be worth much
more? If, on the other hand, the supposed alchemist sold only empty
promises, then a bit of gold was far too high a priceY In Maier's view,
alchemy was either priceless or worthless.
Beneath Maier's discussion of the logic of selling alchemical secrets, one
can detect a broader agenda. The rest of his treatise fired moral, epistemo-
logical, and philosophical salvoes at the folly of those who thought they
could become alchemists with a little training, almost as easily as one could
become a goldsmith or pharmacist. Maier viewed alchemy as a sacred art, a
lifelong project to understand God's mysteries, which required learning,
piety, and years of hard work. He was troubled by the popularization of
alchemy and what he saw as its dilution to the point of not being alchemy at
all. In the context of commerce, however, what is striking about the Examen
fucorum pseudo-chymicorum is that Maier chose to articulate this point in the
language of profit and commercial exchange. As much as he felt that the
market for alchemical goods was responsible for the proliferation of a type
of alchemy of whose goals and practices he despised because it created
shoddy practitioners and the patrons to support them, he also knew that that
market was both vibrant and undeniable. Unable to argue it out of existence,
he chose instead to attack it on its own terms, exposing what he saw as the
flaws of a system that viewed alchemy in terms of profits and prices.

THE VALUE(S) OF ALCHEMY

The critiques of alchemists like Maier, Lauterwald, and Khunrath reveal a


fundamental divide in the community of alchemical practitioners about the

2I6 Tara E. Nummedal


value of their art. Those who purchased and peddled alchemical knowledge
in the decades around 1600 operated in a world in which alchemical value was
defined in terms of utility and profit. Practitioners like Sommering purchased
books and recipes, hoping in turn to sign contracts with patrons and sell
acquired knowledge for thousands of thaler. Patrons like Duke Julius of
Braunschweig-Wolfenbiittel, in turn, purchased this kind of alchemical
knowledge, investing their fortunes in alchemy in order to multiply the nat-
ural resources in their territories. Such practitioners and patrons may have
appreciated more intellectual or spiritual aspects of alchemy as well, of course,
but these considerations rarely entered into their contracts, laboratories, or
alchemical work. These were evaluated in terms of how useful and productive
buyers perceived the alchemy to be. Critics like Michael Maier, on the other
hand, operated in a different value system. They too circulated in courtly cir-
cles and depended on princely patronage, but as alchemists they located their
own worth elsewhere, in their status as pious, learned men with a deep under-
standing of God's mysteries. For them, recipes or processes isolated from that
larger learning were both impossible and, ultimately, worthless.
The irony was that, as much as philosophical and spiritual alchemists
wished to remove alchemy from the marketplace, they ultimately could not.
Even if they did not directly compete with their more practically minded
colleagues for contracts, alchemists like Michael Maier did perceive them as
competition for the right to define what the "true" alchemy was. For even if
philosophical alchemists scorned practical alchemists as frauds and impos-
tors, princes did not - and noble support gave legitimacy to practical
alchemy. As the sixteenth century came to a close, in fact, practical alchemy
seemed to be gaining more and more princely support as princes placed their
fortunes in the hands of its practitioners. The sudden appearance of treatises
like Maier's Examen fucorum pseudo-chymicorum testifies to the vitality he and
others ascribed to this new breed of alchemy, and the extent to which they
feared its triumph. In the end, of course, they were right: enthusiasm for
purely transmutational alchemy would eventually dissipate and come to be
mocked as it is today. This would take centuries, however; even in 1796 a
contributor to a German newspaper would declare that "thousands of hands
and minds" were still at work on alchemy in the Holy Roman Empire. 68 In
the sixteenth and even early seventeenth centuries, both strains of alchemy-
one commercial, one philosophical-might have claimed the right to define
whether alchemy was to be a pious natural philosophy or a useful practice
immersed in the expanding world of early modern commerce.

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 21 7


I wish to thank Janice Neri, Daniel Stolzenberg, Seth Rockman, and the participants of the
1999 Clark Library workshop on "Commerce and the Representation of Nature in Early
Modern Europe" for their insightful comments on early versions of this paper.

I. See, for example, B. J. T. Dobbs, The Foundations of Newtons Alchemy or, "the Hunting

of the Green Lyon" (Cambridge: Cambridge University Press, 1975); R. J. W. Evans, Rudolf II
and His World: A Study in Intellectual History (Oxford: Oxford University Press, 1973); Bruce
Moran, The Alchemical World of the German Court: Occult Philosophy and Chernical Medicine
in the Circle of Moritz of Hessen (1572-1632) (Stuttgart: Franz Steiner Verlag, 1991).
2. One important exception to this trend is Pamela Smith, The Business ofAlchemy: Science
and Culture in the Holy Roman Empire (Princeton, N.J.: Princeton University Press, 1994).
3. Of course, students of alchemy had always spent money on books. Roger Bacon (c.
1215-after 1292) noted as much in his 1267 Opus Tertium (Third work) when he wrote,
"Through the twenty years in which I laboured specially in the study of wisdom, careless of
the crowd's opinion, I spent more than two thousand livres in these pursuits on occult books
[libros secretosJ." As cited (and translated) in E . J. Holmyard, Alchemy, 2nd ed. (Baltimore,
Md.: Penguin Books, 1968), 119.
4- William Eamon, Science and the Secrets of Nature: Books of Secrets in Medieval and Early
Modern Culture (Princeton, N .J.: Princeton University Press, 1994), 125-126.
5. Peter Kertzenmacher, Alchimia Das ist aile Farben, Wasser, Olea, Salia, und Alvmina,
damit mann aile Corpora Spiritvs unnd Calces Prepariert, Sublimiert unnd Fixiert Zubereyten :
und wie man diese ding nutze, auff dass Sol und Lvna werden mage: Auch von Soluieren unnd
Scheydung aller Metall, Polierung allerhandt Edelgestein, fiirtrefflichen Wassern zum Etzen,
Scheyden unnd Soluieren: Und zuletzt wie die giJftige Dampff zuuerhiiten, ein kurtzer bericht
(Frankfurt am Main: C. Engenolfrs heirs, 1570), IV. On the various editions of Kertzen-
macher's text and their differences, see Eamon, Science and the Secrets of Nature, 114- 133.
6. Eamon, Science and the Secrets of Nature, 126.
7. Philipp Sommering's testimony, I Alt 9, Nr. 311, fol. 14ff. Niedersachsisches Staat-
sarchiv Wolfenbilttel (hereafter NStA Wolfenbilttel). For printed accounts of Sommering's
life, see Albert Rhamm, Die betriiglichen Goldmacher: am Hofe des Herzogs Julius von Braun-
schweig: Nach den Processakten (Wolfenbilttel: Julius Zwif31er, 1883),3-5, and Jost Weyer,
Graf Wolfgang II. von Hohenlohe und die Alchemie: Alchemistische Studien in Schlof3 Weiker-
sheim, 1587-1610, ed. Stadtarchiv Schwabisch Hall, the Hohenlohe-Zentralarchiv Neuen-
stein and th e Historischen Verein filr Wilrttembergisch Franken, (Forschungen aus
Wilrttembergisch Franken, 39) (Sigmaringen: Jan Thorbecke Verlag, 1992),283-285.
8. Librum Isaacij and Hexameron Bernardij. I have been unable to identify these books and
their authors, although I assume that the latter were hexameral writings attributed to Bernar-
dus of Treves (fourteenth century). On Bernardus of Treves, see William Newman, "Bernardus
Trevirensis," in Alchernie: Lexicon einer hermetischen WissenschaJt, ed. Claus Priesner and Karin
Figala (Munich: C. H. Beck, 1998), 78. The Book of Isaac may refer to the writings of "Isaac
Hollandus," a possibly mythical alchemist whose writings first appeared in the 1560s. See Julian
Paulus, "Holland us, Isaac and Johann Isaac," in Alchernie, ed. Priesner and Figala 181.
9. Philipp Sommering's testimony, I Alt 9, Nr. 311, fol. 14ff. NStA Wolfenbilttel. For
printed accounts of Sommering's life, see Rhamm, Die betriiglichen Goldmacher, 3-5, and
Weyer, GrafWolfgang II, 283 - 285.
10. William Eamon makes this point in Science and the Secrets of Nature, chap. I.

I I. Clearly Lazar Ercker did not identify himself as an alchemist; in fact, as Pamela Long

correctly has pointed out, he is quite critical of alchemy in his published treatises. In practice,

218 Tara E. Nummedal


however, Ercker's projects seem little different from those of self-identified alchemists. Com-
pare his process here, for example, to Hirschenberg's below. The lines between metallurgy,
mining, and practical alchemy are extremely difficult to draw in this period. One person may
call a particular process alchemy, and another may choose not to use the term. I take the
broadest possible understanding of alchemy, one which is defined by practices rather than the
rhetoric that often appeared in printed treatises. On Ercker and alchemy, see in particular
Pamela O. Long, "The Openness of Knowledge: An Ideal and Its Context in 16th-Century
Writings on Mining and Metallurgy," Technology and Culture 32, no. 2 (1991): 318-355.
12. Lazar Ercker to Herzog Julius in Wolfenbiittel, 3 May 1585, I Alt 9, Nr. 394, fol. 1-2,
NStA Wolfenbiittel.
13. Ibid., fol. I.
14. Ibid., fol. 2.
15. Petr Hlavsa's exact dates are unknown, though he did serve as mintmaster to the king-
dom of Bohemia from 1553 to 1561. See Vaclav Bi'ezan, Zivoty Poslednich Roimberkii, 2 vols.,
ed. Jaroslav Panek (Prague: Svoboda Praha, 1985),283,701,801.
16. One Loth = approximately 14.62 g, thus the amount of metal involved in this transac-
tion was actually quite small: 16 Loth = 233.85 g. On Hirschenberg, whose exact dates are
unknown, see Joachim Telle, "Der Alchemist im Rosengarten. Ein Gedicht von Christoph
von Hirschenberg fiir Landgraf Wilhelm IV. von Hessen-Kassel und Graf Wilhelm von
Zimmern," Euphorion 71 (1977): 283-305.
17. Petr Hlavsa to Vilem z Rozmberku (alias Wilhelm von Rosenberg), 18 January 1574,
Rozmbersky roddiny archiv 25, Statni oblastni archiv TFebon (hereafter SOA TFebofl),
Czech Republic.
18. On Polhaimer, see Weyer, Graf Wolfgang Il., 228-271 and his "Der 'Gold macher'
Michael Polhaimer-Alchemistischer Betriiger am Hof des Grafen Wolfengang II. von
Hohenlohe," Beitrage zur Landeskunde. Regelmaj3ige Beilage zum Staatsanzeiger fur Baden-
Wurttemberg 4 (1993): 7- 11.
19. For a description of this contract, see Rhamm, Die betruglichen Goldmacher, 5. SCim-
mering left Gotha in the chaos of war and Reformation politics, abandoning his obligations,
while Scherdinger subsequently took a new post as a pastor elsewhere. In 1571, however,
Sommering signed another contract with Duke Julius of Braunschweig-Wolfenbuttel.
20. Cost- and risk-sharing arrangements like these worked to the benefit of both parties
involved. By giving up a share of their future profits, alchemists gained the initial investment
they needed to put their knowledge to work (not to mention, of course, the social and politi-
cal benefits that accompanied such a position at court). For their part, princes could hope to
earn back their initial capital outlay once the alchemical work was under way.
21. Alexander Lauterwald, Colloquium Philosophicum. Von der warenn Chimia, Sapientia,
vnd Natura rerum, Wie die von menniglichen vnd allen Liebhabern der Kunst von aller Sophis-
terey vnd betriegerylmag vnterschieden vnd verstanden werden. Alles zu Gruendlicher vnd
warhafftiger widerlegung des groben Phantasierens M. Johan ClaljlBengellebischem
Pfarherrslder durch die Altkumistereylandere lerer Gold machenlVnd also das herrlichelvnd ver-
borgenste Geheimnislso vnter allen Natuerlichen dingenlvnd fuertrefflichsten Gaben Gotteslsehr
wenigen bekant vnd offenbaret istlaus lauterm vnuerstand vnd grobheitldem stinckenden
Kuhemist vorziehen thut etc. (Cologne: Heinrich Netessem, 1597). Lauterwald's dates are
unknown.
22. Ibid., fol. A4v.
23. In exchange, Wagenmann received 4,000 gulden, which he pledged to pay back should
he fail. Agreement between Michael Heinrich Wagenmann vom Hoff and Duke Friedrich,
Stuttgart, 23 December 1598, Bestand 47 (Alchemie Sachen), Biischel 3, Number 6, Haupt-
staatsarchiv Stuttgart (hereafter HStA Stuttgart).

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 21 9


24. Melchior Hornug to Vilem z Rozmberku, 4 April 1585. SOA Ti'ebon. On potable
gold (aurum potabile), see Lawrence Principe, "Aurum potabile," in Alchemie, Priesner and
Figala, ed. 66, and the bibliography there.
25. "Praparation des Stein der Weisen, von A. M. Ziegler, in eigenhandiger Anschrift des
Hzg. Julius," I April 1573, I Alt 9, Nr. 308, fo1. 52-70, NStA Wolfenbuttel.
26. Ibid., fol. 64-65.
27. This double meaning makes sense since, in Paracelsian terms, the healing of the body
was analogous to the healing of base metals; in both cases, a chemical medicine was necessary.
28. For an account of Bragadino's life, see Ivo Striedinger, Der Goldmacher Marco Bra-
gadino (Munich: Theodor Ackerman, 1928). Also Hatto Kallfelz, "Der zyprische Alchimist
Marco Bragadino und eine florentiner Gesandtschaft in Bayern," ZeitschriJi fur bayerische
Landsgeschichte 3 I, no. 2 (1968): 476 - 500, and Kallfelz's article "Bragadino, Marco,"
Dizonario Biografico degli ltalianz~ 13: 69 1- 694.
29. Marco Bragadino, Michael Sendivoj, Edward Kelley, and Heinrich Muller von Muh-
lenfels are examples of the former, while Anna Zieglerin and Philipp Sommering are
instances of the latter. See Hermann Kopp, Die Alchemie in iilterer und neuerer Zeit (Heidel-
berg: Carl Winter's Universitatsbuchhandlung, 1886).
30. These included an "oil from gold" (the recipe for which was to be found in the ducal
apartments (in "a special table in which all the alchemical things are together"), a "salt from
gold," an aurum potabile, the coagulation of a "red water which is supposed to have come
from Prag," a process for "finishing" gold learned from another alchemist named Thurn-
heuser (Leonhard Thurneisser?), a praecipitat from Mullenfels, one of the duke's former
alchemists who had been hanged two years earlier for fraud and lese-majeste, a vial to be set
in the fire for two years, eventually to yield a projection of gold, and one last, somewhat mys-
terious, "special process." Inventory from 28 January- 3 February 1608, Bestand 47, Buschel
9, HStA Stuttgart.
y. Ibid.
32. Ibid. (These salary figures are also listed in the 1608 inventory.)
33. Ibid.
34. Evans, Rudolf II and His World, 201.
35. Ibid., 27 6.
36. Moran, The Alchemical World of the German Court, 25 .
37. See Long, "The Openness of Knowledge"; Phillippe Braunstein, "Innovations in Min-
ing and Metal Production in Europe in the Late Middle Ages," Journal of European Economic
History 12 (1983): 563-591; Hans-Joachim Kraschewski, Wirtschaftspolitik im deutschen Terri-
torialstaat des 16. Jahrhunderts: Herzog Julius von Braunschweig-Wolfenbuttel, ed. Prof. Dr.
Ingomar Bog (Neue Wirtschaftsgeschichte, 15) (Cologne and Vienna: Bohlau Verlag, 1978);
Danuta Molenda, "Technological Innovation in Central Europe between the XIVth and the
XVIIth Centuries," Journal of European Economic History 17 (1988): 63-84.
38. Richard Bonny, The European Dynastic States, 1494-1660 (Oxford: Oxford University
Press, 199 I), 420.
39. Kraschewski, Wirtschaftspolitik im deutschen Terntorialstaat, 127- 128.
40. Ibid., 157. The same year, Julius received 42 percent of his income from his mining
enterprises in Rammelsberg and the Harz (152).
41. Duke Friedrich of Wurttemberg, for instance, displayed his desire to exploit the nat-
ural resources in his territories in 1596 when he announced a reward for the discovery of ore
deposits in his lands. The following year, Friedrich established the city of Freudenstadt and
founded a silver mine nearby. In Bohemia, the Czech magnate Vilem Rozmberk bought a sil-
ver mine in Reichenstein (in Czech Rychleby-today Liberec-in Polish Zloty Stok, located

220 Tara E. Nummedal


in Lower Silesia) and immediately established an alchemical laboratory there. See Vaclav
Bi'ezan, livoty Poslednich Rozmberku, 703.
42. Five different methods had been tried earlier in the sixteenth century to deal with this,
which may explain why the mine was bankrupt by the end of the century. See Danuta
Molenda, "Technological Innovation in Central Europe," 75. Also Bi'ezan, livoty Poslednich
Rozmberku, 703.
43. On Uden: "Schreiben vom August I576, Unterschrift und Datum fehlen," 2 Air 24,
NStA Wolfenbuttel. On Topfer: "Schreiben des Theophil Topfer an Sander vom 5.2.I575,"
Fach 2a, ro, Oberbergamt Clausthal.-Zellerfeld Archiv des Oberbergamtes. As quoted in
Kraschewski, Wirtschaftspolitik im deutschen Territorialstaat, I59- I60.
44. "Neigung des Herzogs Julius zur Alchemie; Befehl an die Beamten," 5 June I576, 2
Alt 24, NStAW. As quoted in Kraschewski, Wirtschaftspolitik im deutschen Territorialstaat,
I59. When Duke Julius hired Sommering (after he had left Gotha and his employment with
Duke Johann-Friedrich there), the alchemist signed a contract that reflects this close link
between mining and alchemy: Sommering promised not only to teach Julius how to make the
philosophers' stone, but also to increase the yield of Julius's mines to 200,000 thaler a year.
Rhamm, Die betriiglichen Goldmacher, 8-9.
45. Georg Honauer to Emperor Rudolf 11,5 January I597, Bestand 47, Buschel I, Num-
ber 10 (part I, unpaginated), HStA Stuttgart.
46. "Documents regarding the trial of Georg Honauer," Bestand 47, Buschel 1- 2, HStA
Stuttgart. Similarly, Alchemists Moritz Lam and Georg von Minden offered Duke Julius of
Braunschweig-Wolfenbuttel a process in I576 "by which copper can be made from the Ram-
melsberg lead." "Die angegeben Neue Alchimisten, Moritz Lam und Georg v. Minden betr.,
Vernehmungsprotokoll vom 6.6.1576," 2 Air 24, NStA Wolfenbuttel. As quoted in
Kraschewski, Wirtschaftspolitik im deutschen Territorialstaat, I59.
47. Pamela H. Smith, "Curing the Body Politic: Chemistry and Commerce at Court,
1664-70," in Patronage and Institutions: Science, Technology, and Medicine at the European
Court, 1500-1750, ed. Bruce T. Moran (Rochester, N.Y.: Boydell Press, (991), 195- 209, and
Smith, The Business ofAlchemy .
48. I have looked at the later 1616 edition: Johann Clajus, Altkumistica: Das ist / Die ware
Goldkunst / aus Mist durch seine Operation vnd Process gut Goldt zu machen / Wider die
betrieglichen Alchymisten vnd vngeschickten vermeinten Theophrastisten von Herrn Johanne Clajo
beschrieben: Neben angehencktem Special Bericht / von allerhand geheimen vnd subtilen raenken
vnd Handgriffen / dadurch die Arg Chymisten vnnd des uebrigen Geldes fein artig zu endledigen /
und an statt der verhofften gueldenen Berge Aschen / Kohlenstaub vnd den lehren Beutel zulassen
wissen. Auch wie mit dergleichen Kuenstlern vnd Gabalierern zu veifahren seige. Mennigklichen
zur Nachrichtung und Warnung zusammen gebracht / Durch Aletophilum Parrhesiensem, Jed ed.
(Mulhausen: Johann Stangen, 16(6).
49. "A commendable art is here described / Which became widely accepted / Long ago
among the ancients / From the beginning and the Creation / [This art] is called "old-cow-
manure" (Altkumisterey) / In which there is no sophistry / No false dealings nor fraud / Just
that which one turns up with a plow / On a field which is / well-fertilized and enclosed." Cla-
jus, Altkuhmistica, "Vorrede" [unpaginated].
50. Ibid .
51. Ibid.
52. Ibid.
53. See Massimo Luigi Bianchi, "The Visible and the Invisible: From Alchemy to Paracel-
sus," in Alchemy and Chemistry in the 16th and 17th Centuries, ed. Piyo Rattansi and Antonio
Clericuzio (Dordrecht: Kluwer Academic, (994), 17-50.

Practical Alchemy and Commercial Exhange in the Holy Roman Empire 221
54- Lauterwald, Colloquium Philosophicum Iunfoliated, 16r-v].
55. Ibid. [unfoliated, 23 v.]
56. See Berend Strahlmann, "Chymisten in der Renaissance (16. Jahrhundert)," in Del'
Chemiker im Wandel del' Zeiten; Skizzen zur geschichtlichen Entwicklung des Berufbildes, ed.
Eberhard Schmauderer (Weinheim: Verlag Chemie, 1973),47 - 55,
57. Michael Maier, Examen Jucorum pseudo-chymicorum detectorum et in gratiam veritatis
amantium succincte reJutatorum (Frankfurt: printed by Nicolai Hoffmann, published by
Theodor de Brij, 1617), 23. Reprinted (and translated into German) in Wolfgang Beck,
"Michael Maiers Examen Fucorum Pseudo-Chymicorum-Eine Schrift wider die falschen
Alchemisten" (Ph.D., Fakultat fUr Chemie, Biologie und Geowissenschaft der Technischen
U niversitat Munchen, 1992).
58. Lauterwald, Colloquium Philosophicum run foliated, 26-26v].
59. On Khunrath, see Elmar Gruber's introduction to Vom Hylealischen, das ist, Pri-mate-
I'ialischen catholischen odeI' allgemeinen natiil'lichen Chaos, del' naturgemiissen Alchymiae und
Alchymisten. (Magdeburg, 1597; reprint, Graz: Akademische Druck- und Verlagsanstalt,
I990), v-xix.
60. This text, "Wahrnungs-Vermahnung an aIle wahre Alchymisten, sich vor den
betrugerischen Arg-Chymisten zu huten," is appended to his Vom Hylealischen, das ist, Pri-
materialischen catholischen oder allgemeinen natiil'lichen Chaos, del' naturgemiissen Alchymiae
und Alchymisten (Magdeburg, 1597).
6L Khunrath, Vom Hylealischen, 268.
62. Ibid., 286.
63. See Karin Figala and Ulrich Neumann, '''Author Cui Nomen Hermes Malavici': New
Light on the Bio-Bibliography of Michael Maier (1569-1622)," in Alchemy and Chemistry in
the 16th and 17th Centuries, ed. Piyo Rattansi, and Antonio Clericuzio (Dordrecht: Kluwer
Academic, 1994), I3 8- 139·
64- Michael Maier, Examen Jucorum pseudo-chymicol'um.
65. Ibid., ro.
66. Ibid., 22.
67. "Maier even cited the law to support his point: "The Civil Code permits people doing
business to haggle with one another while buying and selling, but not that more than one and
a half times the true worth [of the wares] be paid or asked ... Regarding this, see the com-
mentary to the section [of the Civil Code] regarding buying and selling. With regard to the
wares of pseudo-chymici, nothing is in proportion, because they offer only words and
promises for gold. How much empty wind must one accept for the equivalent of a gold
drachma, is impossible to determine. In sum: he who touches tar gets dirty hands, and he who
gets involved with these people only ends up with an empty purse." Maier, Examen Jucorum
pseudo-chymicorum,22
68 . Kaiserlich Privilegirtel' Reichs-Anzeiger (Gotha), 8 October 1796 (Num. 234), col. 6095.

222 Tara E. Nummedal


Time's Bodies
Crafting the Preparation and Preservation of Naturalia

HAROLD J. COOK

S cientific investigation deals with the secular world: the world of time.
Scientists study manifestations of the universe as they unfold. Time
may be the fourth dimension, but it is so important as to commonly
become one of two axes on graphs and charts. Consequently, many of the
methods employed by investigators of nature attempt to slow or quicken
events. In recent decades, some of the most famous visual examples of this
method include serial photographs of movement, slow-motion films of var-
ious behaviors too quick to be observed with the eye, or fast-forward films
of slow transformations. In the seventeenth century, studies such as William
Harvey's on generation involved the serial examination of fertile chickens'
eggs and does' uteruses; his even more famous discovery of the circulation
of the blood equally involved vivisectional techniques that slowed the heart-
beat and pulse to events that could be seen with the naked eye. Some of the
important investigations into the material structures of animal bodies car-
ried out in the seventeenth-century Netherlands also had their roots in
attempts to alter the processes of ordinary time, especially those associated
with decay. The consequences of such trials proved to yield unexpected
results, in making visible structures that could not otherwise be ascertained.

TIME'S VALUE: ACCUMULATION, PRESERVATION, AND THE FUTURE

Most historical discussion of changing concepts of time has focused on the


development of a sense of its uniformity. Time may be an indivisible wave
or a composition of streaming quanta, but in either case the principle that it
is uniform, allowing one moment to be compared to another, is fundamen-
tal, even in an Einsteinian universe where the speed of matter can "slow" or
"accelerate" the time of one object in relation to another. It was famously

223
the invention of mechanical timekeepers that conveyed the view that time is
uniform. The sense that time changes with the seasons, with one's age, with
peace or war, during moments of stress or bliss, accords with human felt
experience. Classically, night and day were divided into twelve parts which
varied with the season: the hours of night were longer in the winter, whereas
the hours of light were longer in the summer. An advantage of hourglasses
and water clocks is that they could be easily altered to accord with such vari-
ations. But mechanical clocks moved steadily, invariably (aside from
mechanical inconsistency), dividing the day into equal hours. Now it was the
night or day that changed according to the hour, not the hour that altered
according to the light. Clocks quickly made their appearance in the towers
of guildhalls and other municipal buildings, striking the hours to regulate
commerce and other activities of large numbers of people: workers in the
Low Countries had been complaining of "working to the clock" from at
least the fourteenth century.' By the lifetime of Galileo and Descartes, both
natural philosophers and musicians could take the uniform nature of time
for granted, "timing" events according to regularized beats. 2 As Norbert
Elias put it, "The significance of the emergence of the concept of 'physical
time' from the matrix of ' social time' can hardly be overrated."3
In the same period, however, another sense of the relationship between
time and human life that had importance for the development of investiga-
tions into medicine and natural history: a growing sense that new methods
of using time could bring material good. The ways of life developing
among early modern capitalists helped to make questions of manipulating
the ordinary processes of time seem natural. Capitalist forms of economy
depended not only on drawing attention to the rapid passage of time, but
also on making work more regularized. The Dutch financial world also
depended on new methods of commerce that extended time: long-term
arrangements. As a world of markets was being transformed into a world
market, "the beating heart of the Dutch economy" was its "entrep6t func-
tion." That is, the Dutch Republic functioned as a kind of clearinghouse, to
which surplus goods were shipped from all over the world for exchange
and redistribution. The goods (and their abstractions in the form of bills of
exchange) were traded daily at the Amsterdam Exchange (Beurs) rather
than at periodic fairs, making it "a mustering field not only for the coinci-
dental surplus production ... but also for information" about commodities
and exchanges worldwide, helping to stimulate collective decisions by mer-
chants on the allocation of capital. 4 Moreover, not only the commodities
themselves, but their future worth , could be bought or sold in the form of
"stocks," which were paper representations of accumulations of material
things. Accumulation of inventory for later (sometimes much later) sale
allowed the universal principle of "buy low and sell high" to operate over the

Harold J. Cook
middle and long term: items could be kept back from the market when they
were otherwise too plentiful, and sold when demand increased. Inventory
investment therefore reduced and spread out financial risk, making trade more
predictable, and allowing for stabler calculation of future income and expense.
Inventory investment also required that goods set aside for future use or
sale made it into the future in good condition. The building boom in dock-
and canal-side warehouses is the clearest sign of the new form of merchant
capitalism: here nutmeg or tea, there Persian rugs or Chinese silks, were col-
lected, itemized, and assigned an estimated value for the market. In the
warehouse, goods could be safeguarded and stored in a manner that pre-
vented decay. Nutmeg, for instance, was shipped back from the Indies
dusted with lime. Some other very valuable fruits came preserved in sugar.
Salting and pickling kept other fish, meat, and vegetables from putrefaction.
Most other goods could simply be dried and kept dry. It was this combina-
tion of methods of accumulation, preservation, and calculation of future
value that allowed the merchant capitalists of the period to flourish. For
instance, the Dutch East India Company (VOC) was able to manage the
price of cloves by accumulating and preserving the spice in its warehouses.
When the price of cloves fell in 1623, they held back cloves from the auctions
to drive the price back up; when the price was so high that the English broke
into the Dutch monopoly, they resorted in turn to dumping to drive the
English out. With the accumulation and storage of commercial goods, there-
fore, a "certain measure of control over the prices" could be achieved, which
together with "the concentration of the East India trade in the intimately
connected London and Amsterdam markets must have meant a greater
transparency of the market."5
As the simple example of cloves illustrates, "investing in inventory [was]
crucial to a smooth functioning of the market." The permanent staple mar-
ket that developed in places like Amsterdam and Rotterdam served to con-
centrate supply and demand, which "reduced the commercial risk, so that
the cost price decreased. As supply was less regular than sales, prices fluctu-
ated. These price fluctuations offered the prospect of future profits and thus
stimulated stockpiling which, in turn, had a stabilizing effect on the price."
Holland consequently became "a central storehouse and exchange" for the
world market. "And inventory investment was-as already indicated-at
the very heart of the Dutch entrepot trade, which in turn was the focal point
of the commercial expansion of the Dutch economy."6 In short, inventory
investment helped to create stability in "the market," hence increasing con-
fidence in it, while growing confidence in turn helped to lower interest rates
and raise the amount of available credit for more investment in exchange
and accumulation. The calculation and sale of future value gave rise to new
forms of material life, with ramifications for intellectual culture as well.

Time's Bodies 225


Like the developing world market, the investigation of nature depended
on the transportation of information and specimens back to the home
metropolis. There collections were accumulated, housed, and preserved,
inventories were taken and sometimes published, and redistribution of the
value-added information and objects occurred. The ways of life that so val-
ued the accumulation of bulk commodities also valued the accumulation of
unique objects, whether works of art or nature. Among the items brought
back to the Dutch entrepot from all over the world were natural rarities and
curiosities of all sorts. These things at first tended to be one-of-a-kind, or at
least scarce, objects, brought back in the bags of seamen or the chests of offi-
cers and merchants. Other specimens were cultivated in botanical gardens.
Over time, a steady trade in naturalia developed, with a few brokers even
buying up objects at dockside and later selling them to collectors. In short,
many people began to collect various kinds of objects from nature, although
they placed them not in shelters by the docks but in rooms in their homes:
the "curiosity cabinets" of burgers, physicians, magistrates, and nobles (fig.
9. I). As in business, too, the collectors of naturalia kept detailed inventories
of what they had. To make their collections more valuable, they tried to fill
them out with new specimens, and they bought books describing the collec-
tions of others as a way of substituting for what they could not acquire
directly. The accumulation and warehousing of material objects-invest-
ment in and preservation of inventory - was part of the creation of value
for both naturalists and merchants.
In the sixteenth and early seventeenth century, most objects kept in
curiosity cabinets had to be dried. Inventories of later sixteenth- and early
seventeenth-century cabinets make this clear. For instance, one of the earli-
est cabinets of natural history and art assembled in the Netherlands, by
Bernardus Paludanus (Berent ten Broecke), contained fruit, grains, and
woods from the tropics, skins with feathers prepared from many birds (such
as the bird of paradise of New Guinea), many species of fish and reptiles, the
horns of a variety of animals, insects, shells, corals, types of earths, stones,
minerals, marbles, precious stones, coins, medals, weapons, clothes and other
objects used by "savages" and foreigners, objects of art done in ivory, rare
woods, precious metals, mummies and funerary furnishings from Egypt,
and so on. 7 Perhaps the fruits were pickled or preserved in sugar as well as
dried; some of the skins were tanned; everything else could have been kept
dried. In another example, the first inventory of the natural history collec-
tion assembled at Leiden University shows that it contained skeletons and
diverse bones of humans, animals, birds, fish, and other items such as horns;
rarities such as mummies and their parts, seven stones surgically removed
from the bladder of Joannes Heurnius (the first medical professor of Lei-
den), a stone from the kidney of a young girl, and so forth; various lists and

Harold f. Cook
&**Lw 9.i Title page from Ole Worm's catalog of his collection of curiousities,
mainly naturalia, published in 1655. All the specimens are dried. They include
human artifacts; stuffed fish, birds, and animals; and skulls, horns, minerals, stones,
salts, earths, shells, corals, seeds, leaves, and roots. Courtesy o f ~ e l l c o m eLibrary, ond don.

placards; portraits and paintings, large and small; surgical and anatomical
instruments; a large Egyptian mummy, windings of Egyptian linen, Chinese
paper, and paintings of exotic fruits, nuts, woods, stones, and so on; and
other odds and ends. There was also the liver of a young woman of seven-
teen anatomized by Otto Heurnius in 1620, and the vital organs of pigs.
While it is possible that these items were pickled or kept in brine (neither of
which preserves the structures for close later inspection), they, too, were
more likely dried.8 So it goes with other collections of naturalia: they were
based on dried specimens. The result was that collectors could see the shape
of things, their forms, but with few exceptions, not their inner structures.
Physicians and pharmacists had long investigated methods for the preser-
vation of small quantities of valuable biologicals. Apothecaries had helped to
foster the majolica pottery industry and perhaps helped to stimulate the
early modern glass industry as well, since both kinds of containers helped
enormously in preserving medicaments by keeping out light, moisture, and
air. Apothecaries and physicians were also among the first to pioneer the use
of chemical processes for the preparation of medicines; in addition to other
--

Timei Bodies
advantages, chemical preparations were not as subject to decay as their bio-
logical counterparts. In the sixteenth century, a new technique for preserv-
ing botanical material became indispensable for the study of herbal
medicines and botany: the first professor of simples at Bologna and first
director of the botanical garden at Pisa in the 1530S and 1540s, Luca Ghini,
developed a method of taking plants or parts of them and pressing them
firmly between sheets of paper while they dried, which preserved their form
(and temporarily their color) for later study.9 These herbaria gave tremen-
dous aid to botanical study, although there remained no substitute for study-
ing living plants in gardens, which consumed huge investments of time and
money in the period. Not all plants could be studied in European gardens,
however, since many exotics, from tropical climates in particular, died
within a season or two. Not until techniques came along like the building of
glass houses coupled with special furnaces in the mid to later seventeenth
century-a subject handled so well recently by Chandra Mukerji-could
the inner structures of many plants be studied.
The study of animals and their parts was even more difficult. Anatomies
had to be carried out with some rapidity, especially when working with soft
tissue, because of the rapid onset of putrefaction; this is one reason that pub-
lic anatomies tended to take place in the cold winter months, despite the
darkness of the season. Animal parts could be studied at leisure only if dried
(as with skeletons) or tanned, neither of which allowed for the investigation
of the structures of the body.
Methods to counteract the natural senescence that came with time had
been long sought, however. Roger Bacon, for instance, argued that "men
used to know what to do about premature physical deterioration: 'per expe-
rientias secretas' it had been discovered and written that this rapid aging is
accidental (having avoidable side-effects) and therefore can be treated. The
medical art cannot achieve this but the experimental art can."10 His works
on the subject were translated into English in 1683.11 Many other philoso-
phers and alchemists-to say nothing of Ponce de Leon and the search for
the fountain of youth - hoped to discover means to prolong human life to at
least the age of the biblical patriarchs. Sir Francis Bacon believed that the
classical third part of medicine (after preserving health and curing disease),
the prolongation of life, would be much improved by his reform of natural
philosophyY Rene Descartes, too, believed that one of the most important
results of his own intellectual reforms would be the prolongation of life. 13
Similarly, finding better means to prevent decay and putrefaction also
became subject to study. By the middle of the seventeenth century, the rapid
growth of new methods of forestalling time's natural processes can be dis-
cerned. Although it was a goal which today seems more modest than the
prolongation of life, this seems so only in retrospect, probably because solu-

228 Harold j. Cook


tions to problems of decay began to be found whereas medicines to prolong
life remained the stuff of legend. We may think the development of meth-
ods for preserving biological specimens worthy oflittle remark only because
we take them for granted. But in ordinary experience, decay and aging
appear to be closely related. In the seventeenth century, then, methods to
preserve the bodies of living things seemed almost miraculous. They also
made the transition from "dry" to "wet" collections possible. The resultant
spin-offs for the investigation of nature had important implications.

PRESERVING A LIFELIKE BODY

At first blush, the motivations for trying to preserve animal bodies in a nat-
ural-like state seem obvious. As Herman Boerhaave explained in his account
of Jan Swammerdam:

Having gone through his courses [in medicine] with the most sudden and
unexpected success, he immediately began to consider how the parts of the
body prepared by dissection, could be preserved and kept in constant
order and readiness for anatomical demonstrations; as such a discovery
would free him not only from the trouble of repeated dissections, but like-
wise from the difficulty of obtaining fresh subjects, and the disagreeable
necessity of inspecting such as were already putrefied. 14

The details of how Swammerdam and others came to develop methods for
preserving whole bodies and body parts is somewhat more complicated,
however.
University-educated physicians and surgeons like Swammerdam were
preceded and stimulated by the work of Louis de Bils, lord of Coppens-
dam me and Bonem (both modest fiefs in Flanders).15 According to the
report of Samuel Sorbiere, De Bits (b. about 1624) had begun dissecting at
the age of thirteen when living in Rouen, and afterward in Flanders and
Rotterdam. Why he developed his interest and how he obtained bodies to
dissect are both unknown. His father and brothers were merchants, and he
himself seems not to have had the kind of good classical education expected
of savants. Yet by 1646-47, De Bils had a family and was in Amsterdam,
where he met the anatomist and surgeon Paul Barbette and the physician-
chemist and anatomist Frans;ois dele Boe Sylvius; a few years thereafter he
had taken up residence in Sluis, in Dutch Zeeland not far from Middelburg,
a flourishing port city. He continued his anatomical studies and in 1651 gave
the University of Leiden a number of preparations made at great expense,
acknowledged in a written testimonial by the new professor of anatomy,

Time's Bodies 229


Joannes Van Horne. It contained a particularly remarkable specimen:
"Above all else is a dried human cadaver that appears to be freshly dead, the
most worthy work for such a theater" of anatomy, Van Horne attested. 16
Shortly thereafter Van Horne saw in The Hague another body "balsamed"
by De Bils, in which the sinews and plump muscles were displayed as if
alive. Apparently working alone, De Bils had found means to prepare
human bodies so that they appeared to be full of life rather than dessicated.
His secret process of "balsaming" was a fantastic new art in both senses of
the word.
De Bils's method was apparently stimulated by the example of Egyptian
mummies. Mummies were common objects in the curiosity cabinets of the
late sixteenth and early seventeenth centuries, and the period also saw the
flourishing of speculation about the wisdom of the Egyptians. 17 Because
the brain and viscera were extracted and the remaining skin and muscle
were hardened by the preservative process,18 Egyptian mummies had little
to offer the anatomist. But they had long held people fascinated by the
length of time they lasted without decay. In an age before modern methods
of embalming, they were true wonders. Most European churches and grave-
yards were places where bodies were buried and decayed, the grounds being
continually redug for the burial of additional bodies; sometimes the bones
that were recovered in the process were collected in heaps under eves around
the outer walls - the charnel houses. Shakespeare's famous "alas poor
Yorick" soliloquy is located in such a graveyard, contemplating the skull of
a past acquaintance. While the wealthy and powerful might have a stone
tablet or other enduring memorial erected in their memory, it was the sign
of a miracle to possess a body that did not rot after death. 19 De Bils himself
was involved in an investigation of the bodily remains of Maria Margaretha
van Valckenisse, mother of a cloister in Oirschot who died in 1658 yet did
not decay but rather gave off a sweet oil, declaring the causes to be natural
and fraudulent rather than miraculous. 20 Mummies were therefore quite
unusual for lasting so long after death.
The power of overcoming putrefaction also made mummies and pieces of
them, or more often the powder from pieces of them (both called mumia), a
sovereign remedy in all kinds of complaints. Karl Dannenfeldt has given us
an excellent account of the early modern debate over mumia. He explains
the historical process as one by which the use of bituminous products in
medicine was transferred to embalmed or desiccated bodies. The precious
seepage of black rock-asphalt or pissasphalt from a mountain in Persia,
locally called "mumiya," became particularly well regarded. By the thir-
teenth century, the resinous, aromatic substance exuded from bodies found
in Egyptian tombs was considered to be a very similar product. Since asphalt
was said to be used by the Egyptians for embalming their dead, the true

Harold]. Cook
mumia could be found in the cavities of the head and body in the "mum-
mies." It was a short step to considering the embalmed flesh-and even the
wrappings-to contain the precious resin.21 Antonius [Mus a] Brasavola's
Examen omnium simplieium medieamentorum (1537), defined "mumia as the
remains of an enbalmed body and the same as bitumen judiacum."22 The
textbook on chemistry by Joachim Tanckius, professor at Leipzig, simply
stated that "Mumia is the arcanum and secret of the microcosm."23 Conse-
quently, European demand for mummies became so high that the Egyptian
government was forced to outlaw the export of mummies, although a large
contraband trade in both true and counterfeited mummies continued
through the early modern period. 24
Medical practitioners also attempted to make mumia themselves .
Paracelsus and his followers, for instance, described mumia as a force in liv-
ing tissue that attacked invading disease semina. Andreas Tentzel's Medieina
diastatiea (1629) "was primarily devoted to mumia, of which he enlarged the
scope and definition. Now there was extraction of the mumia of the aerial
body by interception of the dying breath."25 More practically, however, flesh
from those who died healthy and without disease - especially those who
died a violent death - was thought by Paracelsus and his followers to still
radiate the power of mumia, so that the flesh of the recently deceased could
be used after exposing it to the air for a day and a night. Oswald Croll was
even more precise: the best tincture of mumia was prepared from the flesh
of a "red-haired man twenty-four years old, who had been hanged, broken
on the wheel, or thrust-through, exposed to the air for a day and a night,
then cut into small pieces or slices, sprinkled with a little powder of myrrh
and aloes, soaked in spirits of wine, dried, soaked again, and dried." From
this could be extracted a red tincture, "a quintessence, which could be used
for cures of pestilence, venin, and pleurisy."26 (Perhaps the trade in mumia
affected the debate about cannibalism in early modern Europe) Y It is no
surprise, then, to find Dutch physicians setting recipes for making mumia
alongside receipts for embalming bodies. 28
As for the process by which the Egyptians mummified their dead, the
main ingredients were thought to be myrrh and aloes, as well as other resins.
It is probably needless to remind anyone that in the Christmas story Magi
bear gifts of myrrh and frankincense, two resins from "Arabia" (as Euro-
peans knew the Near and Middle East) very valuable in medicine; closely
related to the resins, according to Dioscorides, were the substances in the
category pitch, including asphalt and other substances related to the original
"mumiya." As one commentator explained: in order to thwart the usual
course of putrefaction, the Egyptians disemboweled the dead and repeatedly
steeped them in bitumen and stuffed them with precious aromatics. 29 Aro-
matic and oily resins were also those things that often went under the rubric

Time's Bodies 23 1
of "balsam," or in English, "balm." It is clear that De Bils was experiment-
ing with various expensive oils and resins such as myrrh when he engaged in
"balsaming" (balsemen remains the Dutch verb for the English "embalm-
ing"). Given the high prices of the imported balms, his experimental costs
must indeed have been enormous.
What De Bils was trying to do went further than what the Egyptians had
done, however. For mummification preserved only the external form of the
body. While the body endured permanently and could be examined without
any effusion of blood or fluids that might offend those with delicate sensi-
bilities, mummies were useless for the anatomist. The bodies were hardened
and the viscera absent. lo De Bils, however, was developing methods of pre-
serving the whole body in a lifelike manner.
De Bils's personal affairs suffered badly in the early r650s, but he kept up
his investigations. After the death of his father, he and his brothers, mer-
chants in Rouen, became embroiled in various lawsuits against one another
about the inheritance. Although he obtained the office of bailiff of Aarden-
burg, the pay was slight, and he seems not to have invested much energy in
the position. 31 But two medical friends in Sluis, Drs. Abraham Parent and
Laurens Jordaen, both of whom had studied at Padua, helped De Bils with
his anatomical work and jointly published a pamphlet on De Bils's investi-
gations on the anatomy of the inner ear. They both moved away in the mid-
r650s, however, lessening De Bils's opportunities for anatomical study.32 By
r657, De Bils was searching for new means of support. A physician in
Brugges (not far from Sluis), Burchardus Wittenberg, wrote a tract highly
praising De Bils's achievements and calling on a prince to support him, so
that his work was not paid for out of his own pocket. Through an interme-
diary De Bits tried to interest professor Van Horne in working with him,
but Van Horne seems to have balked at the probable expense. De Bils did
finally get the financial support of a physician from Middelburg for his
research and publication on the lymphatics. But this publication hit a nerve
with Van Horne, who expressed complete surprise at De Bils's work. Van
Horne quickly turned out a Latin translation of the book, although criticiz-
ing it at the same time. 11 According to the historian G . A. Lindeboom, Van
Horne himself "now applied himself to the making of fine anatomical
preparations"34- a matter to which we will return in a moment.
Given De Bits's successes, the States General of the Dutch Republic issued
an order on 9 August r658 for the public provision of bodies to De Bils,
while new translations into Dutch of anatomical works by Thomas
Bartholinus and Paul Barbette allowed De Bils to study further. His friend
Parent also published a notice again urging support for De Bils's work,
which was so costly-especially the balsams. The city of Rotterdam, to
which De Bils had followed Parent, set up an anatomical theater over the

232 Harold f. Cook


former English merchants' courthouse. De Bils used it for further studies on
his secret method of dissecting and balsaming, and for the display of at least
four dissected and balsamed cadavers. Apparently a "sovereign power"
sought to get him to sell his secret several times, but he refused in favor of set-
ting up his show in Rotterdam, for which he charged an admission of r rijks-
daalder. Despite the high entry fee, his display was heavily attended by
physicians and students as well as the public, from ordinary people to ambas-
sadors and princes. He also held public anatomical demonstrations. His spe-
cial technique was to dissect without losing any blood or other moisture from
the body. For such theatrical presentations he charged even more. 35
But after initial support, encouraging students and others to attend De
Bils's displays and demonstrations, Van Horne turned against De Bils in
writing. He had become proud, telling the world that students learned more
from him in half an hour than from Van Horne in two years. Van Horne in
turn decided that De Bils was a pretender, with neither learning nor gentle-
manly behavior. The Amsterdam surgeon Barbette also turned against De
Bils. He, too, underlined the absence of academic education in De Bils: "Phi-
losophy, chemistry, astronomy, medicine, and daily practice" were absolutely
necessary for understanding the workings of the body, but for two years the
unlettered De Bils had pretended to be the great master overturning all
established learning. De Bils replied to these two sallies with his own pam-
phlet of March r660, in which he ascribed Van Horne's and Barbette's criti-
cisms to jealousy. But in attacking them he also further attacked, and further
alienated, learned physicians generally. The debate continued throughout
De Bils's life (he died in r669)- he had some supporters in the Dutch
Republic as well as antagonists. But the one matter that continued to be
praised even by his strongest opponents was his balsaming of cadavers.36
De Bils promised to reveal the secret of his process for r20,000 guilders.
But he was willing to part with two balsamed cadavers to one Duke Chris-
tiaan for r6,000 guilders. At the end of r66r, it was rumored that he had sold
his secret to a nobleman. This may well have been prompted by the attempts
of Luis de Benavides Carillo y Tolede, marquess de Caracena, a follower of
Don Jan van Oostenrijk, stadtholder of the southern Netherlands, to pur-
chase De Bils's collections for the University of Louvain. After inspection of
some of De Bils's cadavers in November r662, Gerard van Gutschoven, a
professor of medicine there, became quite enthusiastic about the possibility
of obtaining De Bils's specimens. Oostenrijk proposed to the states of Bra-
bant that they purchase De Bils's cadavers and his secret method. By June
r663, eighteen articles had been drawn up by which De Bils agreed to pro-
vide Louvain with five cadavers and all his knowledge, including his secret
embalming process. The process, and his method of bloodless dissection,
would be written in duplicate in Latin, with a Dutch version for De Bils; the

Time's Bodies 233


two Latin copies would be deposited in separate places, in strongboxes sealed
by two keys, one key to be held by the states of Brabant and the other by the
professors of Louvain. Various other provisions ensured that De Bils swore
that he had not and would not reveal the secret to anyone else. In return, the
states promised a payment of 22,000 Rhenish guilders, and a professorship
salaried at 2,000 guilders per year, which would revert to his son after his
death. De Bils would also establish an anatomy theater in Louvain without
charging for admission. By October some changes were made to this draft
contract, and Van Gutschoven began to learn De Bils's secret under his tute-
lage. At the same time, as word of the arrangement got out, people began to
insist in their wills that their bodies be embalmed by De Bils's method. 37
Finally, on 16 April 1664 De Bils's secret was handed over in writing and
shown to Van Gutschoven, who was allowed eleven minutes in private to
read it, after which he stated that he understood the methods of bloodless
dissection and balsaming of bodies. By May, the five bodies De Bils owed the
Louvain faculty were in hand, and the states paid out the 22,000 guilders.
There was as yet no place prepared for the cadavers, and so they were placed
in the basement of the library; after four hot months, they were laid out on
tables under a roof with holes in it, subject to rain and snow, which not only
damaged books in the library but caused some signs of rot to appear in three
of the five cadavers by 1666. This later became known to De Bils's oppo-
nents, who claimed it proved him a fraud. But despite these and financial
difficulties, De Bils remained well regarded in Louvain and the southern
Netherlands: in early 1669 Flanders awarded him the benefice of a canon of
's Hertogenbosch and St. Oedenrode, and made him an honorary professor
of anatomy at the Illustrious School. During that same year, several public
demonstrations of his method were to be undertaken in the northern
Netherlands with the assistance of Tobias Andreae- but De Bils sickened
and died. Andreae lent his help to trying to sell De Bils's secret in Amster-
dam, coming into difficulties with the Leiden-educated physician and sur-
geon Frederick Ruysch in doing SO. 38
The process De Bils had developed and written down in 1664 was as fol-
lows: A tin box [tinne kiste] 8 feet long by 2 Ih feet wide by 3 feet high was
placed in a wooden box trimmed and caulked so as to let in no light and
fixed with iron bands; into the lid of the wooden box was cut a trap door
[schuyve] that could be opened and completely sealed. The tin box would
also be covered at the appropriate time with double wool blankets so that no
light could enter. Into the tin box was introduced 60 pints of the very best
rum, freshly made; 50 pints of Roman alum very finely ground; 50 pints of
pepper very finely ground; 1 sack of salt finely ground, which must be
poured in at this point; 200 large glasses [stoop] of the very best brandy of
Nantes; 100 large glasses of the very best wine vinegar, all of which were

Harold]. Cook
well mixed in the tin box as quickly as possible so as not to let the power of
the mixture get lost [opdat de kracht niet te veel en verlighe van ditto substantie]'
Twenty pounds of finely ground myrrh of the best kind and 20 pounds of
the best finely ground aloes [allouwe] could also be added to the mixture.
The dead body, wound about with a white linen sheet, was immediately
dunked in this mixture, lying on and tied to a wooden platform [stellinghe]
so that at least two feet of fluid covered it. The boxes were closed for thirty
days, except that three days after the body was put in the fluid, the mixture
was well stirred, as it was twice more during the thirty-day period. Each
time the fluid was stirred, the body was also taken out, unwrapped, washed
in fresh brandy, flipped over so as to drain out any moisture via the mouth
(being careful not to damage the hair or finger- and toenails), rewrapped in
sheets, and replaced. After the thirty days, the body was transferred to
another box made like the first with a mixture of rum, pepper, alum, salt,
brandy, and vinegar in the previous proportions, in which it was left for sixty
days (with three stirrings and turnings). The above mixtures were for kings
or others whose bodies were to be displayed in public. If this was not to be
the case, the rum and alum could be left out of the first mixture and the
spices had to be added, and in the second mixture no salt was added, nor
rum and alum. Between the second and third soaking, the body was allowed
to dry. The first box was in the meantime cleaned and filled with a third
mixture, which excluded the rum, alum, and salt but included the myrrh
and aloes; this mixture was stirred several times and the clear liquid that
came to the surface was skimmed off. Then 44 pounds of aloes, 44 pounds of
myrrh, 20 pounds ofJoullie, 20 pounds of cloves, 20 pounds of cinnamon, 20
pounds of nutmeg (all of the best kind, finely ground), 114 pound of amber-
gris, 1/4 pound of black balsam, with a 112 pound of oil of cinnamon were all
mixed together and applied several times to the exterior of the body and
allowed to dry. The body must lie in the third mixture for two months,
being turned over periodically as before, being washed and rinsed with the
clear liquid skimmed off previously. If after all this the body fat had not
completely dried up, the body would be placed in a small, tight stone room
with two ovens burning low, one of which burned 2 pounds of mastix. After
the body was thoroughly dried, the ambergris mixture was applied to the
body again. The specimen could be best kept in a tin box that let in no air. 39

UNFORESEEN CONSEQUENCES: THE STRUCTURES OF THE BODY

Because De Bils had kept his method secret, however, others had to guess at
the means and experiment with possibilities themselves. In March 1661 it
was rumored that a Dr. Hubertus of Leiden had discovered some of De

Time's Bodies 235


Bils's secrets, and later in the year a story was circulating that De Bils had
sold his secret to a nobleman who passed it on to one Burrhus in Leiden-
although nothing more is known. 4o A student at Leiden at the time,
Theodorus Kerckring, "is said to have invented," a means of "preserving
dead bodies by covering them with varnish."41 In another version, he "per-
formed experiments with liquefied amber to preserve corpses."42 Another
contemporary, Gabriel Clauder, thought that De Bils was using salts. A
medical student at Leipzig (and later physician to the electors of Saxony),
Clauder was making a grand tour of Europe and England in 1660 and
1661 43 when he visited De Bils's cabinet. According to his 1679 Methodus bal-
samandi corpora humana, he "applied his moistened finger to one of the bod-
ies, and carrying it to his lips recognized the taste of salts. He started from
this fact to attempt numerous researches, and succeeded in forming differ-
ent compounds." His salt was composed as follows: "Dissolve one pound of
common salt with a pound of oil of vitriol [sulfuric acid] in a crucible, apply
a cover closely luted, and distill it gradually in a sand bath; you may pour off
a spirit very excellent for a lotion; in the bottom of the crucible will remain
a caput mortuum, which should be dissolved according to art, and after evap-
oration, you will have the salt so much esteemed by the author."44
Better known are the investigations of Jan Swammerdam, who matricu-
lated in medicine at Leiden in 1661 after already having had experience with
anatomical work at home in Amsterdam; he quickly became one of the
favorite pupils of Van Horne and of Franc;:ois dela Boe Sylvius. In 1652 Van
Horne had announced the discovery of the thoracic duct; in the early 1660s,
Swammerdam made a durable preparation of it by soaking it in alcohol and
then drying it. 45 But the centerpieces of Swammerdam's cabinet were a pre-
served child of one month, and a whole lamb, both of which he balsamed
using a process simpler but similar to De Bils's. According to Justus
Schrader-a slightly younger student of Van Horne's-Swammerdam's
technique of balsaming was as follows: First, a tin vessel large enough to
receive the corpse was prepared. Into this was set a grate or screen resting
two fingers' width above the bottom, on which the body was placed. Then
oil of turpentine was poured in to a height of three fingers' breadth from the
bottom. The vessel was covered tightly except for a tiny opening, and set
aside for time to do its work. This most penetrating oil entered the pores and
replaced the fluids that caused fermentation and decay, which due to their
weight descended through the screen to the bottom of the vessel while at the
same time the volatile oils evaporated through the small opening in the top,
leaving the specimen coated throughout with the hardened oil, which pre-
vented it from decay.46 Different organs required longer or lesser times: an
embryo took six months, a skeleton about two, the parenchymia of the heart
three, a liver and a placenta one, a spleen ten days, and intestines a month. 47

Harold J. Cook
rSl5~ .9.2 Engraving of a preserved female human uterus as depicted by Jan
Swammerdam, from his Miraculum Naturae Sive Uteri Muliebris Fabrica {1672}.
The structures of these soft tissues were discovered using preservatives and injec-
tions of various kinds. Courtesy of the Universiteit Bibliotheek Leiden.

A few other techniques helped prepare more complicated specimens. With


bodies and organs properly prepared, one could also inflate the vessels with
air, wax, mercury, and other substances. Using such techniques, Swammer-
dam was able to examine the structure of the lung, the follicles of the human
uterus, the ramifications of the vessels of the placenta, and so on 48 (fig. 9.2).
With a few elaborations, the method developed by Swammerdam contin-
ued to be taught at the University of Leiden. For instance, Carel Maets (or
De Maets, Dematius), who had been teaching experimental chemistry at
Leiden since 1669, explained his private method of preserving bodies from at
least 1674.49 He elaborated the method in his Chemia ratianalis of 1687:
"After first removing the intestines, viscera, brain, and all other soft parts, it
is then placed in a lead coffin [cysta] commodious enough for it, where it is
soaked in clear oil of turpentine. After fourteen days, or when the oil has
well penetrated all the parts of the muscles, remove it and wash it with spir-

Time's Bodies 237


its of wine, and put it in a place where it will dry." To preserve the soft tis-
sues, they were first inflated and injected with lukewarm water so as to
evacuate all the blood; then they were washed out with spirit of wine until
no trace of blood remained, after which they were dried in appropriate
shape and soaked in oil of turpentine. 50 Another former Leiden student,
Stephen Blankaart, also wrote about the use of oil of turpentine for balsam-
ing bodies. 51
Thus, the key ingredient for the Leiden experimenters was oil of turpen-
tine. The turpentine commonly in use at the present day, a product of fir and
pine trees, has little relation to the substance called turpentine or terebinth,
much less its oil, in the seventeenth century. At that time the word applied
only to an exudation of the terebinth tree (now called Pistacia terebinthus, or
Chian turpentine). As John Goodyer explained in his 1655 edition of
Dioscorides, the tree grew in "Arabia Petraea" as well as "Judea and in Syria
& in Cyprus, & in Africa, & in the Islands called Cyclades." He also noted
that "The Resina Terebinthina doth surpass all other rosins."52 Twenty years
later, the English military surgeon James Yonge warned that "there is a base
Turpentine-like substance called commonly Terebinth, brought from
France, drawn from the Fir and other Trees, ... which is no more the gum
of the Turpentine tree, than Tar is."53 This is confirmed by the reports of
major sixteenth-century investigators: "Champier said larch-tree resin was
sold for terebinth but Brasavola reported in the mid-sixteenth century true
terebinth was now imported in round lumps from Cyprus to Venice."54 The
oil of true turpentine, as Yonge explained, itself "contain[s] in it the Bal-
sam."55 Moreover, one definition of balsam itself was "an aromatic oily or
resinous medicinal preparation . .. specifically, of various substances dis-
solved in oil of turpentine."56 The oil or "spirit" of turpentine ("they being
names promiscuously given to one and the same kinds of thing") was
obtained after a slow distillation of the resin of the terebinth tree in a retort,
which produced first a white, then a yellow, and finally a red oil, the last of
which was the best. 57 When the red oil was mixed with blood, Yonge
explained, curious things happened, among which was a coagulation that
made it a very useful styptic for stanching wounds. For chemists of the day,
experiments with oil of turpentine were common. One of the experiments
most often repeated by Robert Boyle was the action of oil of vitriol distilled
in a retort with turpentine, which yielded sulfur. 58
Perhaps it is even significant that "oil of turpentine was regarded as very
similar to spirit of wine,"59 for in England, sometime in the 1650s, appar-
ently at the suggestion of William Croon, Robert Boyle discovered that the
spirit-or oil-of wine (something resembling today's brandy) could be
used to preserve anatomical specimens. He had been so excited by news of
De Bils's invention that he published a translation of one of De Bils's pam-

Harold f. Cook
9.3 A prepared child's arm holding naturalia, clothed in a sleeve with lace
(specimen prepared by Rachel Ruysch, Frederik's daughter). From Ruysch, Opera
omnia. Courtesy of Metamedica, Rijksuniversiteit Leiden.
phlets. In I663 his specimens of "a linnet and a little snake, preserved already
four months, entrails and all, without any change in colour, in some spirit of
wine," were to be found in the Royal Society's repository.60 Whether such
trials were stimulated by news of the efforts of De Bils and others is
unknown. Although the results were slightly imperfect, and the liquid had
to be periodically refreshed, the simplicity of suspending specimens in spirit
of wine in a glass container made it a very important discovery. As one can
tell from some of the methods used by the Dutch, spirit of wine was also
used together with oil of turpentine to produce preserved specimens that
could be handled.
The use of oil of turpentine and other materials for the preservation of
anatomical specimens was rich in unintended consequences, however. For
instance, Swammerdam was the first to see the anatomy of insects as some-
thing other than an almost undifferentiated jelly. He could do so because of
his technique using very fine scissors and tweezers, excellent microscopic
technique, mounting platforms of his own design, and oil of turpentine. The
latter preserved the bodies of insects that could not be dried, but more: oil of
turpentine turned the body fats of insects into a kind of lime, which could be
carefully washed away, leaving their fibrous tissue exposed to the eye. 61
When the young Cosimo de' Medici visited his cabinet in I668, therefore,
Swammerdam famously dissected a caterpillar to show how the wings of the
future butterfly were already contained in the body of the caterpillar. The
demonstration held great importance in showing that metamorphosis was
not an alchemical transformation of one kind of matter into another, but
rather an unfolding of parts already present. The grand duke was so
impressed with the skill and novelty of Swammerdam's work that he
offered him I2,000 guilders for his collection of insects-an enormous
sum - if he would bring it to Florence and enter his service, an offer
Swammerdam declined. 62
Methods for preserving bodies also allowed for the development of other
techniques of anatomical investigation. For instance, working with Van
Horne on 2I January I667 on a human uterus (preserved with oil of turpen-
tine, as Schrader noted above) Swammerdam found means to inject the
uterus with wax-a technique he further developed, together with injec-
tions of air - filling out vessels that could not otherwise be discerned. 63 His
cabinet contained a preparation of the lungs in which the trachea was filled
with white wax even to tiniest parts, the pulmonary artery was filled with
red wax, the pulmonary vein with rose wax, and the small orifices of the
arteria bronchialis with a fire-red substance; he showed a liver similarly dif-
ferentiated in balsam and wax. 64 He was able to show that the human spinal
marrow was composed of fibrous nerves by suddenly placing the yet warm
spinal vertebrae in cold water, leaving them there for twenty-four hours,

Harold]. Cook
and then carefully breaking off the bone to expose the marrow-which
again had turned from an undifferentiated mass into tissues. 65 What had
begun as an attempt to defy time by preserving bodies from the process of
decay had also become an experimental technique crucial to the develop-
ment of new anatomical knowledge.
Frederik Ruysch developed the Leiden methods to a high pitch. A fellow
student of Swammerdam's, Ruysch became perhaps the most innovative
anatomist of the late seventeenth century. His cabinet was an extraordinary
sight, full of embalmed and preserved specimens in lifelike poses and dress,
and strange fish and organs in bottles. The centerpieces of his displays were
his thesaurii: dioramas of tiny human skeletons in poses memorializing the
fleeting world of time by (for instance) playing violins made from hardened
body parts, all standing among woods made from hardened arteries and
veins, and rocks made from bladder stones (fig. 9.4). At the same time that
they drew the viewer's attention to the instabilities of time, the specimens
themselves represented permanence in the face of the forces of decay. When
Tsar Peter visited Ruysch's cabinet, there in a cradle lay an embalmed baby
with glass eyes that looked so lifelike and peaceful that he bent down to kiss
the child. In 1717, the tsar purchased this emblem of the new science, along
with the rest of the specimens in the cabinet, for 30,000 guilders; parts of the
collection remain in St. Petersburg. (Ruysch had sold his collections before,
laboriously building up new ones.)66

CONCLUSIONS

By the mid-seventeenth century, the process of secularization-of investi-


gating the world of time by altering time's previously unalterable move-
ment-was well under way, in the world of science as well as commerce.
Perhaps it is not surprising that the most notable investigators had powerful
ties to the world of business: De Bils came from a merchant family,
Swammerdam from a family of apothecaries located next to the VOC ware-
houses in Amsterdam, Ruysch from civil servants so down on their luck that
he became apprenticed to a pharmacist. Probably all lost more money than
they made from their investigations. They turned their material inheritances
into wonders of nature. Yet in distorting the "natural" actions of time as they
did, De Bils, Swammerdam, and Ruysch also "capitalized" on their work by
turning it into money, or could have.
For naturalists, like merchants, the accumulation and preservation of
things was in anticipation of later demand. For the sake of future genera-
tions as well as immediate curiosity, they undertook the investment in intel-
lectual capital: in specimens and an inventory of details about them (fig 9.5).

Time's Bodies
&+a;re 9.4 One of Ruysch's thesaurii (the third) depicting the shortness of life,
constructed from kidney, gall, and bladder stones, trees made from dried veins and
arteries, topped with fetal skeletons in various poses: a central figure looks heaven-
ward, singing a lament ("Ah fate, bitter fate!") while accompanying itself on the
violin; a small figure to its immediate right conducts the music with a baton set
with minute kidney stones; on the far right is a skeleton girded with sheep
intestines injected with wax, a spear made from a hardened male vas deferens con-
veying a message about man's first hour also being his last; to the left is a figure
with a feather, a symbol of vanitas; and in front is a tiny skeleton holding in its hand
a mayfly-an insect on which Swammerdam had written a famous book, center-
ing his moral argument on the supposed fact that the creature lived in its adult
form for only one day. From Ruysch's Opera omni anatomico-medico-chirurgi
(1721- 1727). Courtesy of Middleton Health Sciences Library, University of Wisconsin, Madison.

Harold 1. Cook
@+3.5 The
cabinet of Bernard
Sigfried Albinus,
who carried on the
tradition of anatomi-
cal preparation and
study at Leiden,
illustrating how a
professor's cabinet
had become replete
with wet specimens
by the 1740% Cour-
tesy of Metamedica,
RijksuniversiteitLeiden.

Material progress and utility became the watchwords of contemporary nat-


uralists even when reveling in curiosities. They wished to create enduring
knowledge from fragile and perishable objects by thorough-going investiga-
tion and reporting, which could be handed down to others. But they, too, as
much as merchants, depended on preserving their accumulated objects. Just
as a warehouse of nutmeg would lose much or even most of its value should
mold take over (hence the dusting of it with lime), so a cabinet of curiosity
lost value as its specimens were lost or destroyed. Investing for the long term
might add value to one's transactions, but it also required a struggle against
the processes of decay and putrefaction. New methods of both capitalism
and science therefore depended on working against the forces of transitory
nature in favor of longer term durability. Given an outlook that valued not
only material bodies themselves but the accumulation and preservation of
them, coupled with methods of investigation linked to thinking about how
to get value from time, ingenious people like De Bils and Swammerdam
added to the store of knowledge, now part of our intellectual capital.

Timei Bodies 243


1. David S. Landes, Revolution in Time: Clocks and the Making of the Modern World (Cam-

bridge, Mass.: Harvard University Press, Belknap, 1983),72-76.


2. On musical horology, see Penelope Gouk, Music, Science, and Natural Magic on Seven-
teenth-Century England (New Haven, Conn.: Yale University Press, 1999),202-204.
3. Norbert Elias, Time: An Essay, trans. by Edmund Jephcott, reprint, 1987 (Oxford:
Blackwell, 1992), II5 . Elias emphasizes the phenomenological approach to understanding
time. For one who argues for the absolute existence of space and time, in which space-time
points "stand in causal relations to one another," see Michael Tooley, Time, Tense, and Causa-
tion (Oxford: Clarendon Press, 1997), quotation on 379. For a collection of essays informed by
modern physical science, see Steven F. Savitt, ed., Time's Arrows Today: Recent Physical and
Philosophical Work on the Direction of Time (Cambridge: Cambridge University Press, 1995).
4. Jan de Vries and Ad van der Woude, The First Modern Economy: Success, Failure, and
Perseverance of the Dutch Economy, 1500-1815 (Cambridge: Cambridge University Press,
1997),691,692.
5. Niels Steensgaard, The Asian Trade Revolution of the Seventeenth Century: The East India
Companies and the Decline of the Caravan Trade (Chicago: University of Chicago Press, 1974),
142- 143, 149·
6. P. W. Klein and J. W. Veluwenkamp, "The Role of the Entrepreneur in the Economic
Expansion of the Dutch Republic," in Economic and Social History of the Netherlands, Het
Nederlandsch Economisch-Historisch Archief, Vol. 4 (Amsterdam: NEHA, 1993),28,31 - 32,
33,49·
7. F. W. T. Hunger, "Bernardus Paludanus (Berent ten Broecke) (1550-1633)," Janus 32
(1928): 36 1.
8. J. A. J. Barge, De oudste inventaris der oudste academische anatomie in Nederland (Leiden:
H. E. Stenfert Kroese's, 1934),34-55.
9. Karen Meier Reeds, Botany in Medieval and Renaissance Universities , Harvard Disserta-
tions in the History of Science (New York: Garland, 1991), esp. 35-36.
10. Faye Getz, "Roger Bacon and Medicine: The Paradox of the Forbidden Fruit and the
Secrets of Long Life," in Roger Bacon and the Sciences: Commemorative Essays, ed. Jeremiah
Hackett (Leiden: Brill, 1997), 337-364.
11. The Cure of Old Age, and Preservation of Youth. By Roger Bacon . .. Translated out of

Latin; with annotations and an account of his life and writings. By Richard Browne (London:
Tho. Flesher and Edward Evets, 1683).
12. Sir Francis Bacon, "De augmentis scientiarum," translated as "Of the Dignity and
Advancement of Learning," in his Works, ed. and trans. James Spedding (London, r860), Vol.
4, book 4, chap. 2,390-394.
13. See, for example, his letter to Chanu of IS June 1646; AT IV:441-442.
14. Swammerdam, The Book of Nature; or, the History of Insects, trans. Thomas Flloyd,
revised with notes by John Hilt (London: C. G. Seyffert, 1758), ii; for the original, see
Swammerdam, Bybel der NatuurelBiblia Naturae, ed. Herman Boerhaave, with facing-page
translation into Latin by Hieronimus David Gaubius (Leiden: Isaak Severinus, Boudewyn
vander Aa, Pieter vander Aa, 1737), sig. B.
15. For what follows, I am heavily indebted to Jan Reinier Jansma, Louis de Bils en de
anatomie van zijn tijd (Hoogeveen: C. Pet, 1919)'
16. " . .. sed fidem superat omnem, exsiccatum hominis Cadaver Recenter Mortuum
Diceres tanto Theatro Dignissimum opus." The wooden plaque containing Van Horne's tes-
timony is reproduced on 47 of Jansma, De Bils . De Bils later claimed that he had spentf
40,000 on the preparations, not including his time (66).

Harold j. Cook
17. On Egyptian mummies in cabinets, see both examples above. The famous work of
Hermes Trismegistus was thought to have been written at the beginning of Egyptian civi-
lization; on the fascination with Egypt, see for example Frances A. Yates, Giordano Bruno and
the Hermetic Tradition, reprint, 1964 (New York: Vintage Books, 1969); Thomas C. Singer,
"Hieroglyphs, Real Characters, and the Idea of Natural Language in English Seventeenth-
Century Thought," Journal of the History of Ideas 50 (1989): 49-70; Anthony Grafton, Defend-
ers of the Text: The Traditions of Scholarship in an Age of Science, 1450-1800 (Cambridge, Mass.:
Harvard University Press, 1991), 145-177.
18. Alfred Lucas, Ancient Egyptian Materials and Industries, 4th ed., revised by J. R. Harris
(London: Edward Arnold, 1962),270-326.
19. See esp. Katharine Park, "The Criminal and the Saintly Body: Autopsy and Dissection
in Renaissance Italy," Renaissance Quarterly 47 (1994): I -33; Katharine Park, "The Life of the
Corpse: Division and Dissection in Late Medieval Europe," Journal of the History of Medicine
50 (1995): I I I - I 32.
20. Jansma, De Bils, 70-74'
21. Karl H. Dannenfeldt, "Egyptian Mumia: The Sixteenth-Century Experience and
Debate," Sixteenth-Century Journal 16 (1985): 163- 180.
22. J.R. Partington, A History of Chemistry, (London: Macmillan, 1961) 2:98.
23. Lynn Thorndike, A History of Magic and Experimental Science (New York: Columbia
University Press, 1923-58),8: 106.
24. Dannenfeldt, "Egyptian Mumia," 169-171.
25. Thorndike, A History of Magic and Experimental Science, 8: 414.
26. Dannenfeldt, "Egyptian Mumia," 173-174; Partington, A History of Chemistry, 2:444-
27. William Eamon, "Cannibalism and Contagion: Framing Syphilis in Counter-Refor-
mation Italy," Early Science and Medicine 3 (1998): 1-31.
28. For example, see Carolus de Maets, Chemia Rationalis (Lugd. Batav.: Jacobum Moc-
quee, 1687), 162- 164, and a manuscript of his chemistry course from 1675 and 1676, British
Library, Sloane MSS 1235, fols. 5-5b.
29. Justus Schrader, Observationes et Historiae (Amsterdam: Abraham Wolfgang, 1674),
236: "Notum est, cad ave rum artus ac viscera sibi relicta necessaria ruere in putredinem,
eorumque compagem nunc citius nunc tardius foedit corruptione dissolvi, nec ullum pristina:
integritatis aut forma: vistigium tandem retinere. Hunc consuetum natura: cursum arte non
solum refra:nari sed & cohiberi posse, dudum evicerunt medicata lEgyptiorum funera bitu-
mine ac pretiosis subinde refinis & aromatibus abunde infarcta."
30. Ibid., 236: "quae tamen quum externam solummodo speciem servent, idque obscure,
non item interiorum habitudinem, merito isti lEgyptiaco operi pra:fertur illa ars, quae cadav-
era & eorum fragmina ita obdurat, ut salva permaneat ipsorum textura, idem supersit color,
eadem conformatio, nullo non tempore ac pro lubitu ab Anatomico contemplanda, & quidem
absque cruoris effusione, aut fastidioso madore, quibus delicatiores offendi, & ab inspiciendis
demortuorum visceribus communiter arceri solent."
31. Jansma, De Bils, 48-53.
32. G. A. Lindeboom, Dutch Medical Biography: A Biographical Dictionary of Dutch Physi-
cians and Surgeons 1475-1975 (Amsterdam: Rodopi, 1984).
33. Jansma, De Bils, 53-54·
34. Lindeboom, Dutch Medical Biography, col. 909.
35. Jansma, De Bils, 54-58.
36. Ibid., 58-67.
37· Ibid., 65, 67, 68- 69.
38. Ibid., n 78-79, 83- 88, 90.
39· Ibid., 96-99.

Time's Bodies
40. Ibid., 67·
4I. Partington, A History of Chemistry, 2:20S.
42. Lindeboom,Dutch Medical Biography, 103I.
43. N. F. J. Eloy, Dictionnaire hlstorique de la medecine ancienne et moderne: Ou memoires
disposes en ordre alphabetique pour servir a l'histoire de cette science (Mons: H. Hoyois, 177S),
655- 656.
44. Jean Nicolas Gannal and R. Harlan, trans. and eds., History of Embalming, and of
Preparations in Anatomy. Pathology, and Natural History; Including an Account of a New Process
for Embalming (Philadelphia: Judah Dobson, IS40), 91 - 92,96; see Gabriel Clauder, Methodus
Balsamandi CO/pora Humana (lena: Oan Bielckium, 1679), chap. 5, sec. 3, 12S- 140, on his
view of the method of De Bils, and chap. 6, 140-ISI, for his own method.
45. Reported by G. A. Lindeboom, ed. and comp., Het Cabinet van Jan Swammerdam
(1637-1680) (Amsterdam: Rodopi, 19So), xii, from a letter ofOlaus Borch to Bartholin (which I
have not yet seen), which places the event around 1661-62, when Borch was in the Netherlands.
46. Schrader, Observationes et Historiae, 237: "Paretur itaque vas stanneum corpori pra:-
parando quoad capacitatem apte respond ens, huic immittatur & duorum digitorum a fundo
distantia probe firmetur craticula lignea minutis foraminibus constans, super quam corpus
collocetur, mox oleum terebinthina: infundatur ad trium digitorum eminentiam, & vas leviter
ac minus arcte opertum per justum temporis inter valium in quiete servetur: Sic penetrantis-
simum istud oleum, cadaver is, cui circumfusum est, poris paulatim sese insinuabit, & aquo-
sum laticem, pra:cipuam fermantationis ad corruptelam tendentis causam, extrudet, qui vi
ponderis descend ens, & per craticulam stillans, spatium inter ipsam & fundum progressu tem-
poris occupabit; interea vera simul subtilior balsami portio, ob minus perfecte clausum vas,
exhalabit, qua magis magisque evanescente, tandem corpus concreta olei amurca tan quam
gummosa medulla penitus imbutum duritiem acquiret, ac idcirca posthac extra liquor em in
aperto aere incorruptum absque situ aut tineis perennare facile poterit."
47· Ibid., 23S,
48. Ibid., 23S-240'
49. Sloane MSS 1235: "Collegium Chymicum Secretum / A/D. Carolo de Maes apud
Lugdunenses," 1675 and 1676: f.5, "Modus Condiendi Cadavera."
50. De Maets, Chemia Rationalis, 162-163.
5I. Stephan Blankaart, Neue und besondere Manier aile verstorbene Carper mit wenig
Ukosten del" Gestalt zu Balsamiren (Hannover und Wolffenbiittel: Gottlieb Heinrich Grentz,
1690)' lowe this reference to Tomomi Kinukawa.
52. Dioscorides, The Greek Herbal of Dioscorides, ed. Robert R. Gunther, reprint, 1934
(New York: 1959),49.
53. James Yonge, Currus Trzumphalis (London: Printed for J. Martin, Printer to the Royal
Society, at the Bell in St. Paul's Churchyard, 1679),5°.
54· Partington, A History of Chemistry, Vol. 2,97·
55. Yonge, preface to Currus Triumphalzs.
56. Oxford English Dictionary.
57. Yonge, Currus Triumphalis, 4S-50' Also see William Davisson, Philosophia Pyrotech-
nica (Paris: Joan Bessin, 164°),325 - 326 and William Davisson, Le Cours de Chymie (Amiens:
Michel du Neuf-Germain, 1675), 30S.
5S, Partington,A History of Chemistry, Vol. 2, 494.
59. Ibid., 267, citing Libavius, Alchemia, 1597, bk. 21, tract. ii, c. 36.
60. L. de Bils, "Large Act of Anatomy" (1659), in The Works of Robert Boyle, ed. Michael
Hunter and Edward B. Davis (London: Pickering and Chatto, 1999), Vol. I; D. H. Tompsett
and Cecil Wakeley, J. Dobson, historical intro, Anatomical Techniques (Edinburgh and Lon-
don: E. & S. Livingstone, 1956), x.

Harold f. Cook
61. Jan Swammerdam, Bybel der NatuureiBiblia Naturae, ed. Herman Boerhaave, Latin
translation by Hieronimus David Gaubius (Leiden: Isaak Severinus, Boudewyn vander Aa,
Pieter vander Aa, 1737), sig. 1.
62. G. A. Lindeboom, ed. and comp., Ontmoeting met Jan Swammerdam, Ontmoetingen
Met Mystici, no. 3 (Kampen: Uitgeversmaatschappij J.H. Kok, 1980), 12.
63. Swammerdam, Bybel der natuur: "Hier was it, in Van Hornes huys, op de 21 Januar-
ius, 1667, dat hy de eerste reys, medwasch opvulde de vaten des Lyvmoeders van eene Vrouw,
door een seer nutte onderneminge, dien hy daar naa heeft verbeterd meer, en meer." [sig C]
"Hy verder oeffende vlytig een bysondere konstgreep, door welke hy de deelen der lighaa-
men suyver rynigde van ai, wat daar in was; hier naa blies hy die op, dat sy vollucht waren,
droogde die dan; waar door die styv geworden, haare gedaante behielden, en door die konst
naauwkeurig kosten beschouwd werden, jaa ook net beschreven. Eene uytvinding waarlyk
van de uyterste nuttigheid."[C2]
64. Lindeboom, Het Cabinet van Jan Swammerdam, xvii.
6S. Swammerdam, Bybel, sig. C: "het ruggemerg, nag warm, med de wervelbeenen, waar
in her bevat is, ten spoedigsten moet gelegd in koud water, en 24 uuren daar in gelaten; waar
na de wervels omsigtig gebroken moeten werden; dan werd dit alles so gesien."
66. Antonie M. Luyendijk-Elshout, "Death Enlightened: A Study of Frederik Ruysch,"
Journal o/the American Medical Association 212, no. 1 (1970): 121-126; Julie V. Hansen, "Res-
urrecting Death: Anatomical Art in the Cabinet of Dr. Frederik Ruysch," Art Bulletin 78
(1996): 663- 679.

Time's Bodies
Cartography, Entrepreneurialism,
and Power in the Reign of Louis XIV
The Case of the Canal du Midi

CHANDRA MUKERJI

I n seventeenth-century France, it was quite normal for infrastructural


projects like roads, drainage systems, canals, and bridges to be carried
out not through direct state action, but rather through political funding
of contracts with entrepreneurs. Such structures were often deemed essen-
tial by the central government, and imposed as a duty on regional govern-
ments, but they were constructed by local engineers and laborers under the
supervision of a financier, functioning as entrepreneur. Commerce and
political power were allied with technical skills for material effect.
Since such engineering brought together political, technical, and com-
mercial powers to rework the landscape for politicoeconomic effect, repre-
senting nature-in this case the countryside-was instrumental to the
process. Infrastructural work was, almost by definition, a product of politi-
cal geography. Places where improvements might be plausibly tried were
identified with surveys, and the results of these projects were recorded in
representations that connected the work to larger schemes of territorial
integration and communication. The local political bodies that were given
some responsibility for these projects also commissioned surveys to assess
the feasibility of the work, address any traditional claims to the land in
question, and evaluate the potential usefulness of the results. And the entre-
preneurs who risked their capital in these endeavors used models and maps
to design and promote their constructions, paying particular attention to the
specificities of place that would affect the costs of the work (such as natural
resources available as construction materials). Their engineers, in turn, used
measures of elevation, distance, soil quality, and topography to choose
building strategies. Some combination of commercial calculation and repre-
sentation was at the heart, then, of these infrastructural efforts, and com-
merce itself was furthered through the cultivation and deployment of
representational techniques.
How this kind of engineering depended upon entrepreneurialism and
imagery in late seventeenth-century France is apparent in the case of the
Canal du Midi, which was built in the 1660s to the 1680s in southwest
France, running from Toulouse to the Mediterranean. It was an interesting
project because it was so much more ambitious in scope than most infra-
structural efforts, but it also shared so many features with them. It was as
though a common system for knowing and acting on nature was blown up
in scale so its social and technical contours were made more visible.
In principle, the canal was meant to link the Mediterranean Sea to the
Atlantic Ocean. At Toulouse, the canal approached the Garonne River,
which discharges into the Atlantic Ocean, and near Beziers the canal was to
reach the sea. The structure would not only link two regions, but also two
vast sea-based trading systems, making the canal huge not only in length but
in possibility. More pressingly, however, such an ambitious project was tech-
nically difficult to realize. It had to cross a major watershed to link valleys
with water draining in opposite directions toward two seas. Flooding the
highest point of a canal in this region (particularly in the dry summers) was
a difficult task-one that many saw as impossible-because opening locks
to move boats would also discharge vast amounts of water. Even the port for
the canal on the Mediterranean was a problem. There was no natural harbor
to use, since the coast along the Mediterranean in the region tended to be
flat, full of salt marshes, and easily silted up. Worse, that region of the sea
was plagued by devastating storms that easily destroyed ports or filled them
with run-off sand. Making a harbor substantial and deep enough to accom-
modate trade was not easy. In fact, it was not satisfactorily done in the
period. But despite of the difficulties, a canal was built, and commerce along
it followed. State-based entrepreneurialism, and representations of nature
yielded a new nature-a second nature-a work of "genius" that was a
tribute to and improvement upon the land of France.
We can get a better sense of how geographical representation and com-
mercial culture met at the Canal du Midi first by studying the canal as an
economic enterprise, and then turning to its development as part of a layered
system of representations. With this background, we can consider more sys-
tematically links between the two.

THE CANAL DU MIDI AS A COMMERCIAL ENTERPRISE

The Canal du Midi (in roughly the form it was built) was proposed to Jean-
Baptiste Colbert, controller general of finance, by Pierre-Paul Riquet, a salt
tax (gabelle) collector from Languedoc. He was an odd man to take on this
task. He was neither an engineer nor a scientist. He was not even an experi-

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


enced entrepreneur who had built large numbers of roads and bridges
already for the state. He had, of course, a fortune and his own businesses, but
nothing on a scale to match this canal. In the humanist language that has
mainly typified him, he was a simple man of vision whose genius was recog-
nized by Colbert and manifested in the canal's successful completion. More
to the point, he was a money man who knew how to use finances to deploy
labor power and natural resources for economic effect.
There are a number of political questions that need including in any
account of the canal's success. Why did Riquet think that a powerful minister
like Colbert would authorize and help him finance such a vast enterprise?
Why would a tax collector (a despised social type) from a region known for
its dissidents and tax revolts become capable of retaining a loyal labor force to
realize his dreams? How could a salt tax collector acquire the capacity to
locate a water supply adequate for the canal's watershed area, when so many
others had failed? And how did a man with no engineering background
imagine he could build a port to serve sailors and fishermen? Most of all, how
could a man like Riquet find solutions to technical problems that had evaded
the more demonstrable genius Leonardo da Vinci, who had been asked to
plan a canal for this region a century before? The mythology of the heroic
Riquet that haunts this bit of history raises as many questions as it resolves.
But the fact of the canal's engineering remains, and so does the role of Riquet
in making the venture succeed. As unlikely as the story might be, a not-so-
simple regional tax man whose French was not so good did indeed propose
and bring into being a canal that ran from Toulouse to the Mediterranean-
not quite before his death but shortly afterward.
There were, as one might expect, indeed some good reasons for Riquet's
eventual success. He could propose this project-although he lacked con-
nections at the French court and engineering experience-precisely because
he was a tax collector and financier. Colbert had been encouraging those
with capital, such as tax men, to use their wealth to invest in infrastructural
work.! They could enrich themselves while serving the state, and he would
give them special privileges or revenue to help them. Riquet had a reliable
source of income already because he was an homme de gabelle. Colbert sim-
ply had to give him permission to increase taxes and use that income to
finance the project. At the moment he proposed the canal project, Riquet
had even recently obtained a new territory to tax. The Treaty of the Pyre-
nees made Rousillon part of France, and Riquet was one of the few men sent
to raise revenue there. Unfortunately for him, however, since Catalonia had
no traditional salt tax, violence erupted in Rousillon when it was imposed,
leaving Riquet angry and with very little additional revenue. His attention
to the dissidents in the region became a sore point with Colbert when the
minister thought Riquet was spending too much time in the Catalan city of

Chandra Mukerji
Perpignan and away from the canal. Still, when the project was first pro-
posed at Versailles, Riquet's position as a tax farmer (and proven skills at
raising money) made him a more attractive candidate for this commission. 2
Colbert, of course, did not simply share costs with Riquet, using treasury
funds. He exercised his political muscle to extract financial contributions
from local political authorities, mainly the Etats de Languedoc. He used also
the power of the state to acquire the land for the canal, set down principles
for assessing the value of the properties, and force local authorities to help
with the financing of their acquisition. The minister additionally signed
edicts setting price limits on construction materials and their transport to the
canal; he ordered the roads in bordering towns to be improved; and he gave
Riquet mining rights in nearby mountains - presumably to make him his
own supplier of iron for the locks. Later in the process, he also (and more
reluctantly, against local opposition) supported the sale of (lucrative) offices
related to the canal's administration, which gave Riquet a new revenue
stream for the project. 3 Colbert even authorized the imposition of a new tax
on public houses, inns, and bars in the region (perhaps because they were
profiting so much from the workers), and required nearby towns to house
the workers at local expense. 4
These schemes for financing the canal's construction were both Colbert's
ways of orchestrating the state's participation in the work, and the tax man's
ideas about how to make his investment work. Just as much as Colbert tried
to control Riquet with his favors, the entrepreneur extracted from his patron
means for financing and managing the project. Riquet knew the limits of
state influence in his region, and let the minister know when Colbert needed
to enforce his edicts or extend his list of required contributions from local
authorities. The tax man also was the one to notice when suppliers were
price gouging, and he asked for legal relief from these practices. The assid-
uousness with which Riquet attended to his financial interests may have
raised Colbert's suspicions about the tax man's true interest in engineering.
But the record suggests that both men used their own forms of financial
experience to make the project work. Riquet (an extractor of local revenues
and creator of economic opportunities) was skilled in recognizing where
money could be found, natural resources exploited, and labor power put to
work for politicoeconomic advantage. Colbert was good at recognizing
political and economic opportunities for the state, and using political incen-
tives to promote them and reduce their risks to the treasury and the reputa-
tion of the king. Both men needed all their wits to keep up with the rising
costs and risks of the project, and despite their mutual distrust they managed
to make their alliance work. s
For dealing with local elites and resources, Riquet had family connections
and associations forged through his work to keep his authority intact against

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


the powerful opponents who tried to block his efforts. Local hostility to the
project was predictable. Confiscating tracts of estate properties from land-
holders and demanding large sums of regional tax revenues on top of this-
particularly to finance a canal that would (in principle) yield personal
economic gain-was unlikely to appeal to stakeholders in the region's eco-
nomic arrangements. But Riquet had connections in this group and was
schooled (if not always skilled) in local politics. His father had been a mem-
ber of the Etats de Languedoc-even when an earlier proposal for a com-
parable canal had been evaluated and rejected by the local authorities. He
was aware of the foot-dragging that was endemic to fund raising for state-
sanctioned projects from the Etats. But he was also trained in making peo-
ple give up the money they owed him - against their will. The elites of
Toulouse, Carcassone, Montpellier, and Beziers may have been more visible
and powerful than the persons who usually owed him tax money, but once
Riquet had Colbert behind him, he had the political connections that could
(at least at times) intimidate them. The tax man also cultivated a loyal cohort
of supporters for his scheme among these powerful men. He recruited as
investors many leading politicians, financiers, landowners, and entrepre-
neurs, making loyal advocates for his engineering scheme from this group of
local elites. 6 These were allies he desperately needed and too frequently
alienated as the work progressed and as the list of his local enemies grew.
No skill in political maneuvering was of any value to Riquet, however,
until he first persuaded Colbert to endorse the project, and this was no sim-
ple task. But he had a trump card. He lived in a land known for its tradition
of religious heresy, tax revolts, and antagonism toward the central French
state. French troops had been so frequently called into Languedoc (since the
Wars of Religion in the sixteenth century) that the region was better mapped
(from military surveys) than most other parts of France. The region was
mainly peaceful in midcentury, but the nobility in southwestern France of
the seventeenth century remained Huguenot, and Protestant ranks seemed
to be increasing in size and power. In this context, Riquet chose as local
patron for the canal project (the man he hoped would bring his proposal to
the attention of Colbert) d ' Aglure de Bourlemont, who was about to become
the archbishop of Toulouse. This respected cleric agreed to inspect the plans
and engineering mock-ups for a canal already set up by Riquet at his estate
at Bonrepos. He was impressed enough with what he saw to take the pro-
posal to the very Catholic court at Versailles, and to the minister himself.?
Colbert had other than religious reasons to be predisposed toward this tax
man and the proposal introduced to him by d'Aglure de Bourlemont-
politicoeconomic ones. The minister was already strategically placing new
commercial ventures in dissident regions to create a permanent state pres-
ence, and Languedoc was on his list of sites with economic assets but need-

Chandra Mukerji
ing better control by the crown. Riquet was a good agent for Languedoc. He
had financial skills, knew the land from firsthand experience, was not a dis-
sident, and was more than willing to function as an informant about the
activities of locals that he deemed threatening to his interests and state
power. Important, too, was the fact that he was willing to risk his fortune in
a commercial project of imposing scale that might indeed make his fortune
but could also serve the regime. If the canal were built, the propaganda value
of the structure alone would be enormous. Early in the project, the canal
even seemed to have strategic military appeal. The great military engineer
Vauban argued that the waterway could be made wide and deep enough to
carry military vessels from the Mediterranean to the Atlantic, thereby avoid-
ing pirates by Gibraltar-a great problem for the French navy. Although
this military dream was soon scrapped, the political value of the canal
remained high. s
Infrastructural improvement also had a particular political resonance in
seventeenth century France. The mesnagement tradition of politics in
France-which defined the state as a great estate that needed proper man-
agement 9-had put great emphasis on strategic use of the countryside as a
route to collective wealth and power. By using rational land management
practices, one could (according to this political theory) yield a landscape that
was more Edenic, and that would allow the people in rural areas (rich and
poor alike) to enjoy greater prosperity, increased trade, and more stable social
relations. The canal was easy to identify as just the kind of improvement
needed in the countryside to make it more perfect-a water system that was
less prone to flooding and the strong currents of local rivers and streams.
Linking the Mediterranean to the Atlantic through such a peaceful waterway
was not only a way to increase trade, but-in theory at least-create a bet-
ter political environment. Trade, of course, was meant to be a clear benefit of
the canal. The economies of the Atlantic and Mediterranean areas were so
different that there was every reason to think that demand for goods that
could be moved through the waterway would be strong. The canal's con-
struction in any case made the countryside around the structure more of an
economic asset, and placed the pursuit of trade into the visual field of all those
who lived by it, suggesting new possibilities for commercial activity.
Riquet's strategies for managing the work process also impinged upon
and added a new commercial element to local social arrangements. The
canal was constructed using an innovative and unusually generous wage
labor system. In some sense, the contractual labor force he raised to build the
canal helped to constitute a working class in this area at a very early
moment. There were roughly forty thousand people who participated in the
construction, giving this region a surprisingly extensive set of capitalist social
relations.

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV 253


Riquet paid and treated his workers extremely well because he was a
money man and knew the power of the purse to deliver what he wanted. But
he also needed some way to attract a stable labor force for this vast project. If
locals were inclined to hate the tax man who (at least nominally) directed the
project, they did not object to sharing his wealth. The pay scale at the work-
site was something of a scandal, since common laborers were paid half again
what they would be given for farm work. They also could take sick days off
and were paid for periods of bad weather, when it was physically impossible
to work, and even holidays. This regularity in compensation was practically
unheard of in the period, even in Paris, much less among laborers of Langue-
doc. lO It is true, as Le Roy Ladurie has pointed out, II that the peasants of this
region were already independent farmers rather than serfs in the thirteenth
century, suggesting that there were long-standing modernizing forces affect-
ing labor relations in the region. Still, Riquet's labor contracts were revolu-
tionary, and shocking enough to his contemporaries that they raised
questions about his character and honesty. Local nobles who opposed the
canal apparently suggested that Riquet was only claiming to pay such high
wages so he could actually pocket more money for himself. That is one of the
main reasons Colbert sent a trusted confidant and engineer, La Feuille, down
to check on Riquet, but there was no evidence of impropriety. La Feuille's
reports back to Colbert were, on the contrary, quite clear that many expenses
of the project were legitimately large. Wages were eventually cut, but not to
the level of local agricultural labor.
It should be no surprise that nobles objected to Riquet's labor policies.
Field hands and shepherds from local estates constituted the bulk of the
workforce, and nobles lost power over the laboring poor because of Riquet's
contracts. Moreover, the tax man did achieve surprising loyalty in the work-
force. Even when Riquet raised taxes to supply more revenue for the canal
and there were movements against paying the gabelle, the workers did not
rebel against him. 12 The stability of the workforce meant that the experience
gained from early efforts to build the canal was not lost by exhausted and dis-
affected workers who left to seek employment elsewhere. It was carried over
to the next stage of the project by those who liked the pay scale and work
rules. For a canal that was so complicated to achieve, this kind of continuity
was very valuable; creating a stable workforce was a good investment.
Given the entrepreneurial logic behind the proposal for the canal and the
process of its construction, it is surprising, in the end, that Riquet sought as
his reward for this work not commercial rights to the canal's use, but
domainal rights to the land on which it lay. Riquet demanded (no matter
how inappropriate it seemed to Colbert) to create a landed family of title
through his entrepreneurial skill. In this region of France, where the well-
being of the household (as Le Roy Ladurie has suggested in Montaillou) was

254 Chandra Mukerji


the central cultural value, this probably made sense. Colbert expected social
mores akin to those of the Parisian bourgeois elites, who bought domains
with their profits rather than demanded domains as part of a contract. But
Riquet was not from Paris. Because of his stubbornness on the subject,
Riquet infuriated Colbert, and lost the support from the treasury for the sec-
ond stage of the project, but his household was indeed given as domain a
long, thin stretch ofland that meandered Toulouse to the Mediterranean. 13
In the end, Riquet brought to the engineering of the Canal du Midi not
technical expertise and probably not personal genius, but financial experi-
ence in extracting resources and creating a loyal workforce that could learn
on the job. He was not a thoroughly modern man of finance, but he was an
entrepreneur who was clever with contracts, and he respected the expertise
of others (even laborers). He used the canal as a means for making a com-
mercially more tractable "second nature" whose profitability would better
serve both the interests of his family and the glory of his king.

THE CANAL DU MIDI AND LAND SURVEY METHODS

Cartographic skills, survey measurements, and other means of representing


the landscape were invaluable for this project, but less as a source of accurate
information to shuttle between bureaucrats at Versailles and entrepreneur!
engineers in Languedoc than as tools in an ongoing system for learning
about and solving problems ofland control. I4 The distinction is important.
We often assume that states need information, and that they acquire it by
deploying experts who feed bureaucrats with the kinds of information they
need or want to make administrative decisions. But a project like the Canal
du Midi could not have been developed around the formal knowledge of
geography in the region, and the engineering depended on a continued pat-
tern of problem solving. Subtle characteristics of the topography, soil, water-
sheds, and the like became more apparent as the work progressed, and had
to be taken into consideration. New information about the project was gath-
ered at every step in the project and pointed to problems that had not been
anticipated in the plans. Of course, good data were vital. The more accurate
the information at any stage of the process, the more useful it was to the out-
come. But the most important thing to the project was the capacity to
learn-to engage in ongoing decision making and problem solving. Accu-
rate information about how to build a channel through one part of the land-
scape was useless if the canal was finally routed elsewhere. The project
needed good maps of the area to help route the canal correctly, but it needed
even more the capacity to survey and assess in complex and changing ways
whatever portion of the landscape required engineering attention. IS

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV 255


The primary problem for Colbert (as the political patron for the project)
and Riquet (as the entrepreneur investing his family's wealth and future in
the scheme) was risk. This was an attractive venture that would clearly have
enormous benefits if it could be realized, but the "if' was very large. Projects
of this sort had been proposed for centuries, but they had not been realized
because of technical problems. Certainly Colbert had no interest in under-
mining the reputation of Louis XIV by sponsoring a project of this visibility
that would fail. (The military's later efforts at the Eure River aqueduct
would do just that in the 1680s, but Colbert was much more averse to risk
than was Louvois or Vauban.)16 Riquet, too, had no desire to waste all of his
personal strength and financial assets. But this was not a project that was
guaranteed of success. By all accounts, it was too difficult to achieve. There
was no one person in France who really knew enough about the huge area
that the canal traversed or enough about canal engineering even to say
whether this scheme was practical and whether Riquet's plan was feasible.
Instead, the major actors accepted the risk and took a leap of faith in sup-
porting the canal project. What made them do it? One plausible explanation
is that they had multiple mappings of this region that made them feel they
knew the local countryside and what to do with it. They had canal plans,
road maps, military surveys, maps of water sources, and legends about the
local landscape. This array of formal and informal representational systems
for rendering the countryside not only yielded large amounts of informa-
tion, but a set of perspectives on land which addressed a surprisingly broad
range of human needs and interests.
The growth of humanist geography in the sixteenth and seventeenth
centuries had already provided in western Europe a particularly rich set of
cartographies, emphasizing human material achievements on the land
(cities, wall systems, ports, canals, roads, bridges, monasteries, managed
forests, territorial boundaries, and property lines). These images provided a
wealth of evidence that human beings could indeed rework the landscape
for commercial and political effect. li The problem with these maps as guides
to building the Canal du Midi is that they were developed by different groups
with distinct ways of measuring, recording, and acting on the land. To
reduce the risk of the project and actually create the canal, these distinct
visions and traditions of practice had to be combined. Colbert and Riquet-
for a range of reasons-succeeded in doing this, creating a social learning
system that allowed experts to bring different ideas into the project. The
ironic result was that in this period of so-called state absolutism the solution
to the material problem of territorial control lay not in centralized and
absolute control of the engineering process, but precisely its opposite: the
development of a system of distributed learning that allowed diverse strands
of surveying and engineering to be combined for a common purpose.1 8

Chandra Mukerji
There was a range of skills in geographical measurement available in
France used for representing and acting on the natural world. Measurement
techniques for making elevation studies with precision were taken from
men of the Academie des Sciences; mapmaking repertoires developed by
military engineers for planning battles and building fortresses were used to
manage the canal's incline, build reservoirs, control water intakes, and
design the canal basin and some aspects of its locks; civil surveying tech-
niques from road engineering and property disputes were employed to man-
age the land acquisitions for the canal, help build the bridges, and design the
route; and geographical folklore and traditions of practice from the region
identified places to avoid or use in planning the canal and acquiring
resources for its construction.
Military surveys were the most frequently employed forms of scientific
cartography in France during the early modern period. Maps of coastlines,
cities, strategic canals, and drainage projects, mountainous areas and road-
ways were important strategic tools that took advantage of the measurement
techniques being refined in the period. There are thousands of unsigned
maps made for building fortresses, planning sieges, setting cannon, deploy-
ing troops, re-creating battles, and describing terrain in border regions that
remain obscure testimony to a widespread practice. The maps were a clear
form of political cartography, obsessed with details of the local landscape
that could affect the army's ability to control it. 19
Military cartographers, whether engaged in fortress engineering projects
or planning how to move troops or set up cannon, learned to think primar-
ily topographically. Topographical features of the landscape were natural
barriers and conduits, so they had vital strategic importance. Fortresses
reconfigured the topography artificially, using walls, ditches, and canals to
constitute a new terrain. A set of high bastion walls with a canal between
them was (ideally) an artificial version of a deep gully surrounded by moun-
tains and filled with a daunting river. Army engineers and surveyors who
helped in mounting sieges, tunneling into the battlement walls and rolling
temporary bridges across streams or canals, were in the process also erecting
a countertopography of their own. No wonder military surveyors became
particularly adept at measuring the subtle changes in elevations that gave
character to local regions. 20 A comparable refiguring of the landscape was
precisely what Riquet proposed to deliver with the Canal du Midi, and what
Colbert hoped to bring to fruition when he sent Chevalier de Clerville,
France's leading military engineer, to evaluate and oversee the project in the
name of the state.
Also during the reign of Louis XIV, Colbert stimulated and set apart sci-
entific cartographic work when he established the Academie Royale des Sci-
ences and the Observatoire. The point of these institutions was to promote the

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV 257


sciences in France. To be a center of European civilization-to rival or even
eclipse Italy -France needed to be a leader in the sciences as well as the arts.
A French system of academies, based on Italian precedents but better
funded and organized, seemed the best way to surpass the Italians. 21 Unlike
the military surveyors who worked in a range of locations but still paid
greatest attention to the peculiarities of a particular place (for obvious strate-
gic reasons), the mathematicians of the academy reduced all lands to planar
measurements. The point was to increase the accuracy of simple measures,
not try to account for geographical complexity. This was still valuable to an
engineering project like the Canal du Midi because the canal would work
only if the elevations were accurate enough to create a stable water supply
and build a viable system of locks. That is why the commission that first
inspected the area under the direction of Clerville used academic techniques
for taking elevations to check on Riquet's claims. 22
The least studied of the pertinent survey traditions of the period was the
kind of simple measuring done for plot plans, resource assessments, or civil
engineering projects. This genre of mapmaking became a routine political
tool for policy making under Henri IV, when infrastructural improvements
became important to state policy. Rational land management techniques of
the sort used on individual estates were applied to state policy to promote the
economic well-being of the kingdom. This political approach put great
emphasis on knowing the countryside as a repository of natural resources
and site of potential improvement. Surveying and engineering were closely
aligned as political tools. Although this strategy seemed to be buried along
with the Protestant King Henri IV, the politics of mesnagement was revived
by Colbert in the period of Louis XIV and became part of the territorial pol-
itics used to serve state-based absolutism. This gave civil, forestry, tax, and
estate surveyors new work and social importance. 23
These surveyors did not place their findings in grids of latitude and lon-
gitude like Cassini and Huygens. Like the academicians, the arpenteurs paid
less attention to topography than the military surveyors, but similarly cared
about the specific characteristics of local areas. Like geographes du roi, they
were called upon to address the political status of land holdings, but unlike
all the rest, they acquired skills in resource assessment and worked on civil
engineering projects with local entrepreneurs. They were the ones attentive
to road construction, forest management, estate planning, and hydraulic
engineering-techniques of land improvement. This kind of localism, so
different from that of the military, was obviously essential to the building of
a great canal like the one from Toulouse to the Mediterranean.
The least scientific of local means for representing the land was not a
form of survey at all, but regional narratives describing places. Folklore
might have seemed to have no place in a process of "rationalizing" the land-

Chandra Mukerji
scape through engineering, but that was not the case for the Canal du Midi.
Stories about the countryside as a site of spiritual as well as natural powers
were crucial to understanding where to build a tunnel or how to recognize
the exact position of a watershed. Stories marked sites of natural anomalies
and indicated where human (or superhuman) forces had changed patterns
in nature. This region of France, as part of a pilgrimage route to Cam-
postella in Spain, was particularly rich in sites of miraculous streams, devil-
ish rock formations, and stories of saints and heretics. This region of France
had also been an important part of the Roman Empire, and stories of past
glory made visible the webs of roads, canals, bridges, and burial sites from
this early period that could be used a models for construction of a great
canal. Narratives of place, then, were means of representation that carried
local knowledge about the character of the local countryside, and passed on
understandings of the natural world and past engineering ventures that
were important to the Canal du Midi.

RISK ASSESSMENT AND REPRESENTATION AT THE CANAL DU MIDI

There were two major problems that plagued the project for the Canal du
Midi, which made the risks of trying to build the canal sometimes seem too
great to be worth trying. The first was the alimentation system for flooding
the high point of the canal. If the canal could not be supplied with water, it
could not be built. The other was the port on the Mediterranean. If the canal
could not link trade on the Garonne River to trade on the Mediterranean, it
had less purpose in propaganda value and economic usefulness. The defini-
tion of the canal as the Canal des Deux Mers - the canal of two seas-
depended on finding solutions to these two fundamental problems. But the
project was begun without either of these problems being fundamentally
resolved. The canal was started when the problems seemed soluble, and
Riquet seemed able to solve them-with the kind of help and supervision
that Colbert insisted upon for the work. What made these problems seem no
longer real impediments, but rather practical issues to work through on the
ground, was a demonstration that diverse groups of experts could be
deployed to fashion solutions.
For the first stage of the project, when the water system was in question,
Riquet developed his scheme by assembling his small cadre of experts to
guide him, and Colbert tested his capacity to do the work by setting up a
commission of diverse (and much more powerful) experts to challenge and
refine Riquet's proposal. At the second stage of the project, when the port
and routing of the canal were more pressing, Riquet (already used to but
annoyed with Colbert's continual stream of spies/experts coming from the

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


north) enrolled all those who came to assess the project in thinking through
the problems. He allowed or even encouraged in this period a much more
fluid social arrangement of participants, and gave more autonomy to those
working at different sites. He also took more risks in the project itself as he
became confident that the engineering process would finally yield a canal. In
these ways, the project clearly changed in its second stage. But the result was
still a pattern of distributed problem solving and group learning, using
diverse means of representing land to act on it.

The water system. The first great obstacle to building this canal was design-
ing a water supply system that would flood the high point of the structure,
and the basic system for the Canal du Midi was designed for Riquet with
help from Pierre Campmas, afontainier from Revel, a small mountain town
where Riquet had some land and financial interests. A fontainier's job was to
find water supplies for the town and get them where they were needed. This
work entailed subtle knowledge of local topography, some hydraulic engi-
neering, and experience with seasonal weather patterns. The fontainier had
already worked for Riquet, diverting water to a mill on his property, so
Riquet was aware of his expertise. 24
Campmas was able to reduce the risk of a water shortage for the canal
because he understood so much about the water supplies of the Montagne
Noire. He also knew how to follow the topographical contours of the land-
scape, designing conduits for carrying water safely downhill to where it
could be used. He already knew the rivers on the mountain and the gorges
that could be dammed to collect water during the rainy season. 25 Still,
Campmas was in no position to think about how to design a water supply
system for a large canal. He could deliver water and build supply channels,
but he had no way to know if the supplies would be enough to keep vessels
afloat in the summertime. He could not compute the amount of water nec-
essary for a canal because he had no experience in assessing how many locks
were appropriate for a particular incline. Without an educated guess about
the number of these structures the canal would need, he could never esti-
mate the amount of water the canal would require to stay filled when it was
in use and the locks were dumping water downstream. So his expertise
alone left too much unknown to make the project seem reliable enough to
fund and try.
The adequacy of the water supply could be better estimated, however, with
the expertise of a hydraulic engineer like Riquet's other collaborator, Fran<;:ois
Andreossy. Andreossy had no local knowledge of the watershed, but he was
trained in principles of surveying and engineering. Moreover, he had
recently visited Italy, where he inspected some of the well-known Italian
canals. He could consider the number of locks needed for the Canal du Midi

Chandra Mukerji
and thus the necessary water supply (although he vastly underestimated the
eventual requirements), so he could assess the results of Campmas's efforts.
He was also a fine cartographer, so he could represent the proposed canal and
its relationship to the region in which it would be laid, not only showing its
route but suggesting its fit with the local topography. Thus, he made the
canal project seem conceptually viable and strategically visible in ways
important to gaining the confidence of Colbert and the king. 26
While these three men were individually in no position to think through
the construction of this large canal, together they had the expertise to make
the project seem feasible. Riquet could calculate finances, Campmas could
find and direct water from the Montagne Noire, and Andreossy could
design the structure and define its technical requirements. Still, the three
had little social standing and no obvious authority for making such a grand
proposal. Their solution was to build a mock-up of the canal on Riquet's
estate at Bonrepos to test their engineering designs. They had two little
ponds for a water supply, a set of locks, supply channels, and even a tunnel
to take the water downhill and through the mock canal. If anyone doubted
that these men could actually construct a canal, this was their answer. The
team had already made such a construction - at least on a small scale. It was
a matter of demonstration, not dispute. Riquet showed the model to the
future archbishop d' Aglure de Bourlemont, who then brought Riquet's
plans to the attention of Colbert.27
Colbert was interested in, if not convinced by, Riquet's proposal and sent
Clerville to assemble a commission of experts to study Riquet's plans. Under
Clerville's supervision, this group was to travel through the Montagne Noire
and along the proposed route of the canal, making surveys and assessments of
the engineering proposed by Riquet, and then report back to the minister. 28
The commission had the social authority that Riquet, Campmas, and
Andreossy did not. It included many local notables who had no expertise in
canal construction, but who, like Bourlemont, were necessary to secure
political support for the enterprise. The "experts" on engineering and sur-
veying on the commission were, first, Clerville himself, with his experience
in building fortresses, including canal construction (at least over short dis-
tances). Next there was Henri de Boutheroue de Bourgneuf, whose father
had completed the Canal de Briare, linking the Loire to the Seine,29 and who
himself now managed that canal. As the man with the greatest practical
knowledge about comparable waterways, Bourgneuf was expressly charged
with estimating the amount of work it would require to build Riquet's pro-
ject, and what it would cost. In addition, there was the sieur de La Feuille,
an ingenieur who seemed to have the trust of Colbert, and who was put in
charge of supervising the canal work from 1667 to 1683. The group also
included four niveleurs, two of whom are known: Riquet's colleague

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


Fran\=ois Andreossy, and Jean Cavalier, a geagraphe du roi. Cavalier had
started in the military, but he had become a skilled regional cartographer,
drawing the region's best map, one that was repeatedly copied and repro-
duced for almost a century.30 The other two surveyors, whose names do not
appear elsewhere in cartography, presumably had some local knowledge
and civil survey skills. This meant that the commission contained not only
members of the region's political elite but established experts in the intellec-
tual traditions pertinent to the job. The survey team included a civil engi-
neer/surveyor (Andreossy), a geagraphe du rai (Cavalier), a hydraulic
engineer (Bourgneuf) and two military engineers/surveyors (Clerville and
La Feuille). There was no academician, but the elevations made by Cavalier
were done using the techniques developed by La Hire, one of the most able
surveyors from the academy, so even academic surveying had its effects on
the development of the plan. 31
The members of the commission trained in different traditions of French
surveying and engineering not only tried to anticipate and reduce the risks
of the venture, but also checked the assessments made by the original team.
Riquet's financial expertise was set against Bourgneuf's experience at the
Canal de Briare; Andreossy's designs for the canal and reservoirs were
checked by Clerville and his assistants for their soundness; and Campmas's
water supply plans (as well as the overall canal trace) were scrutinized by
Cavalier, Bourgneuf, and La Feuille to see if the inclines were well com-
puted and the routes topographically well placed. The commission members
did not so much challenge the plan as revise it to make it more effective,
apparently reducing the risk of this daunting but intriguing project. 32
Since the crucial risk facing the commission had to do with whether the
water supply system would be effective in supporting a navigational canal,
delivering adequate supplies when and where they were needed, the com-
mission spent most of its time considering this issue. One important element
in the plan had to be the location of a watershed where the supplies could be
delivered to the canal to flow both toward the Atlantic and toward the
Mediterranean. Although historians have sometimes admired Riquet's skill
in swiftly locating the watershed between the two river systems linked by
the canal, in fact there was little argument about where to do this. There
were known watersheds in the area, the best recognized of which was La
Grave de Naurouze. These were represented less on topographical maps
than in folklore. Riquet and his colleagues simply confirmed with their ele-
vation studies what was already common conjecture among locals. 33 The
problem was not finding the watershed but making a convincing case for its
usefulness for the project. Physical geographers, following Hondius, gener-
ally believed that mountain chains existed between all major river systems.
In fact, many period maps of southwestern France actually depicted a set of

Chandra Mukelji
mountains running through the proposed route of the canal. They were fic-
tional constructs, theoretical assertions, un based on measurements or obser-
vations of any kind. But they had the authority of respected science.34 So,
much of the effort of the survey work was nivellement or studies of elevation
to prove that the canal could in fact be built across the proposed set of valleys
and be supplied with runoff from the Montagne Noire carried to Naurouze.
Given the serious problems of scientific credibility of the project, it should be
no surprise that the surveyors, apparently under the tutelage' of Cavalier, the
geographe du roi, used La Hire's measurement techniques for the commis-
sion's elevation studies. This gave their findings the authority of science, so
that they could be trusted as a basis for countering Hondius. 35
Even with good elevations, however, the plan for the water system was so
complicated that it was not entirely convincing on paper. This is why the
commissioners also asked Riquet to make a channel along the proposed
route from the Sor River (on the Montagne Noire) to the Fontaine de La
Grave a Naurouze. This rigole d'essai was meant to be a small ditch just to
prove the inclines, but it was nonetheless built with some difficulty. Torrents
of rain kept disrupting construction of the rigole, but in October 1665, the
waters arrived as expected at Naurouze, and the demonstration of the ali-
mentation system was complete. 36
The result of the commission's studies on the Montagne Noire and the
test of the water system with the rigole d'essai was not only an engineering
plan for the alimentation system, but also confidence in the canal scheme
itself. The risks entailed in designing a canal that crossed a watershed, which
had impeded the development of such a waterway in this region before, now
seemed small enough to face. Once the commissioners endorsed Riquet's
proposal (with their revisions), Colbert was willing to give out a contract for
at least the first stage of the process. He authorized beginning the canal at
Toulouse, connecting it to the Montagne Noire supplies, and carrying it
(with an impressive set of locks) across the watershed. Completing the work
to the Mediterranean would have to wait until another set of vexing techni-
cal issues were made less daunting. 37

The Port. The port of Cette or Sete, built as the terminus of the Canal des
Deux Mers, was in some ways the exact opposite of the water supply system
on the Montagne Noire. While designing an effective complex of reservoirs
and ditches in the mountains to deliver water to a faraway canal seemed an
obviously difficult engineering task, an outlet for the canal on the Mediter-
ranean appeared relatively straightforward. Building breakwaters and
dredging harbors might not have been simple ventures, but they were at
least more familiar ones. 38 The risk of diminishing the effectiveness of the
canal by giving it no good port at one end seemed small at first, but it loomed

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


larger as the work progressed. To offset the anxiety that resulted from
repeated failures of the seawalls and dredging, more and more information
about the site was accumulated and discussed by different kinds of experts.
The representations of Sete multiplied, providing new hopes for solving
engineering difficulties that seemed intractable. By the time the difficulties at
Cette had become chronic there was no way to reroute or stop work on the
canal. It was already too far advanced. The point was to reduce the risk to the
canal's success that a poor port would create, and to increase the chance of
making the whole enterprise into the flourishing and politically dramatic sys-
tem that had been imagined. Each new layer of representation was accumu-
lated to achieve this end, and gave new reason to revive this hope.
Importantly, thought had already gone into the construction of a new
harbor in Languedoc even before Riquet first sent his original proposal to
Colbert. Clerville had been asked to survey the coast near Beziers to propose
a way of constructing a new harbor. He suggested building one at Cette or
Sete. The engineer was given this task because for both military and eco-
nomic reasons; having no safe harbor in this area was a strategic problem for
France. Colbert was concerned on both grounds because not only was he
finance minister and worried about the French economy, but he was also in
charge of the French navy (such as it was). His success in empowering the
state, pleasing the monarch, and satisfying his own sense of order depended
on making some improvements in this area.3~
The southwestern region of the country was also among the richest in
foodstuffs and manufactures, but poor in transportation. The population (in
this period before the Revocation of the Edict of N antes sent the industrious
Protestants in large numbers to the Netherlands and England) was rela-
tively large and productive. There were textiles and leathers being produced
in the Montagne Noire. There were marbles and wood to be extracted from
the Pyrenees. And wine and other foodstuffs were produced in excess in the
rich inland valleys. All these goods could find new and more profitable mar-
kets if they could be transported along the Mediterranean. 4o But the sea was
full of pirates, the storms in this area frequent and intense, and France was
a country too weak in naval power to support a merchant marine without a
better infrastructure. The lack of safe harbors was simply dangerous. For
these reasons, Colbert had asked Clerville to survey and design a new harbor
somewhere to the west of Marseilles. Near Agde, a large rocky hill rose from
the otherwise flat and sandy coast. This appeared to be the only natural bar-
rier in the region that could be used to protect ships from storms. The hill
was also strategically useful. The promontory provided a site to look out for
enemy vessels coming to attack ships in harbor. On the inland side of the
hill, there was also a large and relatively deep etang or marsh where water
already collected naturally. It seemed possible that dredging this area for a

Chandra Mukerji
harbor and providing a more substantial barrier between the sea and interior
with a well-designed seawall could yield an effective new port. Clerville was
already planning this project when Colbert asked him to head the commis-
sion to study the Canal des Deux Mers. 41
This helps explain why, when the commissioners were reviewing the
plans for the canal route, they recommended constructing a new port on the
Mediterranean just where Clerville had wanted it. It was immediately obvi-
ous to him (and perhaps to Colbert before him) that Riquet's canal could be
connected to Clerville's port and help serve doubly Colbert's plan for stim-
ulating trade in the region. It also seemed possible to Riquet that if he could
construct and have a fundamental economic interest in a vital new French
port, he could become an even richer man and safeguard his family's future
better than he could with the canal alone. In turn, if Colbert could give this
project to Riquet to complete, he would not have to use so much treasury
money or become too reliant on the military engineers for realizing his eco-
nomic policies. (Keep in mind that Colbert and the minister of war, Lou-
vois, were rivals and even perhaps enemies vying for the king's ear and
limited resources.) The project seemed a way to reduce risk for both the
minister and the tax man. The canal would give more commerce to the
port; the port would make the canal a more effective trading route; Riquet
would risk his fortune and reputation; and Colbert would create a whole
new trading system that would be a wonder of the world and an asset to the
French navy.
Unfortunately, dreaming of a new harbor was quite a bit easier than actu-
ally building one along this coast. Silt in local rivers and the storms plaguing
the Mediterranean in this area made the easy discharge of the canal and its
cargo into the Mediterranean more problematic than anyone had thought. 42
Creating a large seawall to enclose the entrance to the harbor was itself a dif-
ficult task, but knowing how to design it was more taxing. It had to be con-
structed so that high seas would not damage ships during storms, vessels
could move easily in and out of the harbor, and sand could be normally dis-
charged along with outflows of water from rivers and streams. Because the
mouths of existing rivers that ran into the sea produced no natural harbors
in this region, there were some fears about the silting up of the new port. But
no one at first realized what a persistent problem this would be for Sete.
The reason that the silting problems were not well anticipated for the
port was that there was no equivalent to Campmas for the harbor. There
were no local harbor masters to consult on design because there were no nat-
ural harbors. In the other parts of France where harbors (mostly on the
Atlantic) had been improved through engineering, the expertise that had
been accumulated in the process was not useful to transfer to Sete. First, the
problems of engineering on the Atlantic were substantially different, and

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


second, these harbors tended to be part natural as well as engineered, which
the port of Sete simply was not.
Riquet grew up in Beziers and was good at imagining the town he would
build at Sete to make it a great depot; he knew how to design a major center
of wealth and power. But he had no obvious solution to the problems of
designing and building the port, and he was spending a great deal of time in
Perpignan rather than Agde. Nonetheless, he became impatient with
Clerville, La Feuille, and others sent by Colbert, who had their own ideas
about the harbor and felt authorized to speak their minds. Colbert was writ-
ing to him as the man responsible for the success of the venture, but he was.
also not letting him make the decisions. The result was another nightmare
of diminished control for Riquet. But it also set up a new system of distrib-
uted cognition for addressing the problems of the port-using different tra-
ditions ofland representation and design. In the end, this situation also gave
Riquet allies when Colbert abandoned him. La Feuille and Morgues could
speak authoritatively about Riquet's honesty and real need for money to
complete the work.
The experts consulted on the port brought a range of skills to the task:
knowledge of mathematical technique (the Jesuit Pere Morgues), Dutch and
Italian traditions of hydraulic engineering (La Feuille), military engineering
(Clerville), and land managementlbuilding techniques used by local survey-
ors and entrepreneurs in the region.
The authority for the work was as layered as the representational tech-
niques used on site. Riquet, of course, was now sure that he was a world-
class engineer, and lost no time in telling Colbert that he should be given
more control of the work. Colbert, who was increasingly concerned about
costs and the losses that resulted from earlier mistakes of measurement and
design on the canal, was not so impressed, but still needed the entrepreneur
to deliver on this project. Clerville, in turn, felt he still had the commission
of building a harbor in Languedoc, and knew more about this kind of oper-
ation. But he was often called elsewhere to work, and this left La Feuille and
Riquet to share primary authority. When the problems with the port were
mounting, La Feuille was sent first to Holland and then Italy by Colbert to
see how more experienced engineers worked with comparable situations. 43
Colbert then insisted that Riquet not proceed on the port until La Feuille
brought the results of these study trips back to Languedoc. This rankled
Riquet, and locked the two men in a struggle for power that deprived both
of them of control.
It is important to notice that unlike the group who worked on the water
system for the Montagne Noire, the men charged with creating Sete
included a military engineer, a school-trained engineer, and a mathemati-
cian, but lacked a man like Campmas. There was no one from the area who

Chandra Mukelji
had routinely worked on local harbors, knew the coast intimately, shared
local narratives of place and used these strands of knowledge for engineer-
ing. The closest equivalent this group had was an odd outsider, Louis de
Froidour.
Froidour was one of the most unlikely contributors to the port design
because he was France's leading forestry surveyor, not an engineer or local
fishing or shipping expert. But what Froidour could provide were a set of
ethnographic techniques that he used to learn insider, local knowledge about
the place where the harbor was being built, and characteristics of the sea that
impinged on design. The first problem in designing Sete was defining what
local fishermen, merchants, and sailors thought would make a good harbor,
and what they understood about the problems in building it. Froidour then
recommended design changes based on what he heard. 44
Interestingly, Froidour seemed to have been called to the canal site origi-
nally not to study the port but to look at the water system in the Montagne
Noire, where he wrote enthusiastically about the great dam being built at St.
Ferreol, and the ingenious system of ditches constructed to carry water to
the Seuil de Naurouze. Froidour was not particularly informed about the
region or even its forests. He had until this point in his career been busy
mainly in the north inventorying the timber reserves there. He had just been
called to the south by Colbert to map the much more dramatically depleted
forests of this drier region. He was the forestry surveyor whom Colbert
trusted the most, but not because he was such a mathematically adept sur-
veyor or knew the forests personally. It was because he was deemed an hon-
est man who sought accurate knowledge more than bribes. His trip to the
Montagne Noire made sense as one more instance of surveillance orches-
trated by Colber, both resulting from and underscoring the minister's inse-
curity about the water system. Froidour was a good spy for Colbert to assess
the technical work because the forestry bureau was actually the office of
water and forests. When he measured France's forests, this often meant
Froidour was studying mountainous areas in which large timber trees grew
and rivers ran downhill that could carry logs to towns. Additionally, as an
"honete homme," Froidour was also a good person to check for signs of
impropriety. A large part of his forestry "surveying" was really interviewing
people about where old forests had gone. If there was nothing to survey
where a forest had been, or an old forest was much smaller in the surveys
than it appeared to be in old maps, Froidour was charged to determine who
had done the cutting, what taxes they owed on the profits, and what fines
they should pay if they cut trees for which they had no authority. He was,
then, someone whose job was in part to locate and document improprieties
and punish transgressors. These were fine credentials for a spy sent to the
Montagne Noire. 45

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


While it made sense for him to inspect the water supply system, Froidour
was not someone who had any apparent experience in building ports. But his
skills as a kind of ethnographer-someone who could acquire local knowl-
edge about places and their uses - made him the next best thing to Camp-
mas. If he had no firsthand acquaintance with local understandings of the
coast, the problems with storms there, fishing practices, uses of and seasonal
changes in the marshes, and the effects of the rivers discharging into the sea,
he was at least experienced in taking oral testimony from local populations
(proces verbaux). Froidour had learned how to ask pertinent questions that
would reveal patterns (good and bad) of resource use. 46
Froidour interviewed sailors and fishermen about the sea, its character,
and uses in the region. He talked to them about why the rivers there had
dug no natural harbors into the sea. He learned in what direction the winds
blew during the fiercest storms to make sure the port would be designed to
provide appropriate protection for vessels in the harbor. An early storm
proved the potential protection of the seawall, but later ones were more
destructive, raising questions of design that Froidour could address.
Through his efforts, he brought more of the local, folk (and oral) represen-
tations of the area to this part of the canal's design and construction. 47 Unfor-
tunately, no permanent solution was found to the problem of silting in the
port. No matter where the seawalls were placed and replaced, the harbor
soon collected sand. But each new representation raised hope for a solution
as the area near Agde became known in increasingly complex ways.
The insistent nature of the silting led to more careful consideration of silt
accumulation in the canal itself, which was contributing to the load of par-
ticulate matter that was settling into the harbor. A number of small rivers
and streams in the area between Carcassone and the Mediterranean had
been tapped to provide water for the canal in this region, but these water
courses carried from the steep nearby mountains runoff that was filled with
mud and sand. At first, the water was sent directly into the canal, but this
was almost immediately revealed to be a poor plan. The powerful floods that
periodically plagued the region damaged the canal walls as surges of mud
and debris crashed into the fragile ditch. The flood-driven materials either
floated on the surface of the water (sticks and small logs) where it damaged
vessels, or sank to the bottom of the canal (silt) where it clogged the water-
way and required dredging. To solve the problems, these feeder streams
were now either (if they were small enough) routed over slightly elevated
stone containers at the side of the canal that held back debris and mud, or (if
they were larger rivers) redirected into aqueducts, running either over or
under the canal. Conduits (with doors) from the diverted rivers were
designed to allow some of the water from these sources to enter the canal
when it was needed. Drains were also built into the side of the main chan-

Chandra Mukerji
nel, particularly near these inputs, not only for expelling excess water that
threatened to make the canal overflow in wet periods, but also to wash out
some of the silt. Along most of the sides of the canal, ditches were dug par-
allel to the main channel to capture general runoff and keep excess water
from the canal. 48 All these efforts were meant to reduce the myriad silting
and flooding problems that plagued the canal, but particularly the Port de
Sete. Nonetheless, storms ravaged it, and silt collected without cease. New
soundings, land surveys, and harbor plans continued to be drawn, and the
experts from the different traditions of representational practice and engi-
neering experience were called in to help-without stable success.
The lack of a reliable port made trade down the whole length of the canal
less appealing than it might have been. The high cost of paying tolls through
the locks also contributed to keeping long-distance trade low. But the canal
as a whole functioned well by the end of the I680s, and short-distance trade
through it was lively from the start. Commercial dreams and cartographic
skills not only proliferated representations of the landscape in the area, but
made a new landscape in the southwest that was identified with its canal.
This in turn became part of the imagery of the region and of France.

THE CANAL DU MIDI

This story tells us a great deal about how cartographic learning and systems
for representing the landscape were mobilized in the end of the seventeenth
century for commercial and political purposes. They constituted cognitive
tools that allowed French entrepreneurs, engineers, and laborers to achieve
a project that was technically beyond the means of the period. Representa-
tional techniques from land surveying to ethnographic work were mobi-
lized for the project, and provided means for reworking the landscape for
advantage.
The engineering of the Canal du Midi or Deux Mers constitutes a good
science studies story about social learning, scientific expertise, and state
power. But it is also a story about entrepreneurialism in late seventeenth-
century France. The entrepreneurs who built the infrastructure for the state
in this period absorbed a great deal of the risk involved in the massive social
changes that we call in retrospect the growth of the modern state or the
development of state absolutism. They helped to rework the landscape in
ways that transformed places into states, and made them something politi-
cally and economically new. They sought and were given economic benefits
from their efforts. They took economic risks for economic gain. But they
were interested less in taking risks than reducing them; they wanted to do
things they knew they could do. Since they did not have the personal exper-

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV


tise for such assessments, what kind of "knowing" did they rely upon to
make good choices in their schemes? An answer perhaps lies in the under-
studied connections between representations of nature and commercial
activity. If someone could see what to do to promote trade and make a map
or plan for it, there was reason to invest in it. When Riquet had a map of the
Montagne Noire and models to demonstrate his alimentary system, it was
time to start building a canal. And so they did, and reworked this region of
the southwestern France to make it a tribute to human ingenuity and terri-
torial power.

1. See, for example, Marie-Joelle Paris, Versailles: Le Grand Aquedue de Bue au de la


maniere de eonduire les eaux au pare. (Bue: Office municipal des associations de Buc, 1986),
47-5 8.
2. A. Marcet-Juncosa, " L'opposition catalane a PP Riquet" in Ed. Jean-Denis Bergasse,
Le Canal du Midi, Vol. 3: Des siecles d'aventure humaine (Millau: Maury, 1984), 143-150. On 15
December 1662, Riquet wrote Colbert from Rousillon describing his ambitions for this pro-
ject.
3. Archives du Canal (AC) Liasse 46, piece I-"Edit creant un droit annuel sur les
cabaretiers (hotelleries, cabartets, tavernes et marchands de vin) dan toute l'entendue du
Languedoc; piece 2-arret du Conseil d'Etat ordonnant une imposition de 40,000 livres sur
les contribuables aux tailles de la Generalitc de Montauban en remmplacement du droit
annuel sur les cabaretiers." 19 October 1671.
The local resistance to the sale of offices is illustrated in exchanges ofletters between Col-
bert and Riquet mentioning the opposition he was facing from the elites in Toulouse, and
reluctance to pay from Montpellier.
AC Riquet to Colbert 20 August 1670 mentions that the "scindics de Languedoc" are
opposed to the sale of:

offices des greffiers consulaires et de preudhommes. Mais comme ... je vous envoyeray un
arrest du conseil pour en ynterdire un, j'ay lieu de croise qu'apres cet example vous trou-
verez toutes les facilites que vous pouvex desirer a en tirer les sommes qui vous doivent
revenir de la ventes desdits offices.
Je parleray au sieur de Bersan des advances que vous souhaitex qu'il vous fasse sur les
gages des greffiers consulaires et preudhommes, encore que M. de Senes me mande que
vous avez touche les trois quarts de revenu des deux annee 1669-1670, et si je puis l'obliger
a vous fournir quelque somme considerables je Ie feray voloniers.

AC Riquet to Colbert 1-10-7: The problems at Montpellier seem to have been less opposi-
tion to the sale of offices than routine non-payment of the fees for the sales.
4. For the opposition, see, AC Liasse 46. In AC Liasse 548 piece 2, there is an order from
the intendant enjoining the communities near the canal's construction site to repair the roads
to facilitate transport to these areas. 23 Mai 1669. AC, Liasse 191 describes the acquisition of
lands. This is a listing of the sales of lands for the canal, which names the seller, the quality of
land, the size of the parcel, and the price of sale. The land is valued by what is grown on it.

Chandra Muke1ji
So, wheat fields are valued less than orchards and gardens. There is no mention of water
rights in these early records, although there are some buildings included on the lands where
the canal is to go. This is particularly the case in the area of Toulouse where the canal enters
a populated area.
5· AC Liasses 35, 37, 39, 45, 46, 48, 19I. It is important to keep in mind that although the
gabelle in Rousillon was new, it was not given to Riquet just to finance the project. In fact, he
was given this area to "farm" before he first proposed the canal to Colbert, and it was because
of his success as a tax farmer that he was in a position to write to Colbert about his scheme on
IS December I6(J2.
6. The kinds of connections he had through the gabelle are visible in the list of those who
invested both small and large amounts in the project. These were people with multiple ties to
Riquet and politicians of many statuses. See Pierre Burlats-Brun and Jean-Denis
Bergasse,"L'Oligarchie Gabeliere, Soutien Financier de Riquet" in Le Canal du Midi, Vol. 4:
Grands Moments et Grands Sites. 125-141.
7. L. T. C. Rolt, Le Canal Entre Deux Mers (Paris: Euromapping, 1994),23-24.
8. For indication of the value placed on this project in this period and before, see AC
Liasse I.
9. For a discussion of the mesnagement tradition, see Chandra Mukerji, Territorial Ambi-
tions and the Gardens ofVeYJallles (Cambridge: Cambridge University Press, I997), 41-42, 45;
Thierry Mariage, L'Univers de Le Nostre (Bruxelles: Pierre Mardaga, 1990), 43. For more
details of this political tradition and garden design see also Chandra Mukerji, "Bourgeois
Culture and French Gardening in the 16th and I7th Centuries," paper presented at Dumb-
arton Oaks 1998, and to be published in Michel Conan (ed.), Bourgeois and Cultural Encoun-
ters in Garden Art, I550-1850 Dumbarton Oaks 200I. For the relationship between this
political philosophy and the Canal du Midi, see Chandra Mukerji, "The Modern State as
Material Accomplishment: Territorial Culture and the Canal du Midi," paper presented at
Bad Homburg 2000.
ro. For a discussion of the work on the canal including the contract, wages, and working
conditions, see Bertrand Gabolde, "Les Ouvriers du Chantier" in Le Canal du Midi, vol. 4,
235-239, and Andre Maistre, Le Canal des Deux-Mers: Canal Royal du Languedoc 1666-1810
(Toulouse: Editions Privat, I99 8), 72-77.
I I. Emmanuel Le Roy Ladurie, Montaillou: The Promised Land of Error. New York: G.
Braziller, I978.
I2. Rolt, Le Canal Entre Deux Mers, 72.
I3. Maistre, Le Canal des Deux-Mers, 95- I I I.
I4. Bruno Latour, "Visualization and Cognition: Thinking with Eyes and Hands," in
Knowledge and Society: Studies in the Sociology of Culture Past and Present, ed. H. Kuklick and
E. Landfed (Greenwich, Conn.: JAI Press, I986) 6, I-40.
IS. This distinction between information and learning is crucial for understanding state
power and the deployment of experts. Experts often work with partial knowledge and are
asked for advice on which they can give only preliminary judgments. Nonetheless, states
make their policies on these assessments, and they have, over many centuries, managed to
maintain a certain level of legitimacy. Having good information helps, but it is not always
possible to get. What is more important to states is the development and cultivation oflearn-
ing systems -like the research branch of DOD. See Chandra Mukerji, A Fragile Power: Sci-
entists and the State (Princeton, N.J.: Princeton University Press, I989)·
I6. Paul Bondois, Deux Ingenieurs au Sihle du LouiJ XIV Vauban et Riquet (Paris:
Librairie Picard, n.d.); Andre Corvisier, Louvois (Paris: Fayard, 1983). for the risks of trying
to build a canal in this region along with the desire to do it, see AC Liasse I.
I7. Fran<;;ois de Dainville, La Geographie des Humanzstes (Geneve: Slatkin Reprints, I969)'

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV 27 I


18. I have argued elsewhere that a system of distributed learning developed at this site
and that it worked as a means of problem solving for the engineering of the canal. See
Chandra Mukerji, "Distributed Cognition and the Canal du Midi," paper presented at the
1997 annual meeting of the American Sociological Association. I don't want to spend time
here belaboring the point. I will not elaborate either the concept of distributed cognition, or
how it should or should not be applied to this history. That is complex issue that requires a
long argument in itself. For work in this tradition of analysis, see Edwin Hutchins, Cogni-
tion in the Wild (Cambridge: MIT Press, 1995). See also Philip Agre, Computation and
Human Experience (New York: Cambridge University Press), 1997; Yjro Engestrom and
David Middleton, Cognition and Communication at Work New York: (Cambridge Univer-
sity Press, 1996).
19. Anne Blanchard, Les Ingenieurs du Roy de Louis XlVii Louis XVI (Montpellier: Uni-
versite Paul-Valery, 1979),42-54. See also Chandra Mukerji, "Engineering and French For-
mal Gardens in the Age of Louis XIV," paper presented at the University of Pennsylvania
symposium "New Approaches to French Garden History," 1998. Unfortunately for the army,
topographical mapping was a technically difficult discipline. The survey work itself was
physically and cognitively taxing. It required attention to two dimensions of measurement
(elevation as well as distance), and drew surveyors to work in rugged landscapes, trying to
join lines of sight in environments in which it was often hard to see. There were drawing
problems as well; contour lines had not yet been developed so there were no good standard-
ized conventions for rendering elevations accurately. Still, military cartographers became
skilled at noting localized changes in topography and came to know the landscape in these
terms along France's borders. P. D. A. Harvey, The History of Topographical Maps (London:
Thames and Hudson, 1980).
20. For material techniques and military engineering, see Alain Manesson Mallet, Les
Travaux de Mars . .. (Amsterdam: Henri Desbordes, 1696); John Muller, The Attack and
Defense of Fortified Places. In Three Parts (London: T & J. Egerton, 1791), particularly plates 9
and 11; M. Belidor, Les science des ingenieurs dans fa conduite des travaux de fortification.et d'ar-
chitecture civile. (Paris: Claude Jombert, 1729); Nicolas de Fer, Introduccion a fa Fortification
dedie a Monseigneur Ie duc de BOUl"{{ogne (Paris: Chez l'auteur dans l'Isle du Palais sur Ie Quay
de l'Orioge a la Sphere Royale. avec. priv du Roy, n.d.); Pierre Rocolle, 2000 ans de fortifica-
tion fran,aise, tome I. (Limoges and Paris: Charles-Lavauzelle, 1973), 175-212; Sebastien Le
Prestre Vauban, De f'attaque de de la difense des places (La Haye: Chez Pierre de Hondt, 1736).
The importance of surveyors and the military engineers to the politics of the period was sym-
bolic as well as practical. Military engineering had been a hallmark of ancient Rome. "Mon-
uments to the greatness of Rome" were celebrated by Louis XIV and his contemporaries, and
stirred them to use French soldiers for comparable work. The army engineers were first
employed for the obvious jobs of building of ports, garrisons, and arsenals, but they were also
used to improve the water supplies for Versailles and Paris, and in setting out drainage and
flood control ditches around rivers and swamps, and laying out canals that flowed where
rivers did not. All these projects required some survey work, and all of them showed up again
on maps. Tantalizingly, these efforts reshaped precisely what was recorded on maps-the
shape of the shoreline, the course of rivers, the topography, and the roads crossing the land-
scape. See Mukerji, "Engineering and French Formal Gardens in the Age of Louis XIV"
1998; John A. Lynn, Giant of the Grand Siecle: The French Army 1610-1715 (Cambridge: Cam-
bridge University Press, 1997); and JosefKonvitz, Cartography in France, 1660-1848 (Chicago:
University of Chicago Press, 1987).
21. Frances Yates, The French Academies of the 16th Century (Warburg Institute: Univer-

sity of London, 1947); Institute de France, Academie des Sciences: Troisieme Centenaire
1666-1966 (Paris: Gauthier-Villars, 1967), ch. 1.

Chandra Mukerji
22. The canal would work only if the segments met and the elevations worked. Alice
Stroup, A Company of Scientists: Botany, Patronage, and Community in the Seventeenth-Century
Parisian Royal Academy of Sciences (Berkeley: University of California Press, 1990), ch.1.
23. Mukerji, Territorial Ambitions 1997. For a discussion of this political approach and its
effects on the region of Languedoc, see Maistre, Le Canal des Deux-Men.
24. Bertrand Gabolde, "Revel: Des Eaux du Sor a la Rigole de la Plaine," Le Canal bu
Midi, vol. 4, 241-244; Fran.;:ois Gazelle, "Riquet et les Eaux de la Montange Noire," in Le
Canal du Midi, 145-147; Malavialle, "Une Excursion dans la Montagne Noire," Societe
Languedocienne de Geographie Bulletin, part 2, 135.
25. Maistre, Le Canal des Deux-Mers, 72; Gazelle, "Riquet et les Eaux."
26. For the hydraulic expertise in Italy in the period, see Gazelle, "Riquet et les Eaux," vol.
4,147- 150; Malavialle, 120-121. For Andreossy's interest in Riquet and Languedoc, see Jean
Robert and Jean-Denis Bergasse, "L'Etrange Destin des Andreossy," Le Canal du Midi, vol. 3,
199- 201.
27. Rolt, Le Canal Entre Deux Mers, 24-27; Ines Murat, "Les Rapports de Colbert et de
Riquet: Mefiance pour un homme ou pour un systeme?" in Le Canal du Midi, vol. 3
(Cessenon: J.-D. Bergasse, 1984), 108. For a discussion of demonstration and representation,
see Mukerji, Territorial Ambitions 1997, ch. 7; Chandra Mukerji and Patrick Carroll, "Mater-
ial Culture Methods and Historical Sociology," paper presented at the1996 annual meeting of
the American Sociological Association.
28. Rolt, Le Canal Entre Deux Mers, 30-31; Maistre, Les Canal des Deux-Mers, 38-41;
Murat, "Les Rapports de Colbert et de Riquet," I II - I 12.
29. Hubert Pinsseau, "Du Canal de Briare au Canal des Deux Mers: Origines et Con-
sequences d'un Syseme inedit de Navigation Artificielle" in Le Canal du Midi, vol. 4, 27-54.
30. For a discussion of Cavalier's work, see Fran.;:ois Dainville, Cartes anciennes du Langue-
doc, XVIe-XVIIle slkles (Montpellier: Societe languedocienne de geographie, 1961,38-40);
Robert and Bergasse, "L'Etrange Destin," 203.
31. Rolt, Le Canal Entre Deux Mers, 31. Dainville, Cartes anciennes du Languedoc,
XVIe-XVIlIe sieeles, 55, 60-61; M.L. Malavialle, "Une Excursion dans la Montagne Noire,"
Societe Languedocienne de Geographie Bulletin, part 3, tome 15,283-314- The first map pub-
lished of the canal plan and its water supply was actually made by the geographe du roi, P. du
Val. See Malavialle, part 4, tome 15,436-439,475-476.
32. Rolt, Le Canal Entre Deux Mers, p.Some of the changes they initiated had to do with
the route of the canal itself. Under their influence, Riquet abandoned the idea of making the
Fesquel and Aude Rivers navigable, and instead agreed to dig a separate canal in these river
beds to connect his new canal with the ancient Canal de la Robine, which could then carry
boats to the Mediterranean. The reason for this shift was probably Bourgneufs acquaintance
with the Canal de Briare. He knew about the tie-ups in shipping that resulted from problems
in navigating these wild rivers. Trade would be easier in a more contained set of canals.
33. Gazelle, "Riquet et les Eaux," 145-146; M. L. Malavialle, "Une Excursion dans la
Montagne Noire," Part I. Societe Languedocienne de Geographie Bulletin, tome XIV, 1891,
280-284.
34. Henri Enjalbert, "Les Hardinesses de Riquet: Donnees Geomorphologiques de la
Region que Traverse Ie Canal du Midi," Le Canal du Midi, vol. 4,129-142.
35. Dainville, Cartes anciennes du Languedoc, XVle-XVIIle sieeles, 47. The idea of using the
Sor River as a source of the water supply for a canal between the Garonne and Mediterranean
had been discussed by a geographe du roi, Pierre Petit in 1663. See Malavialle, "Une Excursion
dans la Montagne Noire," Societe Languedocienne de Geographie Bulletin, part I, 273 ff.
36. Rolt, Le Canal Entre Deux Mers, 35-37; Froidour, Lettre Ii Monsieur Barrillon Damon-
court, Conseiller du Roy en ses Conseils, Maitre des Requestes Ordinaire de son Hostel, Intendant

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV 273


de Iustice, Police et Finances en Picardie, contenant la Relation & la description des Travaux qui se
sont en Languedoc, pour la communication des deux mas, 9 - 10; Gazelle, "Riquet et les Eaux,"
162-164; Dainville, Cartes anciennes du Languedoc, XVle-XVlIIe sieeles, 1961,47; Malavialle,
"Une Excursion dans la Montagne Noire," Societe Languedocienne de Geographie Bulletin,
part 2, 146. Is this the route that Clerville opposed, according to Malavialle (part I, p. 259)1
Although Riquet's project was vindicated by the study, his plan was not adopted without
revision. While the commission was checking surveys and Riquet was preparing the rigole
d'essai, the main proposed rigole was rerouted . We have no direct evidence about the socio-
cognitive context for the switch. Clerville was the first to make note of it in a plan for Colbert,
but the new channel seemed more the work of a local surveyor than a fortress designer; it fol-
lowed the contours of the land more exactly, requiring a longer course but fewer of the costly
tunnels and less complex terracing that military engineers tended to build. Still, Campmas
was not part of the survey team; Andreossy was Riquet's expert there; and we have no knowl-
edge of the backgrounds or contributions of the two unknown surveyors in the party. On
the other hand, the canal engineer, Bourgneurf, was on the mountain at the time, and he
tended to favor simple systems that were less prone to break down. Whatever the dynam-
ics that led to it, all we know is that the shift occurred when the commission's survey team
was working on the mountain, checking elevations and routes for the waterworks; any or
all of the experts may have contributed to this change of plans. The resulting rigole fol-
lowed a path between the original one developed by Campmas, Andreossy and Riquet,
and the suggestions being made for revision by the commissioners. In this way, it was
apparently (even from the scanty evidence we ha ve of its design) a socially negotiated solu-
tion to a technical problem in the water supply. The alimentary system was changed more
dramatically and explicitly by a redesign of the reservoirs on the Montagne Noire. Riquet
originally proposed a set of small reservoirs to capture and hold the water, but Clerville
thought it would be simpler to construct one great holding facility at Saint-Ferreol. There
was a rock base at the end of this high valley where the Laudot River ran; a huge dam
could be built on such a solid base and provide, Clerville surmised, all the needs of the
canal. The great dam erected at Saint Ferreol was a majestic piece of military engineering,
designed for strength using three distinct stone walls filled between with compacted dirt.
This was a common pattern of construction at the fortresses built by French military engi-
neers. The tailoring of the design to local material conditions was also typical of military
engineering in the period, and so it should be no surprise that Clerville would have mad e
such a proposal. The dam, however, was not just a creature of the military. It contained
elements of hydraulic engineering that were also advanced for the period. There was a
complex set of sluices for different purposes. Some were near the top for letting out water
for use in the canal; at the bottom was a door to drain the reservoir for periodic cleaning,
and to let out the inevitable buildup of silt on the floor of the valley. There was a diversion
channel, too, for carrying off excess water or for diverting the river when the reservoir was
being cleaned or repaired. The great dam, then, was constructed using principles from
both civil and military engineering. The holding system originally proposed by Riquet,
with its complex set of small reservoirs, would have resulted in less reliable dams built on
more unstable surfaces; the Saint-Ferreol plan was in this wayan improvement. But in
fact the one reservoir was not adequate to its task. Ironically, the great fortress designer
Vauban, after he had replaced Clerville as France's top military engineer, was brought in
to correct faults in the system, and recommended the construction of additional small
reservoirs precisely where Riquet first wanted them. Once again a combination of civil
and military engineering-surveying was used to improve the alimentary system. See also
Gazelle, "Riquet et les Eaux," Le Canal du Midi, vol. 4, 155-158, 169; Froidour, Lettre Ii

274 Chandra Muke1ji


Monsieur Barrillon Damoncourt, Conseiller du Roy en ses Conseils, MaItre des Requestes Ordi-
naire de son Hostel, Intendant de Iustice, Police et Finances en Picardie, contenant la Relation
& la description des Travaux qui se sont en Languedoc, pour la communication des deux mers,
16-31; Malavialle, "Une Excursion dans la Montagne Noire," Societe Languedocienne de
Geographie Bulletin, part I, 266-272.
37. Details of the rigole d'essai are contained in AC Liasse 2 .
38. For some of the literature that started to appear in the late seventeenth and eighteenth
centuries on hydraulic engineering, see, for example, Bouillet, Traite des Moyens de rendre les
Rivieres Navigables avec plusieurs desseins de jetties. ... Ouvrage tres-utile Ii tous les Ingenieurs, &
Ii tous ceux qui semelent de Bdtimens & de Machines. (Paris: Chez EstienneMichallet, 1693);
Belidor, Architecture Hydraulique seconde partie qui comprend rArt de diriger les eaux des la Mer
& des Rivieres Ii ravantage de la difense des places, du Commerce & de l'Agriculture, (Paris: Jom-
bart, 1756). For the work done on ports in the period, see Josef Konvitz, Cities and the Sea.
Baltimore: Johns Hopkins Press, 1978.
39. Alain Degage, "Le Port de Sete: Proue Mediterraneenne du Canal de Riquet," in Le
Canal du Midi, vol. 4, 265-285.
40. For the riches of the region, and their importance in the discussions of the canal in
Toulouse, see AC Liasse , piece 14- Avis a Messieurs les Capitouls de la Ville de Toulouse et
reponse par lean de Nivelle, ancien Capitaine Chassvants du Canal. 1667, T

Pour rendre ce Canal Royal encore plus glorieux & donner une tres grande commodite au
commerce qui se fit au Royaume d'Espagne pour porter les laines de France, & de faire
tranporter les marchandies & den trees de France & les Royaumes la, faire porter sur la riv-
iere de Garonne depuis sa source, les marbres precieux de toutes sortes de couleurs, &
jaspres de plus beaux qui soient en l'Europe, les pierres a taille pout bastir, Ie bois a con-
struire les maisons & a faire des Vaisseaux qui faire porter sur la riviere de Lariege, Ie fer &
Ie jayer tire des Montagnes de Foix desdites Pirenees ...... seroit necessaires de faire un Canal
depuis la riviere de Garonne au dessus de moulin Chasteau, d'environ deux ou 300 toises de
pong pour joindre Ie Canal Royal, pour lequel il ne faudroit consruire qu'une seule ecluse
contre la Riviere pour recevoir I'eau necessaire pour porter les Batteaux au Canal Royal.

41. For a discussion of the political economy of the region during this period, see Maistre,
Les Canals des Deux-Mers, 15 - 33.
42. Rolt, Le Canal Entre Deux Mers, 35, 49, 76-77; Alain Degage, "Le Port de Sete: Proue
Mediterraneenne du Canal de Riquet," Le Canal du Midi, 265 - 285. See also Froidour, 35-37;
Froidour, Lettre Ii Monsieur Barrillon Damoncourt, Conseiller du Roy en ses Conseils, MaItre des
Requestes Ordinaire de son Hostel, Intendant de Justice, Police et Finances en Picardie, contenant
la Relation & la description des Travaux qui se sont en Languedoc, pour la communication des deux
mers, 48-72 who also participated in the discussions about designing Sete. He brought from
his forestry experience skills in civil surveying and the politics of engineering the landscape.
Dainville, "Cartes anciennes du Languedoc, XVle-XVIIIe siecies (Montpellier: Societe
languedocienne de geographie, 1961) 56-62.
43. AC 10-10-70- Letter from Colbert to Riquet about sending La Feuille to Holland.
44. Louis de Froidour, Lettre Ii Monsieur Barrillon Damoncourt, Conseiller du Roy en ses
Conseils, Maitre des Requestes Ordinaire de son Hostel, Intendant de Iustice, Police et Finances en
Picardie, con tenant la Relation & la description des Travaux qui se sont en Languedoc, pour la
communication des deux mers. (Toulouse: Chez Dominique Camusat, 1672).
45. M. Deveze, "Une Admirable Reforme Administrative: La Grande Reformation des
Forets Royales SOliS Colbert (1662- 1680)," in Annales de L'Ecole Nationale des Eaux et Foreets

Cartography, Entrepreneurialism, and Power in the Reign of Louis XIV 275


et de la Station de Recherches et Experiences. (Nancy: Ecole Nationale des Eaux et Forets, 1962);
Andre Corvol, L'Homme et l'Arbre sous l'Ancien Regime (Paris: Economica, 1984).
46. Froidour, Lettre d Monsieur Barrillon Damoncourt, Conseiller du Roy en ses Conseils,
Maitre des Requestes Ordinaire de son Hostel, Intendant de lustice, Police et Finances en Picardie,
contenant la Relation & la description des Travaux qui se son( en Languedoc, pour la communica-
tion des deux mers, 1672.
47. Ibid.
48. The rebuilding of the canal is described in AC Liasse 16- Travaux d'amelioration.
These documents are mostly describing work by Niquet, but some was by Vauban.

Chandra Mukerji
'Cornelius Meijer inventor et fecit'
On the Representation of Science in Late Seventeenth-Century Rome

KLAAS VAN BERKEL

A mong the many explanations of the so-called Scientific Revolution,


one of the more attractive relates to the temporary lowering of the
social divide between the technical expertise of the craftsmen and
the theoretical knowledge of the scholars. In several countries in Europe,
architects, navigators, craftsmen, and surgeons contributed both to the con-
struction of scientific knowledge and to the introduction of new methods
and instruments. Italian engineers and architects led the way in the fif-
teenth century, but even as late as the last quarter of the seventeenth century
an ordinary Dutch merchant like Antoni van Leeuwenhoek was able to
stupefy the Royal Society of London with his most detailed microscopic
observations of "small animals."
The contribution of these engineers, merchants, and craftsmen to the
new science was not restricted to unconventional ideas, unorthodox meth-
ods, or newly invented instruments. A vital aspect of modern science is
also the introduction of new ways of representing nature and science.
Whereas mathematical sciences in principle needed no more than some
crudely drawn diagrams and figures, the experimental sciences depended
heavily on the skills of artists who could represent the newly discovered
worlds in a way that was convincing for those who were not present. The
credibility of the new experimental science not only required statements
of trustworthy eyewitnesses and elaborate verbal descriptions, but precise,
lifelike, and attractive visual representations. Precise technical drawings
were not enough (not yet, at least); seemingly irrelevant ornamental
details and a lifelike setting of the configurations were just as important.
In a sense, precisely these irrelevant aspects of the representation were the
most essential because more than anything else they generated the illusion
of lifelikeness that mattered so much to the representatives of the new sci-
ence. Therefore, the craftsman who was also an artist could be of utmost
importance for scholars and scientists who wanted to promote experimen-
tal natural philosophy.l
A little-known, but intriguing example of the craftsman who also oper-
ated as an artist and cooperated with practitioners of experimental science is
offered by a near contemporary of Leeuwenhoek, a Dutchman called Cor-
nelis Meijer. He was born the son of a humble wheel maker in Amsterdam,
but in Rome he rose to the the position of a distinguished member of the
Accademia Fisicomatematica, the most important local scientific society. He
was regarded as a successful engineer and an expert in astronomy, but he
was also valued for his contacts with painters and artists and, more to the
point, for his own artistic talents. In the 1680s and 1690 he published lavishly
illustrated books in which he documented his own inventions, and elabo-
rated on some of the experimental designs of the Accademia Fisicomatem-
atica. As a would-be astronomer he may not have left his mark on
seventeenth-century science and the scientific circles he associated himself
with may not have become as famous as the Royal Society or the Academie
des Sciences, but still his career is instructive for the value even rather con-
servative scholars in a place like Rome attached to a lifelike representation of
their experimental investigations.

THE RISE OF CORNELIS MEIJER

Cornelis Meijer is an unknown figure in the history of science, but in the his-
tory of art the experts are not unfamiliar with his name. Actually, the first
historian ever to devote some serious attention to Meijer was an historian of
art, the assistant director and future director of the Dutch Historical Institute
at Rome, G. ]. Hoogewerff. His 1920 article on Meijer in the art historical
journal Oud-Holland still is the essential point of departure for all research on
Meijer. 2 For art historians, Meijer's claim to fame was his relation to the
famous painter Casper van Wittel, well-known for his vedute, his views of
Rome and its surroundings. Van Wittel went to Rome just after Meijer had
arrived there and was hired by Meijer to execute the drawings for the report
Meijer had to write concerning his investigations into the navigability of the
Tiber. There is some discussion about the relative share of Meijer and Van
Wittel in drawing the illustrations in this report, of which several copies and
versions exist. l Some claim that Meijer can be held responsible only for the
cruder drawings, the more sophisticated ones being ascribed to Van Witte!'
But the technical details in even the most refined drawings are unmistakenly
inspired by Meijer (who also hired other artists to work out his drafts), and,
considering his involvement in the representation of science in the context of
the Accademia Fisicomatematica somewhat later, it is quite likely that his

Klaas van Berkel


share in the cooperation with Van Wittel was more important than most art
historians are prepared to admit.
Cornelis Meijer, born in Amsterdam in 1629, belonged to the Lutheran
community in Amsterdam, a fact that might explain why later in his life he
so smoothly changed religion and turned into a pious Roman Catholic. (The
history of the Lutheran community in Amsterdam is full of reconversions to
Roman Catholicism.) Not much is known of his early life, but he seems to
have been an ambitious craftsman who applied for patents for new technical
equipment by the early 1670s. He apparently moved in semiscientific circles,
discussing hydraulics with university scholars and participating in the
research of Jan Swammerdam. If we are to believe what he told his Roman
audience much later, he even acquired enough riches in Holland to collect a
large cabinet of curiosities, containing for instance quite a number of pre-
cious stones; indeed there is some archival evidence that Meijer's claim is
correct. On his departure for Italy, he had his cabinet taken care of by some
of his relatives. 4
In 1674 Meijer left Amsterdam and went to Venice, where he tried to sell
his technical expertise to the Republican government. Why he did so is not
quite clear. In Rome, Meijer told his friends that he had come to Rome in the
Holy Year 1675 to obtain the indulgences the Church had promised to those
who repented their sins, converted to Catholicism, and visited a specified
number of churches in Rome. But Meijer, while still in Amsterdam, had
printed a leaflet that showed all kinds of hydraulic constructions and is
explicitly addressed (in Dutch, that is!) to the Venetian government. s There-
fore it seems more likely that Meijer's main intention was to go to Venice to
make a profit out of his technical expertise and that the excursion to Rome
(where he would stay for the rest of his life) was just a side trip.
Venice had been a longtime ally of the Dutch Republic in its resistance
against Spanish tyranny and popish imperialism. Since the city was con-
fronted with the same problems as the Dutch Republic-rivers and harbors
that were silting up, low-lying farmland that had to be drained, and so
forth-Dutch engineers found ample employment in Venice. 6 Meijer suc-
ceeded quite rapidly in his plans. Some of his proposals were tried and
adopted, and Meijer was put in charge of all the operations. He also obtained
the official title of engineer, a title he valued even more than the monetary
gains to be made in Venice.
However, in April 1675, before Meijer even had started to execute his
plans, he left Venice for Rome. He promised the Venetian government to
return as soon as possible in order to supervise the clearing of the harbor, but
in Rome he found new opportunities for his engineering skills and in the
end never returned to Venice. In Rome, Meijer became involved in a com-
plicated project regarding the defense of the well-known Strada (or Via)

'Cornelius Metjer inventor et fecit' 279


Flaminia against the Tiber. 7 North of Rome, the meandering river threat-
ened to undermine the road that led straight to the Porta del Popolo, which
most pilgrims took as they entered the Eternal City. Clement IX (1667-69)
had ordered the best engineers and architects in Italy to devise a plan to res-
cue the threatened Via Flaminia. A number of them presented their plans to
the cardinals in charge of the project, and the competition was won by the
young Roman architect Carlo Fontana, a pupil of Bernini. But the pope
died, and the new pope, Clement X (1670- 1676), hesitated; according to his
advisors, Fontana's plan was too expensive. Then, as Fontana and the cardi-
nals were still discussing the details of the project, Meijer came along and
suggested, through the Venetian ambassador to the Holy See, a completely
different construction to check the river. Since Meijer's plan was indeed less
expensive than Fontana's, the pope decided to put Meijer in charge of the res-
cue operation, passing over Fontana. In March r676 Meijer began the work
by first removing some obstacles from the river bed and then driving a large
row of piles in the river, in this way deflecting its current, which from then
on no longer threatened to undermine the Via Flaminia (see fig. I I.1). With
this row of piles, in Italian a passonata, Meijer managed to do what a number
of renowned Italian architects had not been able to do. All of a sudden, he
was a well-known figure in the world of Roman architects and engineers. 8
Not long after the beginning of the construction of his passonata, the pope
also asked Meijer to concern himself with another complicated problem for
which Italian engineers had been unable to find a workable solution.
Clement X was very eager to make the Tiber suited for navigation, since this
would greatly stimulate trade and commerce in the papal dominions. Now
that Meijer had proved his abilities, he of course seemed to be the perfect
candidate for writing a report and making proposals to improve the naviga-
bility of the Tiber. Although Meijer was eager to return to Venice and per-
haps to Holland after finishing his commission in Venice, he was more or
less forced to stay in Rome and to accept the pope's orders. During the con-
struction of the passonata, Meijer had spent-as was common in those
days-some of his own fortune to pay workmen and buy materials neces-
sary for the construction. Of course the pope had promised to pay for all the
expenditures and to reward him with an additional and considerable
amount of money, but Meijer soon discovered that unlike the Venetian gov-
ernment, which had paid him on the spot, Roman officials were very slow in
paying their bills. Only by accepting a second commission could Meijer hope
to regain what he had paid out of his own pocket for the passonata. Perhaps
that was the way papal officials strengthened their ties with their clients
without paying them what they were entitled to.
Within a few months, Meijer had traveled all the way to Perugia, had
seen what obstacles there were and had devised plans to overcome all these

Klaas van Berkel


&* //.1 A view of Rome including Meijer'spassonata in the Tiber. From Cor-
nelis Meijer, Delinationi con discorsi delle Reparationi (Rome, 1670). Copyright
Amsterdam University Library.

difficulties. With the help of Caspar van Wittel and some other artists, he
composed an extensive report on the project and offered it to the two cardi-
nals who were in charge. The first version was written in Dutch and Italian,
a second one in Italian only. Meijer did not speak o r write Italian fluently
enough to be able to write the report himself, so an assistant had to translate
the Dutch text into Italian.9 But before the report was finished, Clement X
died, and a new pope was elected: Innocent XI (1676-89). With the new
pope new clients, including architects, engineers, and artists, came to Rome,
so Meijer had to do all his best not to fall into disfavor with the new pope
and his courtiers. In this respect he succeeded, but his plans were not exe-
cuted, and it is almost certain that despite his pleas and requests, he never
was reimbursed for all the expenditures made during his travels to and from
Perugia. This continual effort to regain some of his money kept extending
his stay in Rome beyond what he had anticipated; his intention to stay in
Rome for a longer period of time is also illustrated by the fact that about this
time he sent for his wife, and that she left Amsterdam for Rome.
The construction of the passonata had turned the foreigner from Holland
into a public figure in Rome. Perhaps his fame was also enhanced by the bit-
ter dispute that erupted once the passonata was under construction. Techni-
cally thepassonata was a success, but it also made him a number of influential

'Corneliuc Me+ inventor et fecit' 28I


enemies, who tried to block his further career in Roman society. In a sense,
Meijer's career in Rome after r678 was a constant fight to defend the pas-
sonata and his own reputation against detractors and envious critics.
The most important of these enemies was the architect Meijer had dis-
placed in the project that made him famous: Carlo Fontana, distantly related
to the famous sixteenth-century Roman architect Domenico Fontana and
one of the most promising pupils of the great Bernini. It must have been a
severe blow to the ambitious Fontana that a total stranger, a Dutchman who
could not even speak Italian, was commissioned to do-and did, apparently
successfully-what Fontana had very much wanted to do. From the start of
the operations, he and others (including other pupils of Bernini) tried to pre-
vent the execution of Meijer's plans. They complained about his materials,
the constructions, the amount of money it would cost. Several times during
the construction of the passonata, the work had to be stopped so the papal
administrators could do some investigations. Every time, Meijer was proved
right and his critics proved wrong.
The result of what proved to be only the first round in this conflict between
Meijer and Fontana was the publication, in r679, of a small book on the con-
struction of the passonata. Actually, it is not a book, but rather looks like a
portfolio collection of some engravings representing Meijer's work on the pas-
sonata. It is dedicated to the pope and has an introduction in which Meijer tells
his readers why he published his inventions, but there is no title or title page.
The most complete copy I was able to consult includes two broadsheets con-
cerning two completely unrelated plans, first to drain the Pontine marshes
and second to dig a new canal between Rome and the sea in order to avoid the
silted-up mouth of the Tiber. In his introduction, Meijer makes it clear that
his "book" is meant as a simple and true record of what had been accom-
plished in building the passonata, by no m eans glorifying the man who had
devised it. "By presenting this to the public eye, I do not pretend," Meijer said,
"to acquire the reputation of a learned and scientifically trained person"
("d'acquistare nome d'addottrinata, 0 scientificata persona").l0 Nevertheless,
this booklet was an effective means to defend his reputation and his passonata.
The booklet was only the first move in a long struggle with Fontana. Evi-
dently, Meijer did also worry about his work on the navigability of the Tiber,
fearing that others might publish his inventions without duly acknowledg-
ing his part in the project. He therefore decided to write a more extensive
report on his engineering skills, including both his construction of the pas-
sonata and the solutions for the problems encountered by ships traveling on
the Tiber between Rome and Perugia. The first edition of this book, titled
L'arte di restituire aRoma la tralasciata navigatione del suo Tevere, was pub-
lished in r683 by the printing office of the Camera Apostolica. A second edi-
tion, or so it seems, was published in r 685.11

Klaas van Berkel


Actually, there is some reason to believe that the second edition was not a
second edition at all, but just the completed version of a book that was pub-
lished partially in 1683. On the title page of the "first" edition, the Camera
Apostolica is mentioned as the printing office, but at the end of the book a
private publisher is mentioned, Lazzaro Varese. Varese is also mentioned as
the printer and publisher of the "second" edition, but it is more likely that
during the publishing of the book Meijer changed publishers and that when
the book was complete in 1685, he had the publisher print a new title page
with the name of the new printer. This seems to indicate that Meijer had
some urgent reason not to wait until the book was complete, but to publish
the first part upon its completion. And indeed he had every reason to speed
up the publication, because in 1683 Fontana once again went public with an
attack on Meijer's passonata. After 1678-79, Fontana had become silent on
the passonata. Apparently Meijer had done a fine job, and the destruction of
the banks of the Tiber no longer threatened the Via Flaminia. 12 Behind the
passonata some of the ground once lost to the river was reclaimed, making
further losses improbable. In early 1683, however, Fontana and his followers
charged Meijer once again with having used the wrong material and with
having driven the piles too shallowly into the bottom of the river and so on.
According to Fontana, it was just a matter of weeks or months before the pas-
sonata would collapse, giving free reign to the river again and eventually
destroying the Via Flaminia. The pope was alarmed and ordered a careful
investigation into the firmness of the passonata. Curiously enough, Fontana
himself was put in charge of these investigations, and he immediately started
to drill holes into the passonata in order to establish whether they were rotten
or otherwise weakened. Meijer, who had denied all the charges, protested
and argued that exactly by investigating the passonata in this way the investi-
gators were destroying his constructions, and he urgently called for a halt in
the investigations. The papal administrators accepted his protests and
ordered Fontana to stop the investigation. In the end, it was shown that the
passonata was as healthy as could be and that there was no need to be afraid
of its being ruined by the river. But apparently, Meijer did not feel satisfied
and once again resorted to the means of publication to establish his reputa-
tion, now not only as an experienced craftsman. He hastily collected his
inventions and published his L'arte di restituire.
In a sense, this book is of course a direct continuation of his efforts in the
1670S to persuade future employers to hire him and assign him certain tech-
nical projects. Although it is lavishy illustrated, the book (in three parts)
essentially consists of a series of separate engravings (or series of engravings)
with more or less extensive and more or less scholarly comments. The 1683
book is much more sophisticated than the 1674 broadsheet or the 1679 me,
but the format is essentially the same. As can be seen from the dating of

'Cornelius Meijer inventor et fecit'


some of the engravings, Meijer used many old engravings for his new book,
replacing the explanations by new comments, but hardly changing the
engravings themselves. His L'arte di restituire, therefore, is still essentially a
collection of advertisements of an ambitious craftsman-engineer for whom
visual representations of his plans and projects were more important than
the comments and explanations.
On the other hand, elements of the book suggest that it is much more
than the sample book of an engineer. First, the high quality of the engrav-
ings is quite remarkable. Almost all of them bear the inscription "Cornelius
Meyer inventor & delineavit, Io.Bapt. Falda sculpsit," which indicates that
Meijer hired an expert hand only to engrave the drawings, not to draw the
pictures themselvesY Yet some of the engravings bear the inscription "Cor-
nelius Meyer inventor et fecit," suggesting that he did execute the engrav-
ings himself, without the help of other artists. Meijer was no member of the
Dutch artistic brotherhood in Rome, the Bentveughels, but he evidently had
talent and moved in these circles. His engravings of technical and scientific
instruments are remarkable for their fine details of surrounding persons and
architectural background. It looks as if Meijer wanted to impress his readers
not only with his technical skills, but also with his artistic talents, or, what is
more intriguing, to use his artistic talents to create an atmosphere in which
his mechanical expertise was much more easily accepted by the Roman elite.
Second, the book, especially the third part of it, contains some projects
that are completely new and seem to indicate a new social and intellectual
milieu in which Meijer moved. For instance, he proudly presents some of
the technical projects executed for the grand duke of Tuscany, Cosimo III, in
Florence and some of the Tuscan harbors. Even more interesting are the
engravings of fountains, scales, couches and, quite surprising, the beautiful
designs for reconstructing the main squares of Rome. On some of these
squares former popes had already erected an obelisk, but now Meijer
wanted to use these obelisks as sundials or stardials and to decorate them
with additional sculptures or new pavements. Rather delicate was the pro-
posal for a completely new pavement of the St. Peter's Square, one of
Bernini's greatest achievements. Around the famous obelisk (relocated in the
1580s by Carlo Fontana's distant relative Domenico Fontana) Meijer wanted
to decorate the pavement with the four systems of the world that were being
discussed by scholars at that time, the systems according to Ptolemy, Tycho
(both of them acceptable to the Church), Copernicus, and Descartes (both
unacceptable) (fig. II.2). Although strictly spreaking Copernicus's book had
not been put on the Index in 16r6 and even Jesuit mathematicians were
allowed to discuss Copernicus as long as they considered his system as a
purely mathematical hypothesis, it was still very uncommon, to say the least,
to confront the pope with these world systems.

Klaas van Berkel


f Y . 2 Meijer's plan for a new pavement for St. Peter's Square, displaying
the Tychonic and Ptolemaic systems of the world (in the text, he also included the
Copernican and Cartesian systems). From Cornelis Meijer, Carte di restituire
(Rome, 1685). Copyright O Amsterdam University Library.

T H E A C C A D E M I A F I S I C O M A T E M A T I C A ROMANA

The reason for introducing these architectural designs with their scientific
contents is elucidated by Meijer himself in the explanation belonging to his
engraving of a balance for establishing whether a crown was made from gold
or silver (Archimedes's famous device) (figs. 11.3 and 11.4). While he was
writing his book on the Tiber, he says, some high officials at the papal court
stimulated him to include some of these other inventions too. Evidently, he
had established relations with these courtiers during his negotiations regard-

'CorneliusMkjer inventor et f e d ' 285


ing the passonata and the project of restoring the navigability of the Tiber,
and they had become interested in Meijer's other inventions too. Meijer does
not say who they were, but he does mention in the book Giovanni Giustino
Ciampini, a high official at the papal court and a well-known scholar.
Ciampini (1633-98) is mainly known for his research on the churches of the
early Middle Ages, but he was also interested in modern science. 14 His house

<..StStftu7<e 11.3 Old ways and new ways to discover the amount of gold in a certain
piece of metal. From Comelis Meijer, Nuovi ritrovamenti (Rome, 1689). Copyright ©
University Library Leiden.

Klaas van Berkel


//.4 Instruments invented by Meijer to discover the amount of gold in a
certain piece of metal. From Cornelis Meijer, Carte di restituire (Rome, 1685).
Copyright 0 Amsterdam University Library.

near the S. Agnese in Agone was already the meeting place of a number of
learned circles when in 1677 Ciampini, in conscious imitation of the defunct
Florentine Accademia del Cimento, also founded a society for the cultiva-
tion of natural science, the Accademia Fisicomatematica Romana. Ciampini
owned a decent collection of scientific instruments, which he put at the dis-
posal of the members of his academy. Although it had no official charter
(even its name varied over time), the meetings of the Accademia were regu-
lated according to strict rules. The members met every week, one of the
members acted as secretary, and the discussions and suggestions for further
research were carefully recorded in a large notebook, which is still extant (at
least in part). Many of the members of this group of virtuosi were simply
amateurs, but a small number of them might be called professionals. The
Jesuit Francesco Eschinardi, professor of mathematics at the Collegio
Romano and highly regarded for his astronomical research, was perhaps the
most important of these professionals.'5 Although the first secretary of the
Accademia had the intention to publish the minutes of the meetings, just as
the Accademia del Cimento had done, the Roman Accademia as such never
published anything. But some members did, for instance Eschinardi, who in
the 1680s published several short pamphlets about his observations of several

'Cornelius Meijer inventw et fecit ' 287


comets. Some of these pamphlets have the form of a letter to the Italian
astronomer Giovanni Domenico Cassini, who was in charge of the oberva-
tory of the Academie des Science at Paris. Through personal contacts like
these, but also through the Giornale de' Letterati, a learned journal edited by
Ciampini, the Accademia Fisicomatematica tried to inform scholars and sci-
entists outside Rome of its existence and its activities. 16
Ciampini introduced Meijer to the Accademia in 1680, not only because
of his construction of the passonata (although that still remained his main
claim to fame),17 but also because Meijer was said to have been an active
member of the Dutch scientific community before his departure to Italy in
1674. Ciampini urged him to present his inventions and discoveries to the
Accademia, and as Meijer relates in his book, Ciampini also deemed them
worthy of inclusion in the proceedings of the academy. From then on Meijer
was a prominent member of the Accademia Fisicomatematica, as he proudly
indicated on the title page of his book, and to please his fellow academicians
he also included several of his inventions in the last part of his book.
Meijer soon developed a specialty of his own. From the very beginning he
seems to have had a prediliction for astronomy and for the observation of
comets more in particular (part of the design for the new pavement of the St.
Peter's Square was a list of all comets since Christ's birth up to the seven-
teenth century). Coincidentally, in 1680 and 1682 two (some thought three)
beautiful comets appeared in the skies, and Meijer had a perfect opportunity
to show the other members of the academy his expertise and his knowledge
of the literature. ls And although we should not overestimate Meijer's astro-
nomical expertise, with the help once more of some of his Dutch friends
(Abraham Genoels, a painter known for his mathematical skills) he was at
least able to draw some very detailed figures of the course of some comets in
his later publications (fig. I 1.5). But he certainly did not restrict himself to
comets. If we compare the notebooks of the first secretary of the academy
and the publications of some of the members, the correspondences with the
contents of Meijer's books are remarkable. The experimental work of Torri-
celli and that of Boyle were much discussed by the members of academy and
also playa prominent part in Meijer's later works.
The works Meijer published after 1685 all follow more or less the same
pattern. In 1689 he published his Nuovi ritrovamenti (New discoveries),
essentially a collection of new inventions of very diverse character. 19 This
book was expanded and republished in 1696 as his Nuovi ritrovamenti divisi
in due parti, while in the same year he published his L'arte di rendere i fiumi
navigabili in varii modl~ con altre nuove inventioni e varii altri secreti, divisa in
tre parti (The art or making the rivers navigable in various ways, with other
new inventions and several other secrets, divided in three parts). Actually,
these books are not really new books, because each new book consists of a

Klaas van Berkel


is&+ / / . 5 Meijer's representation of the orbits of the comets of 1682 and 1684.
From Cornelis Meijer, Nuovi ritrovamenti (Rome, 1689). Copyright 0 University
Library Leiden.

reprint of the former with a number of new topics and illustrations added.
In general, the new material consists of two kinds of information. The first
concerns new technical inventions. Meijer was hired for solving all kinds of
problems all over Italy, and the results were incorporated in the new edi-
tions. The second kind of additions concerns material that had been dis-
cussed in the Accademia Fisicomatematica. There are no notebooks of the
academy after the first two years, but we can reasonably assume that Meijer's
books are at least in part reproducing what the members of Ciampini's acad-
emy had been discussing in the 1680s and 1690s. Of course, the books were
Meijer's, and he adhered to the format he had already chosen for his Carte di
restituire. But we may safely assume that the other members liked his books
for that very reason, because the heavy stress on the visual representation of
the new experimental science was just what they needed. None of the other
members is known for his artistic or representational skills; all their Pam-
phlets are purely verbal reports on their research. Meijer is the only member
with the talents to draw his own illustrations and as a craftsman-become-
virtuoso the language of art was his favorite means of presenting the result
of the work of the academy to the learned world. His explanations are a

'Cornelius Meijer inventor et fecit ' 289


rambling concoction of new and old knowledge, including some fragments
from ancient authors whom he evidently was not able to read, but his draw-
ings are detailed, lifelike, and trustworthy representations of instruments
and experiments. His limitations therefore were, at least in the context of the
new science, at the same time his main assets.

DEFENSE AND DEFEAT

With the publication of L'arte di restituire Meijer undoubtedly had in mind


to show the world that he was not just a plain Dutch engineer, unskilled in
the arts and the sciences, but a member of the scientific elite of Rome. That
is why on the title page he proudly announced himself as a member "dell'
Accademia Fisicomatematica Romana." From the contents of the book it is
clear that Meijer even pretended to be an architect, just as Fontana was, and
it is no coincidence that he especially dwelled on the reconstruction of the
pavement of St. Peter's Square, because his arch enemy Fontana had just
begun to write an impressive book about St. Peter's Church.
Meijer's book certainly was a success, but his new standing as a virtuoso
seemed only to rekindle the feud between Meijer and Fontana rather than
end it. The rivalry between the two men appears to have inspired both of
them to compose new books-books they might not have contemplated if
they had not needed material to counter each other's charges. Meijer in this
conflict was the underdog, but it is remarkable that Fontana, who by now
had become the most important architect of Rome, an all-round artist who
was in the process of seeing through the press his monumental work on St.
Peter's Church, still was trying to ruin the career of someone who, as far as
we can see at least, did not pose any real threat anymore to his reputation. In
1694 and in 1697 Fontana and his followers once again openly charged Mei-
jer of having used the wrong material for his passonata, suggesting that
within a short while it would collapse after alU I1 New investigations were
ordered, with the same result as before: there was nothing wrong with the
passonata.
Apparently, Meijer was not sure that this would be the last time that he
and his passonata were to be criticized by Fontana and his followers.
Although he had already written several memoranda to defend his pas-
sonata, he now mobilized all his connections in Rome to deal a final blow to
all his critics and collect all the money the papal government owed him.
With the linguistic, rhetorical, and juridical help of the Roman lawyer
Francesco Maria Onorati, a member of the Accademia Fisicomatematica and
someone who had helped Meijer before, in 1698 the engineer published an
enlarged edition of an earlier Latin Memoriale as his Apologia per la passonata

Klaas van Berkel


fatta sopra il Tevere fuori di Porta del Popolo in difesa della strada Flaminiana. 21
This publication is a very detailed and rhetorically very sophisticated recon-
struction and defense of Meijer's career in Rome, with special emphasis on his
building the passonata. The self-presentation in this volume is somewhat dif-
ferent compared to Meijer's other publications. He repeats the story about the
passonata and the obstructions he had encountered during and after its con-
struction. But he also has his readers believe that he is a descendant of a rich
aristocratic family in Holland, his brother being the viceroy of the Dutch East
Indies. There is no independent evidence to confirm these claims, quite the
opposite, but it is clear that Meijer in his Apologia publicly posed as an aristo-
crat who was not treated accordingly by the papal bureaucrats and their Ital-
ian clients, Fontana being one of them. Fontana himself is not attacked right
away; although his name is not mentioned, the architect (referred to as 'un
gran Virtuoso') is even praised for his fine knowledge of architecture. But in
the second half of the Apologia Meijer subtly points out that Fontana had
overseen some details that were crucial to the construction of the passonata
and that only the superior technical expertise of Meijer, who stood in the tra-
dition of the celebrated Dutch engineers, had been able to build a construc-
tion strong enough to withstand the pressure of the water in the river.
In the end, Meijer was able to ward off all the charges brought against
him by Fontana and others, but in an indirect way, Fontana did succeed in
ruining Meijer's career in Rome. Even though the Dutch engineer proved to
be able to execute several difficult constructions in several places in Italy, he
never obtained the rewards promised to him by the papal administration.
He constantly had to pay for his expenditures from his own pocket. In the
Apologia he complains how he had to sell his cabinet of curiosities in Ams-
terdam just to stay alive in Rome. We are not able to verify this story, but it
is quite sure that dealing with the pope ruined Meijer financially.
Lack of success is also what characterizes Meijer's efforts to drain the
Pontine marshes-another proposal worthy of a Dutch engineer. 22 Though
others had tried to drain this region to the south of Rome before and had
never succeeded in doing so, Meijer was convinced that with his superior
skills he would finally succeed. Already around 1680 and later in the 1690s,
he drew up several contracts with wealthy Roman nobles and even dreamed
of obtaining a noble title after having opened the land for cultivation. In the
1690S he indeed hired workmen to dig canals and build sluices in the
marshes, but in the end the inhabitants of the surrounding villages and
towns successfully obstructed these operations. They rightfully feared that
they would lose the opportunity to acquire some additional income by fish-
ing in the marshes and by herding some of their livestock in the drier sec-
tions of the plain. So by the time of his death, in 1701, Meijer had still not
succeeded in draining even a small part of the marshes, and in his last will

'Cornelius Meijer inventor et fecit'


he handed over this task to his son Otto. Otto Meijer bravely continued the
work, but after a few more years he too had to acknowledge defeat.

CONCLUSION

As Hoogewerff saw it, the career of Cornelis Meijer foundered on the jeal-
ousy of the Italians. Notwithstanding the success of his passonata near the
Via Flaminia and the willingness of some amateurs of science to admit him
to the Accademia Fisicomatematica, he ended up impoverished and disillu-
sioned. His books, however, remain an important source for the history of
science and technology in Rome in the second half of the seventeenth cen-
tury. They contain information about the Accademia Fisicomatematica that
cannot be found anywhere else. In combination with his Apologia these
books also inform us about the social conventions regarding the practice of
science and technology in the age of baroque. And finally they draw our
attention to the visual aspect of science and technology. What Meijer had to
say on the course of comets or the effects of air pressure or the existence of
the void is not very important perhaps, but the way in which he presented
his conclusions does indeed merit our attention. As I suggested, in his repre-
sentation of science Meijer is still very much indebted to the customs and
conventions of the class of commercially operating engineers in Holland to
which he originally belonged. Yet this was exactly what the new scientific
elite in Rome needed in order to present their research in a trustworthy way
to the learned world. Meijer and the Accademia made a perfect fit. A closer
study not only of the texts, but also of the engravings these texts pretend to
explain, as well as further investigations into the contacts Meijer had both
with artists and with other engineers in Holland and in Rome might even-
tually shed more light on the development of the visual language of the new
experimental natural philosophy.

1. A recent discussion of the importance of visual language in propagating the new exper-

imental philosophy, especially the eclectic philosophy of the Leiden professor Arnold Sen-
guerd, can be found in Gerhard B. Wiesenfeldt, Leerer Raum in Minervas Haus. Experimentelle
Naturlehrean der Universitiit Leiden, 1675-1715 (Amsterdam: Edita, 2001).
2. G. J. Hoogewerff, "Comelis Jansz. Meijer, Amsterdamsch ingenieur in Italie (1620-
1701)," Oud-Holland 38 (1920): 83-1°3. This article is mainly based on Roman sources,
including Meijer's writings.
3. Giuliano Briganti, Caspar van Wittel e l'origine della veduta settecentessa (Rome: Bozzi,
1966); An Zwollo, Hollandse en Vlaamse vedute-schilders te Rome, 1675-[725 (Assen: Van Gor-

Klaas van Berkel


cum, 1973), 12off.; Giuliano Briganti, Caspar van Wittel. Nuova edizione a cura di Laura Lau-
reati e Ludovico Trezzani (Milan: EJecta, 1996).
4- Ellinoor Bergvelt and Renee Kistemaker, eds., De wereld binnen handbereik. Nederlandse
kunst- en rariteitenverzamelingen, 1585-1735 (Zwolle: Waanders, 1992) 51, 249-250, 324-
5. The Dutch text reads: Verscheyde aanwijzingen en bequame middelen, aan d'Hartogh van
Venetien & om alderhande Schepen, heen en weer, over rivieren en ondiepten te halen. A copy is to
be found in the Venetian Archives. See note 6.
6. Roberto Berveglieri, "Tecnologia idraulica olandese in Italia nel secolo XVII: Cornelis Jan-
szoon Meijer a Venezia (gennaio-aprile 1675)," Studi Veneziani, N .S., 10 (1985): 81 -91 (with illus-
trations); Idem, Inventori straniet1 a Venezia (1474-1788). Importazione de tecnologia e circolazione
di tecnici, artigiani, inventori (Venice: Istituto Veneto di Scienze, Lettere ed Arti, 1995),152-166.
7. Much has already been written on this subject. See for instance: Cesare d'Onofrio, It
Tevere e Roma (Rome: Bozzi, 1970), esp. 80-85; Paola C. Scavizzi, Navigazione e regolazione
fluviale nella Stato della Chiesafra XVI e XVIII secolo (il caso Tevere) (Roma: , 1991).
8. Roman correspondents of Leibniz, like Christian Albert Walter and Ehrenfried
Walther von Tschirnhaus, repeatedly informed the German philosopher and scientist of the
Dutchman and his passonata-strikingly without ever mentioning his name. Walter simply
referred to "un habile Hollandois" (a competent Dutchman), while Tschirnhaus, whose
report is quite detailed, talks about "Holland us." G. W. Leibniz, Siimtliche Schriften und
Briefe. Dritter Reihe, Zweiter Band (Berlin: Akademie Verlag, 1987) 33, 317, 383-384' (I
kindly thank Liesbeth de Wreede for drawing my attention to these letters.)
9. Versions of the report are to be found in the Biblioteca Corsiniana and the Biblioteca
Nazionale in Rome and in the Staatsbibliothek in Munich. They all are dense!y illustrated and
offer a wealth of material for comparing Meijer's and Van Wittel's contributions to the final report.
10. The copy of Meijer's booklet in the Vatican Library is titled L'arte di restituire aRoma
la tralasciata navigatione del suo Tevere (Rome: Varese, 1679), but this is only added in hand-
writing. The copy in the University Library at Amsterdam does not have this title, and it is
probable that the title is just a mistake. In the Ciornale de' Letterati of 1680 there is a review
of a book by Meijer that bears the title Delinationi con discorsi delle Reparationi del Tevere fatte
da Cornelio Meyer Olandese con alcum' pensieri circa la disseccationi delle Paludi Pontine, e fare un
nuove alveo al Fiume (Rome: Lupari, 1679), and this is exactly what the book is about. (The
review also gives the correct printer.)
I!. L'arte di restituire aRoma la tralasciata navigatione del suo Tevere, divisa in tre parti. 1.
Cl'impedimenti, che sono nell'alveo del Tevere da Roma a Perugia, e suoi remedii. 2. Le difficolta,
che sono nella navigazione del Tevere da Roma fino al mare, e suoi remedii. 3. Nel quale si discorre
perche Roma estato fabricata, e mantenuta su Ie sponde del Tevere, e si tratta d'alcun' altre propo-
sitioni proficue per 10 stato ecclesiatico. Dell' ingegniero Cornelio Meijer (Rome: Stamperia
della Reverenda Camera Apostolica, 1683).
12. Not everybody was convinced that the passonata would hold. In April 1678 Von
Tschirnhaus wrote to Leibniz: "Satis bene success it, sed cum non tam bene atque tam alte in
nostris regionibus extructum videatur, nescio an sit futurum durabile." Leibniz, Siimtliche
Schriften und Briefe 3.2, 389. This clearly is an echo of Fontana's (unfounded) criticism.
13. There is one engraving drawn by Caspar van Witte! and a few by other artists.
14. Dizionario Biografico degli Italiani 25 (1981): 136-143.
15. According to the entry in the Dizionario Biografico degli Italiani 43 (1993): 273, Eschi-
nardi "fosse in qualche modo il principale animatore."
16. On the Accademia: W. E. Knowles Middleton, "Science in Rome, 1675-1700, and the
Accademia Fisicomatematica of Giovanni Giustino Ciampini," British Journal for the History
of Science 8 (1975): 138-154. Since the author was interested mainly in the history of the ther-
mometer, his account of the activities of the Accademia is rather one-sided. More information

'Cornelius Meijer inventor et fecit' 293


is offered by Salvatore Rotta, "L'accademia fisicomatematica Ciampiniana: un' iniziativa di
Cristina?," in W. di Palma a.o., Cristina di Svezia. Scienza ed alchimia nella Roma barocca (Bari:
Dedalo 1990),99-186. As the title of the article already suggests, Christina of Sweden had
nothing to do with the Accademia. Ciampini and others may have hoped to place their soci-
ety under the protection of the former queen of Sweden, but she did not have the financial
means to do so. On the Ciornale: Jean-Michel Gardair, Le 'Ciornale de' letterati' di Roma
(1668-1681) (Firenze: L. Olschki, 1984), esp. ch. 7.
17. When Marcantonio Celli, fellow of the Accademia, wrote a letter to Cassini, inform-
ing him about what was going on in Rome, he also mentioned Meijer and introduced him as
the man who had constructed the passonata and thereby had saved the Strada Flaminia.
18. There is for instance a reference to Bartholomeus Schimpffer, Kurze Beschreibung des
dunckelen Cometen so an no 1652 den 8. Decembr erschienen darauffgemeiniglich sonderliche Enderun-
gen und Verwirrungen zueifolgen pflegen (Frankfurt am Main: Johann Phillips Weiss, 1653).
Schimpffer, an obscure almanac maker and astrologer in Halle (Saxony), predicted that the comet
would reappear in 1682 and since around that time indeed a new comet was seen (the comet of
Halley), on Meijer's suggestion the members of the Accademia ordered the book from some library
in Rome (nowadays it is very rare), had someone translate it, and then discussed its contents.
19. Nuovi ritrovamenti dati in luce dall' ingegniero Cornelio Meyer per eccitare l'ingegno de'
Virtuosi ad augmentarli 0 aggiungervi maggior pelfettione (Rome, 1689). Among the topics dis-
cussed by Meijer in this book we find eyeglasses, couches, the furnishing of a cabinet of
curiosities-including a chemical laboratory-the introduction of silkworm breeding in
Italy, and even some medecine. In the context of the treatment of illnesses caused by the little
animals seen through the microscope Meijer claims to have been present at some of the
anatomical dissections of Jan Swammerdam (who died in 1680), which once again indicates
that Meijer's interest in science dates back to his early life in Holland. Also included is a chap-
ter in which Meijer cites a number of biblical texts on the correct payment of businessmen
('Tenor Sacrae Scripturae de Mercede')!
20. In 1694, the year in which his magnum opus on St. Peter's Church was published (II
tempio Vaticano e la sua Orzgine, in Italian and Latin), Fontana also had the time to issue a Dis-
corso sopra Ie cause delle inondazioni del Tevere antiche e moderne d danno della cittd di Roma, e
dell' insussistente passonata fatta avanti la Villa di Papa Ciulio III per riparo della via Flaminia
(reissued in 1696). See Helmut Hager, "Le opere letterarie di Carlo Fontana come autorapp-
resentazione," in In Urbe Architectus. Modelll~ disegnz~ misure. La professione dell' architetto,
Roma 1680-1750, Ed. Bruno Contardi, Giovanna Curcio, (Rome: Argos, 1991), esp. 177-187.
21. The full title runs Apologia di Francesco Maria Onoratiper la passonata fatta sopra il Tevere
fuori di Porta del Popolo in difesa della Strada Flaminiana con Ie direttione del Signor Cornelio
Meyer,famoso ingegniere olandese all'eminentissimo e reverendissimo prencipe il Sig. Cardinale Cio.
Francesco Albano Segretario de Brevi di nostro signore. In Roma MDCXCVIII. In the foreword
Onorati confesses to be the mouthpiece of Meijer himself, "havendo io qui quasi la sola parte
di traduttore & interprete del suo linguaggio." This foreword also contains a direct reference
to the Dutch as being such able hydraulic engineers: the passonata was built "secondo i suoi
principii di far Passon ate all'usanza Olandese, che in quelle parti sostengono con artificio
incomparabile sopra il dorso della passonata il peso di tutte Ie acque di que! gran Mare che
bagna rOlanda." Onorati in his capacity of "ministro deputato per la Reverenda Camera" had
been the papal supervisor during Meijer's construction of the passonata, which means that he
had a vested interest in clearing Meijer of all charges brought against him by Fontana.
22. Meijer's efforts are relatively well documented by Hoogewerff, "Comelis Janszoon Mei-
jer," and J. Korthals Altes, Polder/and in Ita Iii'. De werkzaamheden der Nederlandsche bedijkers in
vroeger eeuwen en het Italiaansche polderland voorheen en thans (The Hague: Van Stock urn, 1928).

294 Klaas van Berkel


CONSUMPTION, ART, AND SCIENCE
This page intentionally left blank
Inventing Nature
Commerce, Art, and Science in the Early Modern
Cabinet of Curiosities

PAULA FINDLEN

"But is it a basilisk? "


"It's a saltwater fish that charlatans usually arrange in the form of a
basilisk, and it helps them deal with peasants in the piazza when they
want to sell their balsam."
-Carlo Goldoni, Famiglia dell'antiquario

CURIOSITIES FOR SALE

I n 1653, a curious book appeared in the city of Venice: Niccolo Serpetro's


Marketplace afNatural Marvels. Serpetro's encyclopedia was one of many
such volumes that satisfied the seemingly infinite desire for wonders in
the early modern period. It followed a rich publishing history of broadsheets,
natural histories, and encyclopedias of the strange and unfamiliar that char-
acterized the sixteenth-century love affair with the marvelous, and that cat-
aloged the many pleasurable and terrifying ways in which nature made
manifest the hand of God in the world. l Serpetro drew liberally from this
tradition to create his own theater of wonders. But he added one innovation
that was entirely his own: he placed his marvels in the marketplace, a teem-
ing piazza in which merchants and customers bargained over goods for sale,
and wares were displayed for all to see (fig. 12.1). It was the most fitting loca-
tion that he could imagine for the pursuit of wonder.
It is surely appropriate that a book published in the city of Venice-still
a thriving center for trade and commerce despite the challenges of such
northern cities as Amsterdam, Marseilles, and London, and the Spanish
gateway to the Atlantic, Seville - should imagine the world of marvels to
be a marketplace. Serpetro explained his metaphor as follows: "Since in a

297
& 2 . Charlatans selling their wares in Piazza San Marco, Venice. Source:
G. Franco, Habiti d'huomini e donne (Venice, 1609). Courtesy of Marquand Library of Art
and Archaelogy, Princeton University Library.
famous marketplace the wealthiest merchants come from many different
countries to show gems and the most precious and admirable things that one
finds in various provinces of the world, thus, in this work I tried to transport
from the most celebrated authors the rarest and most delightful marvels that
the Author of Nature has produced." He made his book a literal market-
place, dividing it into porticos, loggias, and shops so that passersby could
"walk easily" among them. 2 Each chapter was an imagined purchase, or at
the very least a bit of window shopping in the marketplace of marvels. Ser-
petro had probably taken the idea of the marketplace from works such as
Tommaso Garzoni's Universal Piazza (I585), a popular encyclopedia that col-
lected all the professions of the world into an imaginary piazza. 3 But he was
also an astute observer of his times. Nature was for sale in many market-
places throughout Europe. It was a commodity bought, sold, bartered, and
exchanged - the centerpiece of a series of transactions that connected the
world of commerce to the study of nature.
Shopping for natural curiosities was indeed possible by the time Serpetro
wrote his Marketplace of Natural Marvels. We need only think of the Dutch
tulip craze at the beginning of the seventeenth century to recall just how fren-
zied the market for a particular curiosity could become. 4 More generally, how-
ever, the growing popularity of cabinets of curiosities - private, princely, and,
in a few cases, institutional collections that grew in size and scope throughout
the early modern period - gradually transformed the act of collecting nature
into a business. It produced a world of entrepreneurs who saw nature in new
ways because of the culture of collecting. In February r644, John Evelyn
described the experience of walking through the merchants' stalls in the Isle
du Palais in Paris. One shop, in particular, caught his attention, a place called
Noah's Ark. There "are sold all curiosities naturall or artificial, Indian or
European, for luxury or use, as cabinets, shells, ivory, porselan, dried fishes,
insects, birds, pictures, and a thousand exotic extravagances."5 Evelyn had
found a cabinet of curiosities in which everything was for sale. Collectors
missing some choice item for their cabinets could depend upon the proprietor
of Noah's Ark to supply them with a sample-for the right price.
The idea of a shop filled with curiosities seemingly contradicted the
humanist ideal of scientific collecting as a series of exchanges among schol-
ars in which objects were freely given as an act of friendship; they accompa-
nied and embellished the words that described them. 6 Many collectors
accumulated the majority of their artifacts through travel and through the
generosity of other scholars with whom they regularly exchanged letters,
images, and specimens. But it was also possible to buy a cabinet of curiosities,
or at least its most important parts, by the early seventeenth century. Such
purchases were costly luxuries-not an act of scholarly inquiry into nature
but a sign that the pleasures of collecting involved more than the single-

Inventing Nature 299


minded pursuit of knowledge. If having a collection was one means by which
a prince or a merchant might proclaim his ability to command the world, cre-
ating a microcosm in which to receive visitors and to demonstrate his place in
a world of global commerce and conquest, then a collection was indeed worth
something. Many objects did have a precise monetary value, even if scholarly
collectors chose to ignore this aspect of the passion for curiosities.
The market for marvels produced more than one type of collector, and all
of them in different ways responded to the exigencies of the marketplace. In
addition to thinking about learned naturalists such as Ulisse Aldrovandi, who
created a theater of nature in late sixteenth-century Bologna in order to know
more about the natural world, we need to consider a different kind of collec-
tor who understood the idea of profiting from wonder. The early seventeenth-
century Augsburg merchant Philipp Hainhofer, for example, not only acted as
purchasing agent for rulers who sought out luxury goods, but explicitly made
his cabinets of curiosities, filled with objects acquired from merchants at the
Frankfurt fairs, to sell them. He speculated in nature. "When someone pre-
sents me with a foreign object for my Kunstkammer," he once said, "I experi-
ence more pleasure than ifhe had given me cash."7This comment that reflects
the ways in which curiosity and commerce worked harmoniously together.
Curiosities, after all, might be a good investment in terms of the favors and
ultimately business they might bring from certain patrons.
Hainhofer was perhaps one of the earliest collectors, following in the wake
of the Fugger merchants who had acquired many objects for princely
patrons, to recognize that the value of the cabinet of curiosities was not sim-
ply intrinsic. By the late seventeenth century, tales of the fantastic sums that
princes were willing to pay for a good cabinet circulated among connoisseurs
of such things . The grand duke of Tuscany's efforts to acquire the Dutch nat-
uralist Jan Swammerdam's collection of insects - and his expertise as a natu-
ralist-for I2,000 guilders were well-known and only increased the status of
this particular collection. s Around the same time, the duke of Modena
became so fascinated with Manfredo Settala's gallery of curiosities and inven-
tions in Milan that he attempted to purchase it. Negotiations had fallen apart
by the early r660s, but when John Ray and Philip Skippon visited the newly
installed cabinet in the ducal palace in February r664, visitors were still being
told of the outrageous sums the duke had been willing to pay for a collection
he never succeeded in buying. 9 The economic ability to afford a famous cabi-
net, in other words, had become a measure of one's status.
By the eighteenth century, natural history cabinets were put up for public
auction, and sale catalogs began to appear with greater regularity, reflecting
a full-scale commercialization of collecting culture. In the late sixteenth and
seventeenth centuries, however, it was rare to have an entire cabinet made
purely for profit. Individual objects might be heavily embedded in the world

300 Paula Find/en


of the marketplace, and occasionally princes might try to buy a collection, but
the majority of cabinets emerged from an individual collector's passion for
things. Yet even these collectors, however removed from the world of com-
merce, had to contend with the ways in which their fascination with nature
was fueled by a variety of individuals who had unique access to natural
objects and understood the profitability of nature. No collector could entirely
remove himself from the marketplace. How they interacted with it reveals a
series of interesting connections between commerce, science, and art.

TRAFFICKING IN NATURE

Early modern natural history was a product of the new material abundance
that flowed into European cities from all corners of the world. It extended
the medieval culture of buying and selling nature into new domains because
knowledge quite literally grew in proportion to the expansion of European
trade. lo While humanists who studied nature for the pure pleasure of
extracting knowledge might scoff at those who used their knowledge to turn
a profit, the fact remains that nature had always been a commodity to the
rest of the world. Since the Middle Ages the spice trade between western
Europe and the Levant- dried bits of nature that traveled thousands of miles
to satisfy the taste for the exotic - had shaped the commercial image of
nature. Columbus's attentiveness to the wonders of the New World in 1492
was hardly disinterested curiosity. He was not simply looking for the mon-
sters described in Pliny's Natural History, but also went in search of nature
for profit-cinnamon, balsam, aromatic woods, and unusual animals and
plants to delight the palate as well as the eye, and to cure the diseases of the
Old World with the nature of the New. 11
The profitability of nature was closely tied to its medicinal uses. Nature
provided the ingredients for a vast array of medicines in the ancient phar-
macopeia, from simple herbal remedies to more highly prized items such as
balsam, bezoar stones, and all of the key ingredients to create that panacea of
panaceas, theriac. 12 Merchants, apothecaries, and physicians together created
an economy of natural objects. They bought those parts of nature that they
could not cultivate and acquire on their own and transformed them into
medicines. The rarest and most exotic medicines, dependent on ingredients
from the Levant and later the New World, were usually composed of costly
ingredients. Serpetro's image of marvels lined up under the porticos for sale
was not at all improbable: it reflected the reality of the most marvelous
aspects of medicine.
If commerce and medicine established the essential contours of traffick-
ing in nature, faith placed a high premium on a different set of unusual

Inventing Nature 301


objects. By the thirteenth century the Crusades had created a lively trade in
relics, but also in natural objects that conformed to ancient accounts of the
marvels of the East. 13 The spoils of Christian conquest included a kind of
mythologized conquest of nature: Egyptian crocodiles, ostrich eggs, alleged
unicorn's horns, griffin's claws, and other examples of exotic nature, real and
imagined, began to appear in churches and treasuries throughout western
Europe. 14 While not as highly prized as sanctified relics, such objects
reflected a growing interest in the fantastic parts of nature described in
medieval bestiaries and other Christian allegories of nature that privileged
certain animals as harbingers of God's will. The basilisks, griffins, and drag-
ons found in the Bible and in such works as Pliny's Natural History became
more than paper fantasies of natural omens. Increasingly, they were actual
objects created to satisfy the taste for such curiosities. Such objects did not
disappear at the end of the Middle Ages but enjoyed a certain revival in the
age of the Reformation. In an era fascinated with reports of omens and
prodigies that signified God's will in a world of divided faith, they were
fully integrated into the cabinets of curiosities.
In all of these different ways, curiosity about the natural world shaped the
marketplace of marvels. The fascination with wonder helped to create a
kind of individual skilled in buying, selling, and creating wonder. Curiosity
created its own commerce-a world of specialists in natural curiosities that
we can only glimpse indirectly through accounts of nature in the early mod-
ern period. Such individuals did not aspire to interpret nature, but to sell
nature to those who created knowledge out of the raw ingredients of the
marketplace. Extraordinary things demanded a special expertise to acquire
and invent them, which gave them economic as well as symbolic value.
Understanding more precisely how learned collectors acquired the wonders
that they prized brings us into closer contact with the marketplace that they
were often loathe to discuss. It was a world filled with mountebanks and
charlatans who cultivated ties with physicians, apothecaries, and merchants
in order to sell their vision of nature to a public consumed by curiosities.
The tensions between those who sold nature and those who interpreted it
are evident in letters that accompanied objects in circulation among natural-
ists. When a learned collector crossed the imagined boundary between sci-
ence and commerce, he was subject to scathing criticism. In the 1590S, for
example, naturalists complained to each other about the practices of the
Basel physician Felix Platter who, according to one source, refused to make
gifts of his curiosities, selling "everything he has."15 This rather unusual
comment about a learned naturalist suggests that by the end of the sixteenth
century the line between science and commerce was increasingly blurred, if
it had ever been clear. Buying nature in the marketplace was a commonly
accepted practice among naturalists, a necessity to increase and replenish the

302 Paula Findlen


storehouse of knowledge. Selling nature, however, was an activity unworthy
of a natural philosopher. Or was it? Platter's decision to sell what he pos-
sessed suggests that it had become possible to put a price on the time and
expertise required to find and cultivate something rare, even for the pur-
poses of study.
Selling curiosities in a cabinet, or selling an entire cabinet, represents the
final step in a series of transactions that began the moment a curiosity
became available. Following an object from the beginning to the end helps
us to understand more precisely how science and commerce intersected. Let
us take the case of a curiosity that entered Ulisse Aldrovandi's collection in
Bologna in 1579, a gift of one of his regular correspondents, the Genoese
patrician Bernardo Castelletti. Castelletti exemplified well the meaning of
friendship in late Renaissance natural history. He routinely procured new
curiosities for Aldrovandi to describe in his great, unpublished natural his-
tory, and asked nothing in return but the pleasure of corresponding with a
famous naturalist. The gifts he gave to Aldrovari'di arrived in his hands by
many different avenues, and included curiosities he purchased in the public
piazzas of his city. In February 1579, Castelletti sent a letter announcing the
imminent arrival in Bologna of the most marvelous fish he had ever seen:
"What's more, you will have a fish that is one of the rarest and most extrav-
agant parts of nature in the sea." He described how he acquired this rather
ugly fish with bulging eyes:

It was given to me dried, as I send it to you, by the fisherman who caught


it, who, upon seeing it had such strange features, didn't throw it back into
the sea, as fishermen usually do with all the other useless fish. Indeed he
kept it alive as long as he could, and then had it dried to show to people as
a miraculous thing. ' 6

The ingredients in the story are the stuff of which cabinets of curiosities
were made and replenished: a useless fish, an ambitious fisherman, an audi-
ence eager to pay to see natural oddities, and collectors who could not resist
acquiring them. This was quite literally the experience of nature in the mar-
ketplace.
The dried monster made its way to Bologna and Aldrovandi added it to
his museum. He may have even had his artists illustrate it and dictated a
description to his scribes, in preparation for its inclusion in his Natural His-
toryY This at least was Castelletti's fear a few years later. Apologetically and
quite reluctantly, he informed Aldrovandi that the marvel had been invented
by the fisherman who sold it to him. "I am sorry to have to tell you that in the
description of the fish sent to you some years ago, I was deceived .... " Wor-
ried that he had compromised the veracity of Aldrovandi's account of

Inventing Nature
nature, he confessed that his words as much as the object itself were not a
reliable source of information. They, too, had been bought and sold in the
marketplace: "they are those that the fisherman sold me."!8 A clever vendor
of the nature's bounty had tricked a gullible humanist into believing that all
the monsters found in Pliny's Natural History truly might be acquired for
one's museum, if only one looked hard enough. Castelletti had forgotten the
golden rule: caveat emptor.
Such episodes give us further insight into the way in which the commerce
in natural curiosities responded directly to the collector's passion for the
exotic and unknown. A city like Genoa was a trading zone for natural
curiosities. When the grand duke of Tuscany commissioned his botanist at
the University of Pisa, Francesco Malocchi, to acquire curiosities for the uni-
versity garden and its museum during the summer of 1599, Malocchi
planned an itinerary that made Genoa his final destination. Malocchi's buy-
ing trips were, in essence, a merchant's itinerary to the port cities ofItaly. In
each city, he encountered men who had curiosities to sell, and made pur-
chases for the grand duke which were recorded in his ledger. In April r604,
for instance, Malocchi acquired an entire "whale skeleton"-a rare prize for
an early modern natural history museum - in the port city of Livorno.!9 He
was more successful than the French royal surgeon Ambroise Pare, who was
fascinated by "a head of a large fish in the house of a rich merchant" in Lyon
that he hoped to acquire for King Charles IX.20 Unfortunately the fish was
quarantined with the family during a plague epidemic, and that was the last
Pare ever saw of it. These and other anecdotes suggest that naturalists rou-
tinely visited merchants who owned and sold curious things.
Knowledge of nature could not increase without the commerce in nature.
Naturalists had to come to terms with the marketplace in order to pursue
curiosities. Digging further into Aldrovandi's correspondence we find indi-
cations that he knew some of the famous charlatans of his day who made
and sold curiosities in the piazza, and considered them an interesting source
of knowledge as well as artifacts. In April 1568, for example, a correspon-
dent from Piacenza described their mutual acquaintance "Master Leone
who sells his wares in public often, and is known to all the apothecaries in
Venice."2! Leone Tartaglini of Foiano was a famous mountebank known to
most collectors of natural curiosities in late sixteenth-century Italy. He
inhabited the Venetian piazza famously depicted as filled with men of his
profession. Naturalists traveled from cities as dispersed as Lucca, Piacenza,
Bologna, and Verona to see his cabinet of curiosities in Venice, which was an
early precursor to the Parisian Noah's Ark that Evelyn described. Many of
the objects Tartaglini possessed were evidently for sale. Among other things,
he specialized in the sort of extravagant fish that Castelletti admired and pur-
chased in Genoa. Visitors to Venice reported that he had a book illustrating

Paula Findlen
all of his dried fish-a book Aldrovandi, among others, wanted very much
to seeY While many naturalists collected images of curiosities in order to
create a complete archive of the natural world, in Tartaglini's case, such a
book might have served the additional purpose of advertising the kind of
nature that he sold.
The image of the seller of nature as a mountebank appears not only in
descriptions of Tartaglini's activities in Venice, perhaps the most famous
vendor of curiosities of whom we have a precise description, but also
informed other accounts of the buying and selling of curiosities. In Novem-
ber 1663, the English traveler Philip Skippon encountered a mountebank
named Rosachio, an astrologer who sold medicines in Piazza San Marco.
Skippon was evidently fascinated by Rosachio; he followed his initial
encounter with the mountebank by visiting Rosachio at home in order to see
his "collection of rarities." In it was a flying serpent-or at least an alleged
flying serpent since Skippon described it as having "a long furrow on either
side, in which were cartilaginous parts (he said) when it was alive, that served
for wings."23 Skippon's traveling companion, the great English naturalist
John Ray, evidently did not find the alleged dragon worthy of note since he
neglected to include it in his own journal of the same voyage, but the fact
remains that a century after Tartaglini had succeeded in getting all the great
naturalists of Italy (and undoubtedly other regions) to visit his cabinet in
Venice, mountebanks were still selling the same bits of artificial nature to the
heirs of Aldrovandi. In the 1672 catalog to his museum in Verona, Count
Lodovico Moscardo continued to discuss the "swindlers and charlatans from
Dalmatia" who sold examples of the basilisk in his museum. 24
The network of people who bought and sold nature was composed of
more than just charlatans and random fishermen who showed exotic fish in
the fish markets. Let us return for a moment to the fact that Master Leone
of Venice was known to all the apothecaries of the city. Were they as much
the source of his curiosities as he was of theirs? In his History of Animals
(I558), the Swiss naturalist Conrad Gessner informed his readers about
"apothecaries and others who usually dry rays and shape their skeletons into
varied and wonderful forms for the ignorant."25 Rather than condemning
charlatans, Gessner blamed apothecaries for facilitating this trade, indeed
accused them of inventing fraudulent curiosities. We can find traces of rela-
tionships among apothecaries and mountebanks in surviving correspon-
dence. The Veronese apothecary Francesco Calzolari, for instance, was so
intrigued by reports of Master Leone's activities that he sought out the artist
who had illustrated the Venetian's curiosities. 26
Scholarly collectors recognized that the pharmacy was both a world of
wonder and an extension of the marketplace. They entered it expecting to
find an invented nature (fig. I2.2). Visiting apothecaries was an important

Inventing Nature
& a 2.2 The apothecary Francesco Calzolari's museum in Verona, filled with
many strange fish and reptiles hung from the ceiling. Source: Benedetto Ceruti and
Andrea Chiocco, Musaeum Franc[isci] Calceolarii Iun[ioris]. Veronensis (Verona,
1622).Courtesy of the Biblioteca, Universitania, Bologna.

part of the collector's itinerary. John Ray took pleasure in "a so-called siren's
rib" in the apothecary Jean van der Mere's collection in Delft and visited the
apothecary Mario Salb in Verona, who claimed to have the "reliques of Cal-
ceolarius his M~seum."~' Possibly one of the items surviving from Calzolari's
museum that did not especially impress Ray in 1663 was the unicorn's horn
that Aldrovandi saw when he visited his shop at the Sign of the Golden Bell
in Piazza dell'Erbe in 1571. Aldrovandi was too polite to tell Calzolari that
it was a fake, but he privately noted "that there is no doubt that it is not a
true example."2s Such objects were also the ordinary stuff of any cabinet of
curiosities. But apothecaries, who practiced a certain alchemy on nature to
create their medicines, must have seen the fabrication of natural objects as a

306 Paula Findlen


demonstration of professional skill-the ability to manipulate nature. They
filled their shops with those marvels, real and imaginary, that helped to sell
their medicines and reminded people of the apothecary's close connections
with the world of art to which they were officially joined in towns where
painters and apothecaries belonged to the same guild because both trans-
formed the raw ingredients of nature into art. 29
Collectors understood that the more unusual nature seemed to be, the
more likely it was a product of their own demand for a certain kind of won-
der. They repeatedly offered advice on how to discern the difference
between an authentic and fabricated version of nature. The Milanese physi-
cian Girolamo Cardano advised his readers to inspect the joints and sutures
of marvelous creations in order to see if they had been put together by
human rather than divine hands. 30 But the possibility of fraud did not make
collectors any less interested in acquiring them - quite the opposite since
invented bits of nature were highly prized. One of the less well-studied
aspects of the cabinets of curiosities regards the significance of objects that
purported to be natural while actually being artificial. These fabrications
allow us to understand how commerce and science helped to create the art
of nature in the early modern period.

INVENTING THE HYDRA AND THE BASILISK

The most popular fabrications of the sixteenth and seventeenth centuries were
hydras and basilisks. They took their place in the cabinet of curiosities among
the many different kinds of dragons that fascinated early modern collectors.
Flying dragons, eagle-fish, and other hybrids of the imagination emerged
from the pages of medieval bestiaries and church and princely treasuries to fill
Renaissance museums. They did so according to rules of art that were best
expressed in a passage from Leonardo da Vinci's notebooks. Describing how
to make an imaginary animal appear natural, Leonardo wrote:

You know that you cannot make any animal without it having its limbs
such that each bears some resemblance to that of some one of the other
animals. If therefore you wish to make one of your imaginary animals
appear natural-let us suppose it to be a dragon-take for its head that
of a mastiff or setter, for its eyes those of a cat, for its ears those of a por-
cupine, for its nose that of a greyhound, with the eyebrows of a lion, the
temples of an old cock and the neck of a water-toftoise.3 !

The rules of good painting applied no less to the three-dimensional con-


struction of an imaginary animal. In order to be convincing, it had to origi-

Inventing Nature
nate in nature. Leonardo's contemporary Albrecht Durer also believed that
art emerged from nature, and he strove hard to give the beasts of the Apoc-
alypse a more anatomical appearance. le Leonardo's example of a dragon was
hardly casual because it was indeed the imaginary animal of choice. It was
the most fantastic and symbolically potent animal in the Christian imagina-
tion, worthy of multiple inventions across the centuries.
The hydra and the basilisk - two of the most elaborate kinds of dragons
described in ancient and biblical sources-had a level of complexity that
many other natural inventions did not. A unicorn's horn was the horn of a
narwhal. A griffin's claw was often a bison's or ox's horn. Many inventions
of nature, in other words, were entirely natural. They simply involved an act
of reinterpretation in order to see the imaginary in the real. Objects that
took shape through the manipulation and transformation of nature
belonged to an entirely different category. They were truly works of art in
which one could take pleasure in the possibilities that nature suggested to
the human mind.
Not coincidentally, they were also objects on which one could put a
price-repositories of economic as well as spiritual capitalY Conrad Gess-
ner described a hydra that had been brought from Turkey to Venice in 1530
and acquired by the king of France (fig. 12.3). "It is appraised at six thou-
sand ducats," he wrote in 1560. Like Durer's rhinoceros, Gessner's hydra
was an image derived from an image. He lifted it from broadsheets such as
Durer's popular Whore of Babylon (1498) that depicted the seven-headed
beast of the Apocalypse with vivid clarity for a public eager to see signs of a
world in turmoil. An encyclopedia such as Conrad Lycosthenes's, Chronicle
of Prodigies and Portents (1557) was probably the more direct source for Gess-
ner's illustration. Evading the issue of its truth or falsehood, Gessner chose
instead to comment on the hydra's art. "The ears, tongue, nose, and faces are
different from the nature of all species of serpents. But if the author of such
an invented natural thing were not ignorant, he would be able, with great
artifice, to trick observers." 34
Aldrovandi agreed with Gessner's assessment of the hydra. After receiv-
ing a hydra from a Ferrarese noble who wanted to know if it was authentic,
Aldrovandi responded that, given the confused description of the hydra
among the ancients-an animal with three, four, seven, nine, or even ninety
heads - it was hardly surprising that no one knew the truth about it. He
reflected on how others had profited from this uncertainty: "it is no wonder
that in our age some have been deceived by the miraculous artifice with
which these hydras are faked from other bodies and put together, as they
have also done with the flying dragon - which however does exist in
nature-trying to imitate it by using a species of marine ray, as one can see
in my study." His assessment of the hydra of Ferrara was mixed. The body

Paula Findlen
GF ~ 2 . 3The King of France's many-headed hydra, valued at 6,000 ducats in
1560. Source: Conrad Gessner, Nomenclator aquatilium animantium. Icones animal-
ium (Tiguri, 1560). Courtesy of Department of Special Collections, Stanford University Libraries.

and tail came from the "true flying dragon that is born in Arabia and
Egypt," but the heads had come from different animals and one could see
that various parts of the dragon-its wings and its hind legs-had been
removed to give it the appearance of a hydra.35It was a half-true specimen,
a wonder of nature transformed into a work of art by the desire of nature's
artisans to turn a profit. Better, in short, than the hydra of San Marco in
Venice, which he declared to be patently false.36For this reason, Aldrovandi
engraved it for eventual inclusion in his History of Serpents and Dragons.
Undoubtedly because hydras were often found in state treasuries -the
Venetian doge had a fine example with nine heads, for instance3'-visitors
talked more self-consciously about their monetary worth (and went to
famous naturalists such as Aldrovandi to see if they would authenticate
them, which surely increased their value). Skippon, for instance, admired
the seven-headed hydra in the duke of Modena's gallery, originally a gift of
the Holy Roman Emperor Charles V to the Gonzaga family of Guastallo.
He, too, refused to say with certainty that the hydra was a fake, though he
commented on "the head being like that of fitchet, or of that kind, the body
and feet were of a rabbet or hare, and the tail was made of common snake's

Inventing Nature 309


skin, the back and neck covered with the same." But he was very precise
about the origins of the hydra since it was part of a collection bought from
"Zennon the apothecary for 300 doppii" when the d'Este family was unable
to acquire Settala's collection. 38 An apothecary's hydra , of course, had its
price - and it was much lower than the value of objects normally associated
with a ducal treasury.
The process of inventing nature fascinated early modern naturalists.
While they might condemn mountebanks for preying on the gullible and
the ignorant, they could not contain their own delight in understanding how
nature could be invented or suppress their admiration for the artistry
involved in making monsters. As naturalists collected and inspected the
variety of nature with greater regularity, they began to put into print their
observations on nature's fabrications. Dragons were a focal point of this dis-
cussion and, more often than not, such discussions appeared in ichthyolo-
gies-further underscoring the idea that the point was not to talk about
dragons sui generis but to discuss how to make them from fish. In his Natural
History of Strange Fish (1551), Pierre Belon described the passion of many
people for dragons "made for pleasure such as those that we see counter-
feited with rays disguised in the manner of a flying serpent."39 Conrad Gess-
ner's complaint in 1558 about fraudulent apothecaries came in the midst of a
lengthy discussion of dragon-making in his History of Animals. In a chapter
on rays, he described in great detail how such monsters were made. "They
bend the body, distort the head and mouth, and cut into and cut away other
parts. They raise up the parts that remain and simulate wings, and invent
other parts at will."40 Understanding the possibilities of the ray as a dragon
in potentia was the first step in appreciating the art of the dragon.
The ability of many naturalists to look critically at the anatomy of the
hydra, the basilisk, and many other kinds of dragons reflected the shifting
religious and intellectual climate. In the early decades of the sixteenth century,
such creatures were sufficiently charged with religious meaning that it would
have been heretical to suggest that they were anything less than God's will. By
the 155os, it had become possible to inspect these portents as examples of
nature's variety and to suggest that human intervention made them approxi-
mate people's fantasies of a terrifying nature. Increasingly, such objects seemed
to evoke pleasure more than horror. 41 The Renaissance dragon, after all, was
usually no more than a couple of feet long. Cardano simply couldn't imagine
how many of the specimens he saw could fly. J. c. Scaliger contented himself
by observing: "The skin is like that of a ray."42 One wanted to know how they
were made while avoiding the question of whether they existed.
Naturalists actively collected and traded these physical talismans of the
medieval and Reformation culture of portents-no longer clear demonstra-
tions of the mysterious ways in which God's will manifested itself in the

310 Paula Findlen


world, but increasingly desirable items for cabinets of curiosities. In 1573,
the French surgeon Ambroise Pare recalled a marine monster that Cardano
sent Gessner, "which had a head similar to a bear and hands almost like a
monkey, and the rest of a fish."43 Such descriptions reveal the pleasure natu-
ralists took in understanding how harmonizing the many parts of nature
into something new and unexpected might be an art unto itself. Examining
the griffin's claw in the treasury of Charles V, for instance, Cardano
reflected: "perhaps by carving out an ox's horn, art invents nature."44 The
ability to dissect the bestiary that made the beast gave credence to the idea
that knowledge did transform how one looked at an object. If commerce
responded to curiosity by inventing what people wanted to see, then science
responded to art by understanding that the boundaries between nature and
art were there to be crossed. We need only think of the jeweled boxes that
German and Italian artisans made in the shape of crocodiles and dragons,
the French artisan Bernard Palissy's ceramic re-creations of nature, or the
flying dragon chandelier that Durer designed in Nuremberg, which used
the natural shape of a stag's antlers for its wings, to recognize how the idea
of making art from nature was a central theme of the late Renaissance. 45
By the time Aldrovandi's On Fish appeared posthumously in 1613, it
reflected the new sensibility of late Renaissance natural history toward the
idea of inventing nature. While indebted to all previous publications that
had discussed fabulous creatures in the cabinets of curiosities, Aldrovandi's
book improved upon them by showing the artifice of inventing dragons and
basilisks from rays in greater visual detail. His work included no less than
two images of a "ray dried and shaped in the form of a dragon" as well as a
"sea-eagle" that he declared to be patently false (fig 12-4).46 Examining these
images, we can see explicit efforts to demonstrate the artifice of the object in
question while retaining the canonical form of the dragon.
Aldrovandi's images of flying dragons provided the introductory mater-
ial for a chapter on the basilisk in Bartolomeo Ambrosini's edited version of
Aldrovandi's History of Serpents and Dragons (1640). A small, solitary African
dragon described by Pliny and Galen, it was reputedly so poisonous that it
could kill someone with its breath or its glance, dry plants, and break stones
in half.47 It quickly became the canonical example of a work of nature trans-
formed into a work of art (fig. 12.5). Aldrovandi reported that the great
physician Girolamo Mercuriale had found a "basilisk's cadaver" in the trea-
sury of the Holy Roman Emperor Maximilian II. Be (or his editor
Ambrosini) tactfully chose not to comment directly on the imperial basilisk,
restricting himself instead to condemning those "imposters" who frequently
made basilisks "out of small dried rays."48
Aldrovandi's comments on the invention of the basilisk rested on the more
extensive critique of this animal composed by the imperial physician Pier

Inventing Nature
&$*~;ue Y2.4 The "dragon formed a ray" in Ulisse Aldrovandi's museum in
Bologna. Source: Ulisse Aldrovandi, Depiscibus (Bologna, 1613). Courtesy of Depart-
ment of Special Collections, Stanford University Libraries.

Bafillfcus ex Raia e&&s pro&#&lupinbpi&s.

& ~ U J W Y2.5 Aldrovandi's basilisk. Source: Ulisse Aldrovandi, Sevpentum et dra-

conum historiae libro duo, ed. Bartolomeo Ambrosini (Bologna, 1640). Courtesy of
Department of Special Collections, Stanford University Libraries.
Andrea Mattioli. In the expansion of his 1544 commentary on Dioscorides's
De materia medica, the leading handbook on medicinal simples since antiquity,
Mattioli added a section on poisons. The final chapter of his popular com-
mentary was devoted to the basilisk. "The variety of stories makes me easily
believe that one can't determine anything about this animal," wrote Mattioli,
"or know what its true history might be among all the stories told."49
Nonetheless he proposed a few logical questions about the idea of the basilisk
that reflected the growing numbers of specimens in cabinets of curiosities.
How could something so dangerous that it could kill men instantly be so eas-
ily captured? If it were so small, how could men, observing it from a safe dis-
tance, see enough of its features to report on details such as the three points on
the crested head, or the crown that it was often thought to wear? The basilisk,
after all, had no Hercules to slay it like the hydra, nor a tale equivalent to the
decapitation of the deadly Medusa. Only divine providence, or human delight
in the endless invention of nature, could bring it into the museum.
The longevity of the basilisk, well beyond the period in which there was
any doubt about its authenticity, suggests the importance of understanding
the relations between science and art in the early modern period. Certainly
the decision to make basilisks a prominent part of the iconography of nat-
ural history was a contributing factor. Each image created a prototype of an
object that could be made by looking at its engraving, and remade by copy-
ing these images into new natural histories. Aldrovandi's fake dragons
enjoyed a wide circulation in the seventeenth century. They reappeared in
later editions of Aldrovandi's posthumous natural history and eventually
found their way into Joannes Jonstonius's Natural History of Serpents (1657).
In this work, Jonstonius brought together all of Aldrovandi's images to
demonstrate the art of inventing nature (fig. 12.6). When it came to the
basilisk, Jonstonius offered no lengthy discussion of its physical form, cus-
toms, and mythology, as earlier naturalists had done. He simply noted:
"They are formed from a ray, just as one can see from this image. Preserved
in the Bologna Museum."5o
By 1622, connoisseurs of basilisks could enjoy a competing image of this
dragon by turning to page 90 in the new and improved catalog ofCalzolari's
museum written by Benedetto Ceruti and Andrea Chiocco, two Veronese
physicians in contact with the apothecary's grandson (fig. 12.7). There was a
splendid portrayal of all the unique features of the basilisk, with a level of
detail that no previous image had captured: the diadem decorating its
crested forehead, the scales covering its wings and tail, the strange fins on
which it balanced, and, most importantly, the act of flight. The engraver had
succeeded in bringing the basilisk to life. Lest there be any confusion, Ceruti
warned his readers: "You should know, lest any lies are discovered in our
nomenclature, that this is neither a basilisk nor a dragon, but a fish from the

Inventing Nature
42.6 Joannes Jonstonius' reproduction of Aldrovandi's invented
dragons. Joannes Jonstonius, Historia naturalis de serpentibus libri II (Amsterdam,
1657). Courtesy of the Biblioteca Nazionale Centrale, Florence.
~~ ~2.7Calzolari's basilisk in Verona. Source: Benedetto Ceruti and Andrea
Chiocco, Musaeum Franc[isci] Calceolarii Iun[ioris] Veronensis (Verona, 1622).
Courtesy of the Biblioteca Nazionale Centrale, Florence.

sea-an ill-shaped ray of course- worked into this shape by the hand of an
artisan." He invited his readers to admire how Calzolari's monster, "exhib-
ited for viewing," imitated the shape of the ba~ilisk.~'
Calzolari's engraved basilisk quickly supplanted Aldrovandi's illustra-
tions as the canonical depiction of an object that did exist, even if the animal
did not. When Lodovico Moscardo published two catalogs of his own col-
lection in 1656 and 1672, he reproduced Calzolari's image. Of course in
Moscardo's case it is not unlikely that he had Calzolari's actual basilisk, since
both collectors came from the same city. It was in regard to this particular
basilisk that he offered the opinion that it "had been shaped in this way by
swindlers and charlatans from Dalmatia, and shown by them in public
stands to the people as a true basilisk."52The image, in other words, now
fully demonstrated the art of invention and the collector's role as a critical
consumer in the marketplace of marvels.
The power of Calzolari's image and its circulation in various catalogues
throughout the seventeenth-century attracted visitors to Moscardo's
museum who wanted to inspect the art of the basilisk. Catalogs gave objects
a double life; visitors experienced them both in word and image, before see-

Inventing Nature 3'5


ing them in the museum. In 1687 Maximilian Misson stood in front of
Moscardo's basilisk, which he had surely encountered first in print, and dis-
cussed how it was made in greater detail than any of his predecessors: "the
invention is most pleasing and a thousand people are fooled by it." He added
to Gessner's original description of 1558 by noting the way in which a darted
tongue was neatly fitted into the fictitious mouth, and claws and enamel eyes
were added "with some other little parts dexterously put together." With great
pleasure, he concluded, "And voila! The invention of the basilisk."'53 The
Verona basilisk did not disappear from view in the next century, but gained
further currency as naturalists more aggressively cultivated their reputation as
debunkers of ancient superstitions. When the catalogue of the Nuremberg
apothecary Basil Besler and his son Michael Rupert's cabinet appeared in 1716,
it also contained an image of the 1622 Calzolari basilisk, further cementing its
reputation as the measure of this particular wonder (fig. 12.8).
Even museum catalogs that did not include the image of Calzolari's
basilisk made reference to it as the best instance of the engraver's art captur-

r.S3r~ 12.8 Basil and Michael Rupert Besler's basilisk, with a demonstration of
how it was made from a ray. Source: Rariorum Musei Besleriani quae olim Basilius et
Michael Rupert Besler collegerunt, aeneisque tabulis ads vivum incisa evvulgarunt: nunc
commentariolo illustrata a Johanne Henrico Lochnero (Nuremberg, 1716). Courtesy of
the Biblioteca Nazionale Centrale, Florence.

Paula Findlen
ing the fabrication of nature. The papal physician Johan Faber, for instance,
referred to the "most beautifully engraved figures" of the basilisk in flight in
the Ceruti and Chiocco catalog to highlight the unusual appearance of a
dragon's skeleton he included in his edition of the Spanish royal physician
Francisco HernPndez's famous Treasure of Medical Things of New Spain
(1649) (fig 12.9).~~
Cardinal Francesco Barberini's dragon had been carefully
inspected by Faber, who pronounced it authentic in every respect. One way
to demonstrate its authenticity was to depict it in a manner different from
the Calzolari basilisk because it had become the canonical image of the
invention of nature.
There were many technical reasons to admire the Calzolari image. In
comments such as Faber's we get a glimpse of the naturalist as a savvy con-
sumer of the art of printing as a technique for reproducing nature. By the
time Lorenzo Legati composed the 1677 catalog of Marchese Ferdinand0
Cospi's museum in Bologna, he was no longer satisfied with Aldrovandi's
depictions of the basilisk. While referring his readers to the images of 1640-
which were actually woodcuts done at the end of the sixteenth century-he
told them that the image of the Calzolari basilisk more closely approximated
the object he was trying to describe. "Other than being most finely engraved

a%- Y2.9 Cardinal Francesco Barberini's dragon in Rome, as described by


papal physician Johan Faber. Source: Francisco Hernindez, Rerum medicarum
novae hispaniae thesaurus seu plantarum animalium mineralium Mexicanorum Historia
ex Franczsci Hernandez, ed. Johan Faber (Rome, 1649). Courtesy of Bancroft Library,
University of California, Berkeley.

Inventing Nature 3'7


in copper, it also articulates the spines and roughness of the tail that one does-
n't observe in the first figures."55 Legati did not follow Moscardo's example of
including the image, on the presumption that readers of his catalog would
simply turn to a copy of the 1622 catalog to confirm his opinion. In a much
more decisive way, Legati reminded his audience that depicting a dubious
nature was a special kind of art.
Only one seventeenth-century scholar took the image that Faber created
in the 1640S to be a better likeness of a dragon. Both the image and descrip-
tion of Faber's dragon appeared prominently in the German Jesuit Athana-
sius Kircher's Subterranean World (r 664). ,6 But this was hardly surprising
since Kircher was in the midst of dissecting a dragon's head with the Bar-
berini librarian Hieronymus Lancia after a flying dragon made its appear-
ance in Rome in r660. Kircher was perhaps the last naturalist to believe
passionately in the reality of any papal dragon he saw, even though he knew
well the stories of basilisks invented from rays. His successor as curator of
the Roman College museum, the Jesuit naturalist Filippo Bonanni, tactfully
chose not to discuss the Barberini dragons, confining his comments instead
to a splendid example of a dried ray "sold by some as a basilisk." It was
surely one of the two rays "formed by art" that Giorgio de Sepi described in
the 1678 catalog of Kircher's collection. 57 By the end of the seventeenth cen-
tury, the vast majority of naturalists agreed with Ceruti that hydras,
basilisks, and dragons existed only to the extent to which artisans and
engravers could bring them to life.

MONSTROUS CODA: THE LAST HYDRA FOR SALE

No account of the early modern invention of nature, however, ends without


the moral that belief is more powerful than any number of criticisms that
might demolish it. The last hydra to preoccupy the community of naturalists
belonged to two merchants in Hamburg. In the city of Amsterdam, the
apothecary Albert Seba routinely enjoyed the company of visitors to his
famous cabinet of curiosities. Around 1720, he began to hear tales of the
hydra of Hamburg (fig. 12.10). At first, he dismissed it as a mere fable. A
year later, a minister told him the same story and brought him an image of
the hydra. But what finally convinced Seba that he needed to know more
about the hydra was its price. "When I heard that it was for sale for 10,000
florins, a detail he confirmed, the immensity of the sum reawakened my
desire to have a faithful copy of it." Seba's response to the hydra was not all
that different than his contemporary Antonio Vallisnieri's reaction to the
basilisk. Vallisnieri, one of the famed professors of natural history who prac-
ticed the kind of critical, microscopic natural history that his mentor Mar-

Paula Findlen
/2./0 The hydra of Hamburg, engraved for Albert Seba. Source: Albert
Seba, Locupletissimi rerum naturalium thesauri accurata descriptio, et iconibw art$-
ciosissimis expressio, p w universam physices historiam (Amsterdam, 1734), vol. I , table
CII. Courtesy of Bancroft Library, University of California, Berkeley.

cello Malpighi pioneered, kept a basilisk in his collection in Padua in the


early eighteenth century because he could not believe the "high price" that
an "Armenian trickster" had gotten for it.58
Commerce indeed was the final wonder of the art of inventing nature.
Both truth and falsehood had their price. Seba immediately wrote to a fel-
low apothecary in Hamburg, asking his opinion of the hydra. "He assured
me that it was in no way a work of art, but truly one of nature." 59 The
apothecary Natorp provided the "faithful copy" that Seba requested for him
to see what kind of hydra it was. Seba subsequently circulated it widely
among connoisseurs of curiosities by making it the most dramatic illustra-
tion in the 1734 catalog of his Amsterdam collection, even though he had no
direct claim on the hydra. Shortly thereafter, young Linnaeus would declare
that it was probably the fabrication of monks-not unlike the "basilisk's
tongue in two pieces" and the "two basilisk skeletons in pieces" that the
abbot Matteo Priuli kept in his collection in Padua at the end of the seven-

Inventing Nature
teenth century. Debunking the hydra became part of Linnaeus's mythology
as a modern naturalist. 60 Yet what we miss in such an account is Linnaeus's
admiration for the hydra of Hamburg- a response to the fabrication of
nature that he shared with Seba and all the naturalists who preceded them.
The price on the hydra may have declined precipitously thereafter, but it
was still a work of art. 61

1. See especially Lorraine Daston and Katharine Park, Wonders and the Order of Nature
1150-1750 (New York: Zone Books, 1998); and Jean Ceard, La nature et les prodiges. L'insolite
au siezieme sieele (Geneva: Droz, 1977); Ottavia Niccoli, Prophecy and People in Renaissance
Italy, trans. Lydia G. Cochrane (Princeton, N.J.: Princeton University Press, 1990); and
William Burns, An Age of Wonders: Prodigies, Providence, and Politics in England, 1658-1727
(Manchester: Manchester University Press, 2001).
2. Nicolo Serpetro, It mercato delle maravlglie della natura overo storla naturale (Venice,

1653), n. p., "Introdutione per chi legge."


3. Tommaso Garzoni, Piazza universale dl tutte Ie professionl del mondo, ed. Paolo Cherchi
(Turin: Einaudi, 1996).
4. See Anne Goldgar's essay in this volume.
5. John Evelyn, The Diary of John Evelyn, ed. William Bray (London: Bickers and Son,
1906), Vol. 1,51 (Paris, 3 February 1644).
6. The role of gifts in early modern natural history is discussed in Paula Findlen, "The Econ-
omy of Scientific Exchange in Early Modern Italy," in Patronage and Institutions: Science, Technol-
ogy and Medicine at the European Courts, 1500-1750 (Woodbridge, u.K.: Boydell & Brewer, 1991),
5-24; Giuseppe Olmi, "'Molti amici in varii luoghi': Studio della natura e rapporti epistolari nel
secolo XVI," Nuncius 6 (1991): 3-31; and Brian Ogilvie, Observation and Experience In r:'arly Mod-
ern Natural History (Ph.D. diss., University of Chicago, 1997), esp. 8, 129 - 130, 241-242.
7. Hans-Olof Bostrom, "Philipp Hainhofer and Gustavus Adolphus's Kunstschrank," in
Orzgins of Museums: Cabinets of Curiosities in Early Modern Europe, ed. Oliver Impey and
Arthur MacGregor (Oxford: Clarendon Press, 1983), 91. For further discussion of this
approach to a cabinet of curiosities, see Pamela H. Smith, The Business ofAlchemy: Science and
Culture in the Holy Roman Empire (Princeton, N.J.: Princeton University Press, 1994).
8. See Mark Meadow's and Harold Cook's essays in this volume for more on Hans Jacob
Fugger and Jan Swammerdam.
9. John Ray, Observations Topographical, Moral, & Physiological; Made in a Journey Through
part ofthe Low-Countries, Germany, Italy, and France (London, 1673),237; and Philip Skippon,
An Account of a Journey Made Thro ' Part of the Low-Countries, Germany, Italy, and France, in
A Collection of Voyages and Travels, eds. A. and S. Churchill (London, 1752), Vol. 6, 565.
10. Many of the essays in Cultures afNatural History, ed. Nick Jardine, James Secord, and

Emma Spary (Cambridge: Cambridge University Press, 1995), touch upon this theme.
I!. The commercial aspect of Columbus's interest in nature should be considered along-

side those elements described in Mary Campbell, The Witness and the Other World: Exotic
European Travel Writing 400-1600 (Ithaca, N.Y.: Cornell University Press, 1988); and Stephen
Greenblatt, Marvelous Possessions: The Wonder of the New World (Chicago: University of
Chicago Press, 1991).

320 Paula Findlen


12. Theriac was an ancient medicine, described in great detail by Galen and composed of
numerous ingredients that allegedly cured all manner of poisons. It was increasingly taken as
a preventive medicine in an era of frequent plague epidemics. See Findlen, Possessing Nature:
Museums, Collecting, and Scientific Culture in Early Modern Italy (Berkeley: University of Cal-
ifornia Press, 1994), ch. 5.
13. Patrick Geary, Furta Sacra: Thefts of Relics in the Central Middle Ages (Princeton, N.J.:
Princeton University Press, 1978), indicates the extent of the passion for relics whose value
escalated to such a degree that people stole as well as bought them.
14- Daston and Park, Wonders, 69, 74.
15. This marvelous passage is discussed in Ogilvie, Observation, 241.
16. Biblioteca Universitaria, Bologna (hereafter BUB), Aldrovandi, ms. 136, Vol. 9, C.5v
(Castelletti to Aldrovandi, Genova, 22 February 1579).
17. See, for example, Ulisse Aldrovandi, De piscibus libri V (Bologna, 1613),401. "De cen-
trine" describes fish he acquired from Castelletti, who called them "verae ac genuinae."
18. BUB, Aldrovandi, ms. 136, Vol. 9, c.129r (Castelletti to Aldrovandi, n.d.)
19. Archivio di Stato, Pisa, Universita, 530, c.2r (Spese occorse nel viaggio fatto da un sem-
plicista per ritrovare piante e minerali); 518 (16 April 1604). This material is also discussed in
Lucia Tongiorgi Tomasi, "Arte e natura nel Giardino dei Semplici: dalle origini alia fine del-
I'eta medicea," in Giardino dei Semplici: [,Orto botanico di Pisa dal XVI al XX secolo ed. Fabio
Garbari, Lucia Tongiorgi Tomasi, and Alessandro Tosi (Ospedaletto: Pacini, 1991) 162.
20. Ambroise Pare, On Monsters and Marvels, trans. Janis L. Pallister (Chicago: University
of Chicago Press, 1980), 128.
21. BUB, Aldrovandi, ms. 38, Vol. 4, f.46 (Antonio Anguissola, Piacenza, 12 April 1568).
The Italian original literally describes Master Leone as someone who "monta in bancho
spes so" - the origins of the word "mountebank."
22. The fascinating story of Leone Tartaglini has been reconstructed in Achille Forti, "Del
drago che si trovava nella Raccolta Moscardo e di un probabile artefice di tali mistificazioni:
Leone Tartaglini da Foiano," Madonna Verona 8 (1914): 26-51; and idem, "II Basilisco esisente
al Museo Civico di Storia Naturale a Venezia e gli affini simulacri finora conosciuti. Contrib-
uto alia storia della ciarlataneria," Atti del Reale Istituto Veneto di scienze, lettere ed arti 87, part
2 (1928 - 29): 225 - 238.
23. Skippon, Account of a Journey, 517. Compare with Ray, Observations.
24- Lodovico Moscardo, Note overo memorie del Museo del Conte Lodovico Moscardo nobile
veronese (Verona, 1672), 235. Moscardo called them "ciurmatori, 0 Zaratani." Achille Forti
makes the interesting suggestion that the spelling of the last word indicated a kind of charla-
tan who came from Dalmatia into northern Italy so I have indicated this possibility in the
translation. Forti, "Del drago," 30.
25· Conrad Gessner, Historia animalium (Tiguri, 1558),4,945.
26. Mario Cermenati, "Francesco Calzolari e Ie sue lettere all' Aldrovandi," Annali di
botanica 7 (1908): 48 (Verona, 16 December 1571).
27. Ray, Observations, 27, 219.
28. BUB, Aldrovandi, ms. 136, Vol. 5, f. 179r. This episode is also discussed in Conor Fahy,
Printing a Book at Verona in 1622: The Account Book of Francesco Calzolari Junior (Paris: Fon-
dation Custodia, 1993), 20.
29. For more on this subject, see Pamela Smith's forthcoming book on artisans and science
in early modern Europe.
30. Girolamo Cardano, De rerum varietate, in Opera Omnia (Leiden, 1663), Vol. 3, 342. Also
discussed in Daston and Park, Wonders, 167.
31. Leonardo da Vinci, The Notebooks of Leonardo da Vinci, ed. Irma A. Richter (Oxford:
Oxford University Press, 1952), 167. The original is in ms. A, f. 20r.

Inventing Nature 321


32. Colin Eisler, Durer's Animals (Washington, D. C.: Smithsonian Institution Press, 1991),
3II -3 I2.
33. This formulation is found in Daston and Park, Wonders, 74, in their discussion of the
medieval treasury. Since the treasury was the original location of many hydras and basilisks,
it seems all the more fitting.
34. Conrad Gessner, Nomenclator aquatilium animantium. leones animalium (Tiguri, 1560),
362-363. Compare with Conrad Lycosthenes, Prodigorum ac ostentorum chronicon (Basel,
1657 ed.), 538-539. For further discussion of the religious significance of the hydra, see Myth-
ical Beasts, ed. John Cherry, (London: British Museum Press, 1995),20,35-36.
35. BUB, Aldrovandi, ms. 21, Vol. 4, c. 89v.
36. Ulisse Aldrovandi, Serpentum et draconum historiae libro duo, ed. Bartolomeo
Ambrosini (Bologna, 1640),387. See the second hydra, which is evidently the hydra of Fer-
rara: "Hydra septiceps Equitis de Corneto affectatoris olim Sereniss. Ducis Ferrariae."
37. Edward Topsell, Historie of Four-Footed Beastes (London, 1607),202.
38. Skippon, Account of a Journey, 565. "Zennon" is probably Giacomo Zanoni, a well-
known Bolognese apothecary whom many collectors visited in the mid-seventeenth century.
See also John Evelyn's account of a fifteen-headed hydra in the Villa Ludovisi in Rome that
he saw in 1645; (The Diary ofJohn Evelyn, ed. E. S. de Beer [Oxford: Clarendon Press, 1996,
1951], Vol. 2,391).
39. Pierre Belon, L'histoire naturelle des estranges poissons (Paris, I55I),I8r. For a general
discussion of fake animals, see E. W. Gudger, "Jenny Hanivers, Dragons and Basilisks in the
Old Natural History Books and in Modern Times," Scientific Monthly 38 (June 1934):
5II - 523; Richard Carrington, Mermaids and Mastodons: A Book of Natural and Unnatural His-
tory (New York: Rinehart & Company, 1957); and Peter Dance, Animal Fakes and Frauds
(Berkshire, UK: Sampson Low, 1976).
40. Gessner, Historia animalium, 945. Compare with Maximilian Misson, Nouveau voyage
d'Italie, 5th ed. (La Haye, 1731), Vol. I: 161-162.
41. See Daston and Park, Wonders; and Niccoli, Prophecy.
42. Carrington, Mermaids and Mastodons, 69-70. I have modified Carrington's translation
of the passage in order to conform with other translations of "raia" as "ray."
43. Pare, On Monsters and Marvels, 109.
44. Cardano, De rerum varietate, 343.
45. Eisler, Durer's Animals, 265, 321. An excellent example of a carved dragon can be found in
the Museo degli Argenti in Florence, part of the Medicis' famous collection of worked objects.
46. Aldrovandi, De piscibus, 437,443-444- I have wondered if the sea-eagle might not be
Castelletti's bulging-eyed fish? The most detailed discussion of the sources of Aldrovandi's
imagery can be found in Erminio Caprotti, Mostn~ draghi e serpenti nelle silografie dell'opera di
Ulisse Aldrovandi e dei suoi contemporanei (Milan: Gabriele Mazzotta, 1980).
47. For a classic description of the basilisk, see Lycosthenes, Prodigorum, 22; also Jacques
Grevin, De venenis libri duo (Antwerp, 1571), ch. 18.
48. Aldrovandi, Serpentum, 364.
49. Pier Andrea Mattioli, De i discorsi di M. Pietro Andrea Matthioli ... Nelli sei libri di
Pedacio Dioscoride Anazarbeo (Venice, 1585 ed.), 1526-1527,
50. Joannes Jonstonius, Historia naturalis de serpentibus libn II (Amsterdam, 1657),34-
51. Benedetto Ceruti and Andrea Chiocco, Musaeum Franc{isct) Calceolarii Iun{ioris}
Veronensis (Verona, 1622),90-91.
52. Moscardo, Note, 235·
53. Misson, Nouveau voyage, Vol. I, 161.
54. Francisco Hernandez, Rerum medicarum novae hispaniae thesaurus seu plantarum ani-
malium mineralium Mexicanorum Historia ex Francisci Hernandez, ed. Johan Faber (Rome,

322 Paula Findlen


1649), 818. This text is discussed in greater detail in Silvia de Renzi, "Herodotus and the
Microscope: Investigating Dragons in Seventeenth-Century Rome" (unpublished paper). I
thank Dr. de Renzi for providing me with a copy.
55. Lorenzo Legati, Museo Cospiano (Bologna, 1677), 81.
56. Athanasius Kircher, Mundus subterraneus (Amsterdam, 1678), I03-108.
57. Filippo Bonanni, Musaeum Kircherianum (Rome, 1709), 270; Giorgio de Sepi, Romani
Collegii Societatis Jesu Musaeum Celeberrimum (Amsterdam, 1678), 27. He concluded: "They
are deformed rays and putative basilisks and, in the author's opinion, made by art and not by
nature." There is no doubt that this opinion was also Kircher's, suggesting that he shared
Aldrovandi's view of the distinctions between true dragons and false basilisks.
58. Antonio Vallisnieri, Opere fisico-mediche (Venice, 1733), Vol. 3, 370. Vallisnieri"s biog-
rapher C. Lodoli also discussed how he showed the basilisk to "reveal the monstrous decep-
tions perpetrated in other museums where much was made of miraculous works of nature
such as basilisks, fabulous hydras, petrified bread and fungi and other similar nonsense."
Ibid., Vol. 1, LVI. This latter passage is discussed in KrzysztofPomian, Collectors and Curiosi-
ties: Paris and Venice, 1500-1800, trans. Elizabeth Wiles-Portier (London: Polity, 1990), I04.
59. Albert Seba, Locupletissimi rerum naturalium thesauri accurata descriptio, et iconibus arti-
ficiosissimis expressio, per universam physices historiam (Amsterdam, 1734), Vol. 1, table I02, 159.
60. Pomian, Collectors, I03. On the hydra of Hamburg, see Colin Clair, Unnatural History:
An Illustrated Bestiary (London: Aberlard-Schuman, 1967), 2II-212, 235; and Dance, Animal
Fakes, 33-36. For an excellent introduction to Linnaeus, see Lisbet Koerner, Linnaeus: Nature
and Nation (Cambridge, Mass.: Harvard University Press, 1999)'
61. Samuel Butler's Hudibras recounts a less well-known story - a send-up of the whole
cult of fictitious nature-about a dead rat that James Bobart found in the Oxford physick
garden in the late seventeenth century and turned into a dragon to see who would believe
him. Yet after he revealed the joke, the dragon remained in the cabinet of curiosities associ-
ated with the Oxford anatomy theater, "looked upon as a masterpiece of art." In Dance,Ani-
mal Fakes, 59.

Inventing Nature 323


Nature as Art
The Case of the Tulip

ANNE GOLDGAR

" W h e n Homer sang in ancient times at Corinth, no one lis-


tened to his verses. In our own era in Paris, Poussin earned
too little to live." These lines bewailing philistinism, pub-
lished by Nicolas de Valnay in r669, were written, rather surprisingly, in
defense of the tulip. Valnay, controleur of Louis XIV's household and a
member of a loose group of curieux devoted to flowers, expressed surprise at
the preference some felt for other curiosities, such as paintings, medals, or
porcelains. Look at such things as long as you like, he wrote, but you will
always be looking at the same thing. Not so with the wonderful annual
variety of flowers, of which the tulip was the queen. The beauties of paint-
ing, moreover, are all in design, execution, and color; but "I challenge the
entire Academie de Peinture to imagine flowers better than natural ones, to
execute them in complete perfection, or ever to approach the colors of
Flowers." If you own a painting, you will always have only one, but bulbs
have the advantage of multiplying themselves. The consequence of this
(although Valnay did not put it this way) is social: you can give a rare flower
to a friend and yet still keep the same thing, not a copy, for yourself. These
arguments against painting, Valnay said, could also be made against
medals, porcelains, and other fashionable rarities: "when reason is com-
bined with taste, beautiful flowers will hold the first rank among the plea-
sures of sight."l
Valnay's view that the times were too "blind and insipid" for flowers like
tulips-that they suffered from the poor taste of the public, as painting and
poetry had also sometimes suffered-is something of an exaggeration. 2
Although the tulip would not find again such passionate advocacy as it did
in the period leading up to the tulipmania of r634-37, it enjoyed a healthy
popularity, particularly among a circle of professional men, merchants, and
gentry, both in France and in equivalent social groups in other European
countries such as the Netherlands, England, Italy, and the German states.
But what is interesting in Valnay is not chiefly the strength of his views,
views expressed by other enthusiasts over the years, such as Crispijn van de
Passe in the Netherlands, Jean Franeau in Flanders, Sir Thomas Hanmer in
England, or the Sieur de La Chesnee Monstereul in France. The interest lies
rather in Valnay's treatment of tulips as works of art. This essay will explore
the complex of ideas and attitudes that allowed such a comparison, and, fur-
ther, the social consequences that sprang from making it.
Grappling with the relationship between art and nature was hardly new
in the seventeenth century. As Lorraine Daston and Katharine Park have
delineated in a richly suggestive chapter of their Wonders and the Order of
Nature, the opposition between the forces of nature and art was an ancient
paradigm, still crucial in the seventeenth century even as natural philosophy
began first to break it down and then to change its terms. 3 From the days of
Zeuxis and Parrhasius, artists have made a strong case for the superiority of
their craft to the creative powers of nature. 4 With the strengthening view in
the Renaissance that the artist's ingegno was akin to that of the creating God,
the view of art as surpassing nature became ever more influentia1. 5 Early
modern comments about floral still life make this clear. Painters of flowers
were considered to improve on nature if they were talented at depicting tex-
ture, and, as Paul Taylor has pointed out, the desire to idealize the flowers in
paint and thus perfect imperfect nature was a powerful impulse for artists. 6
Thus the triumphs of Zeuxian grapes were brought up to date with verses
praising Daniel Seghers' painted roses. According to Huygens in r645,
"Nature as judge, conceded defeat in the contest:/The painted flower ren-
dered the real one a shadow."7
But art had not quite won this battle with nature, and Seghers could
prove just as potent an argument for the other side. Proponents of the glo-
ries of flowers, for example, also cited Seghers, but now to stress his sup-
posed inability to do justice to them in paint. Flower breeders John Rea and
his son-in-law Samuel Gilbert favored real tulips over their portrayal by
"Pater Zegers, a Jesuite in Antwerp, famous for painting flowers," Rea com-
menting of a tulip called the Agate Hanmer, "Her Native Beauties shaming
Art, I Once did that famous Jesuite try I To copy out her Majesty; But falling
short of his desire, I He left his Pencil to admire."g Despite artists' tradition
of challenging God and nature with their ingegno, in the seventeenth cen-
tury some voices, including, naturally enough, those of various botanists and
gardeners, continued to maintain that art could do nothing to imitate the
beauties and wonders of some natural objects.
The traditional division represented here, seen in the usual classification
of collections into naturalia and artificialia, was precisely one that the later
sixteenth and seventeenth centuries delighted in undermining. 9 There was a

Nature as Art
pleasure in blurring the boundaries between art and nature, to be seen in
artificial objects resembling natural ones, such as Palissy's ceramics or the
mechanical rainbows and songbirds to be found in the Villa Aldobrandini at
Frascati. 1O More important for our purposes are the many natural objects in
collections turned half into artificialia by gilding, etching, carving, or artistic
arrangement. Coconuts, ostrich eggs, or rhinoceros horns transformed into
reliquaries; nautilus shells etched and gilded into luxurious beakers; rein-
deer antlers fashioned into candelabra: all testified to the desire of artists and
collectors to intertwine nature with art. 11
A coconut carved with a biblical scene or a nautilus cup engraved with
the image of other shells and sea creatures not only illuminates the status of
objects that are half art and half nature. Such objects also demonstrate to us
a desire for art to conquer nature. A nautilus shell itself was sometimes not
enough; the hand of man had to alter it and beautify it as well. These atti-
tudes have been elaborated in recent work on the Kunstkammer and its
objects. What has not so far been examined by scholars, however, is a further
step along this path. Man could impress himself upon natural objects by
gilding, etching, and engraving; but there were also objects that, although
remaining entirely natural, yet were evisaged as art. The tulip is one such
object, fascinating and enigmatic precisely because of this special mixture of
art and nature. What, then, did it mean for Valnay to place it in the same
category as a painting or a piece of porcelain?
Even in 1669, when Valnay was writing, tulips remained exotic items.
They had arrived in Europe in the mid-sixteenth century, either through
seeds sent home by the imperial envoy Ogier Ghislain de Busbecq, as most
sources recount, or, possibly, through trade between Turkey and Italy,
France, and the Low Countries. The first tulip in Europe was described by
Conrad Gessner in De Hortis Germaniae (1561) as growing in 1559 in the
garden of Johann Heinrich Herwart in Augsburg. By the late sixteenth cen-
tury, botanists and collectors had developed a passion for tulips. Such collec-
tors included both professional men with an interest in plants, particularly
doctors and apothecaries, as well as a variety of elite groups ranging from
merchants to aristocrats. The special excitement generated by tulips
stemmed first from their foreign nature; they were prized along with the
other exotic bulbs that arrived in Europe, largely from the Ottoman Empire,
in the same period, including flowers such as the iris, crocus, and hyacinth.
But tulips were particularly valued because of their unpredictable and
exciting capacities for variation. Unlike most other flowers, tulips could
change from year to year in coloration and form, and the propagation of
tulips could present new forms never before seen. Already in 1597 Gerard's
Herball reported that to recount all the different sorts of tulips "would trou-
ble the writer and weary the Reader," and by the time of Rea's Flora of 1665

Anne Goldgar
we hear that "so numerous are the varieties, that it is not possible that any
one person in the world should be able to express, or comprehend the half of
them."12 These qualities led to high prices for many tulips and, eventually, to
a futures trade in the Netherlands. But even in countries unaffected by the
futures trade, the tulip remained a favorite flower until it was eclipsed by
the hyacinth in the eighteenth century.
Given the reigning issues about art and nature, the question of the
authorship of tulip varieties was a live one in the period. Who or what actu-
ally created the tulip? Many works on gardening alluded to tulips, and to
flowers in general, as small but perfect works of God, or of both God and
nature. For Petrus Hondius, whose poem Dapes Inemptae, ofde Moufe-schans
(1614) praised the garden of his patrons in Neuzen, gardening was a way of
honoring the name of God and his creation,u Similarly, Jean Franeau,
extolling the gardens of the gentry and aristocracy of the Southern Nether-
lands, compared a flower to a school or a beautiful book of which "the
author is this great God, who, as a schoolmaster, teaches the lessons in his
own words."14 But these protestations address a countervailing trend which
we have already seen in the field of painting or sculpture: the desire of man
to claim credit for himself. The commercial breeder Samuel Gilbert,
acknowledging in 1682 that man can playa role in gardening, hedged his
bets in a verse mainly devoted to the planning, shaping, and protection he
would give to a garden:

Assisting Nature by industrious Art;


To perfect every Plant in every part,
But not like some, whose crimes to rise so high
Boldly to pull down Heavens Deity.1 5

Gilbert's concern that an emphasis on man deprives God of his rightful place
speaks to the questions raised by those who believed that man could, indeed,
"assist nature."16
The actual assistance man could provide was minimal, but this relative
powerlessness was not acknowledged by contemporaries. We have seen that
it was the variety of tulips that made them so popular. This variety was
achieved either by cross-breeding or by "breaking" tulips through an aphid-
borne virus. Neither process was known in the early modern period. Gar-
deners had no knowledge of the sexuality of plants, so that any crosses which
occurred would have to take place by accident; and although there was some
speculation that beautiful tulips were diseased-"in the same way that a per-
son in agony turns different colors when through a contagious malady he
approaches death," wrote a derisive La Chesnee Monstereul in 16S4-this
was neither universally accepted nor understood. 17 Yet long experimentation

Nature as Art
had assured some botanists and gardeners that, if they were grown from seed
rather than bulbs, certain tulips were more liable to vary in color, although
such variation could take up to ten years. Jacques Garret, a Dutch apothecary
living in London, was described by John Gerard as having "undertaken to
finde out if it were possible, the infinite sorts by diligent sowing of their
seedes, and by planting those of his owne propagation, and by others received
from his friends beyond the seas, for the space of twentie yeeres, not being yet
able to attaine to the end of his travaile, for that each new yeere bringeth
foorth new plants of sundrie colors not before seene."18 Later annotators of
Dodonaeus's Cruydt-Boeck noted in 1618 that sowing seeds was an uncertain
business, and that in any case tulips could get worse as well as better;19 but the
very fact that sowing seeds could produce new tulip strains produced a con-
fidence that man could indeed use his art on the natural tulip.
Not surprisingly, much of the writing on the alteration of tulips focused
on the choice of seed. If gardeners could not actually predict the outcome of
sowing seed, at least they could try their art in choosing only the seed of
superior tulips, and by culling bulbs of flowers that did not meet their stan-
dards. The eighteenth-century diary of the Lancashire gardening enthusiast
Nicholas Blundell is full of notations such as "I Examain'd my Tulops and
marked some of the best of them to be preserved & the worst to be
destroyed."20 But the idea that man could control nature went further than
this. The aesthetic and commercial value of tulips, as well as the tedium of
taking ten years to grow them from seed-described by the Haarlem
breeder Nicolas van Kampen as "unpleasant" and "useless"21-1ed early on
to more direct attempts to intervene by art into the processes of nature.
Some of these promised shortcuts to beautiful flowers, it is true, remained
close to natural processes. The theory that tulips changed color through dis-
ease led to experiments in weakening flowers so that they would fall sick
more easily. John Rea suggested in 1665, for example, that "more vulgar"
tulips might be dug up just before flowering and laid in the sun "to abate
their luxury, and cause them to come better marked the year following"; a
yearly alternation of good and poor soil was thought to have the same
effect. 22 Rea assured the aspirant gardener that such methods would produce
tulips that "might be taken for much better flowers than they are, especially
if a new name be put upon them, as some flower-merchants about London
use to do."23
Other procedures to alter tulips, however, were much more self-consciously
a form of art. Like an engraver carving designs on a nautilus shell, garden-
ers set about by intrusion to change their flowers. It was said that cutting
two bulbs in half and sticking them together would produce a cross-breed;
that new varieties of tulip with exotic colors could be bred by steeping the
bulbs or seeds in colored water, ink, paint, or even "mixing a number of

Anne Goldgar
ingredients with pigeon dung" and burning the ground with it. 24 Many of
our accounts of such interventions come, it is true, from those casting asper-
sions on these methods; John Evelyn, for example, warns us to "trust little by
mangonisme, insuccations, or medecine, to alter the species, or indeed the
forms and shapes of flowers considerably," and John Parkinson dismisses as
"meere tales and fables ... the many rules and directions extant in manie
mens writings to cause flowers to grow yellow, red, greene or white, that
never were so naturally.... [W]hen they come to the triall, they all vanish
away like smoak."25 Comments of this sort indicate, however, the prevalence
of such views; and contemporary gardening books, such as Giovanni Battista
Ferrari's Flora, seu de jlorum cultura of 1633, took such advice seriously.26
Moreover, even those ridiculing these methods did not necessarily believe
that there was no art to changing the appearance of tulips. La Chesnee Mon-
stereul, who took time to ridicule the Rouennais who burned up his entire
garden with a fire made with pigeon dung, affirmed flatly that, although
this particular method was laughable, efficacious means of producing new
tulips did exist. "It only remains for me to discuss whether by Art one can
embellish those which have not yet attained their peak of perfection .... I
have no difficulty in saying that one can, and that without doubt by Art they
can be rendered capable of changing into something better."27
Not only did gardeners, then, give their endorsement to the idea that
man's art could triumph over the processes of nature, but they consciously
conceived of what they were doing in these terms. La Chesnee Mon-
stereul's discussion of the transformation of tulips plunges straight into the
topic. "There is no doubt that it is not only in this point that Art surpasses
nature, of which we can see the effects, but in many other things which
[Nature] begins, & which Men complete & perfect through their indus-
try."28 Florists and gardeners, according to other writers, "know how to aid
nature by an artifice which industry and time has taught them"; flowers
are "natures Choicest dishes, advantag'd by Art."29 These gardeners were
not mere spectators to the wonders of nature, but active participants in
changing it, for, as Jan van der Groen wrote of gardens in 1669, "nature
can, through art, be shifted, decorated, put into good order, and made
ornamental and pleasurable."30
Thus even those pleading, as Samuel Gilbert did with such concern, that
flowers were the work of God had to admit that, ultimately, "our Art, with
Madam Nature joyn[s]."31 But the power to control nature-a power of
which tulips were, ironically, a poorer example than industry or
agriculture-was not the end of the construction of tulips as artifacts. The
language used by gardening writers and the names given to tulips by grow-
ers show a distinct mental association between the flowers and both art and
craft. In floral still life, it was said that the technically gifted made roses look

Nature as Art
like silk and tulips like leather,32 but the same kind of comparisons were
applied not only to paintings but to the flowers themselves.
The trope that tulips outdid the work of any painter-"it is impossible
for painters and dyers to imitate the colors of them"33-demonstrates a cog-
nitive link between the flowers and the art of painting. But the mention of
"dyers" in this same passage from a 1697 gardening treatise points us toward
the wider associations of tulips in the period. There were constant references
to the flowers as part of a wider world of man-made luxury objects. That
both paintings and other artificialia should provide comparisons with tulips
is perhaps not surprising, since painting had such strong connections with
craft and with craft guilds, in the Netherlands at least. In the chief towns of
Holland, painters and sculptors were members of the same guild as crafts-
men such as embroiderers, glassworkers, and goldsmiths. Although painters
increasingly came to emphasize the dignity of their profession, they did not
seek in this period to emancipate themselves from the structure of a craft
guild; indeed one artist who also was involved in the tulip trade, Frans
Pietersz. de Grebber, was reported to "get by nicely on embroidery" when
his skills as a portraitist were not in demand. 34 Given both the rarity and the
price of tulips, however, a more important explanation for their linkage with
costly crafts is surely the role of the luxury trades in the maintenance of sta-
tus. Like visible material wealth, tulips were precious and costly collector's
items. The close associations of tulips and luxury are evident; the names and
descriptions of tulips remind us continually of a shiny, varied, patterned
world of cloth, enamel work, and polished stone.
Cloth, carpets, and embroidery are among the earliest images applied to
tulips and the gardens containing them. Gardens decorated with a variety of
colors reminded viewers of a tapestry or carpet; Marie de Brimeu, Princesse
de Chimay and a friend of Clusius', referred in 1591 to the garden he had
stocked for her with plants sent from Frankfurt as "the riches of your tapes-
tries [which] truly surpass by far those of gold & silk, as nature surpasses arti-
fice."35 John Parkinson was still musing in 1629 on such comparisons, and on
the possibilities for garden design using tulips. "[A]bove and beyond all oth-
ers, the Tulipas may be so matched, one colour answering and setting of
another, that the place where they stand may resemble a peece of curious
needle-worke, or peece of painting."36 That this was not merely a metaphor
is evident from the Iardin du Roy Tres Chrestien Henry IV of 1608, a collection
of engravings of plants in the king's gardens. The author, Vallet, far from
being a gardener, was a professional embroiderer, and the accompanying
verses in praise of the book make much of his abilities with the needle. His
designs were evidently intended at least in part as patterns for embroidery.37
Not only whole flower beds, but also individual tulips called to mind the
luxury of elegant cloth. Clusius, in first describing his tulips in his Rariorum

Anne Goldgar
plantarum historia, resorted repeatedly to images of silk shimmering with
two colors, such as a silk with a golden warp and a red weft, or a silver silk,
made in the same way and known to the Germans as Silberfarb. 38 Such
images quickly appeared in the vernacular. Among the earliest names to be
given to tulips (the first flowers of which individual cultivars were named)
were "Goude Laeckens" and "Silver Laeckens," gold cloth and silver cloth,
appearing in other countries as "drap d'or" and "drap d'argent" and as
"Cloth of golde" and "Cloth of sylver."39 Other names referring to cloth
included "Saey-blom" (say-flower). While says were not the most elegant of
cloths, the economy of Holland depended on them; and the other materials
could not have been more redolent of luxury. No cloth was more expensive
or prized than silk embroidered or shot with gold or silver thread-silver
tissue was the standard material of bridal gowns at the European courts 40_
and silk and satin clothing was similarly confined to the elite. 41 Like embroi-
dery, which bespoke an ability to pay for goods whose manufacture was
highly labor-intensive, clothing or furnishings made of costly materials were
effective signals of leisure and high station. Fine textiles, because they
demonstrated marvelous workmanship, were also commonly to be found in
collections and Kunstkammern.42
It was natural for an expensive flower like the tulip to be compared with
such goods, but, as Clusius's more specific descriptions indicate, it was not an
idle comparison. The petals of the tulip were also thought actually to resem-
ble silk, satin, or velvet. One Dutch gardening book recommends matter-of-
factly that a good source of seed for the best tulips was a flower that was
"Satijn-agtig" (satinlike),43 and in the more florid prose of works dedicated
to praising the tulip, references to elegant cloth abound. To take only one
example, Jean Franeau's 1616 ode to the tulip, the Iardin d'Hyver au Cabinet
des Fleurs, intended to remind the growers and collectors of the aristocracy
and gentry in the Southern Netherlands of their treasures during the winter
months, is liberally peppered with references to embroidery, silk, taffeta, sil-
ver cloth, and silver needlework. Tulips were "dressed" in rich "mantles"
which, far from being natural, were "full of artifice." The tulip called the
Duc van Tholl was said to be the work of a skilled tailor: clothing worthy of
a great prince or duke, so richly was it embroidered in gold. This was the
work of nature, Franeau wrote, but also of "les grans," who cultivated tulips
in their gardens. 44
The luster of tulips, if it did not come from shiny satin or silk, was also
said in many texts to be enamel work or vermeil, a metaphor that was mixed
happily with the many references to clothing. 45 But the other favorite com-
parison for the tulip, with its veined and streaky markings, was polished
stone such as marble and agate. Again, the names of tulips give us a hint of
how they were perceived. A common designation in the Netherlands was

Nature as Art 33 1
ghemarmerde, marbled, and names such as Ghemarmerde de Goyer or Ghe-
marmerde Liefkens became usual. In French, tulips were "marbrees" or
"jaspees," and whole classifications of tulips based on this comparison grew
up. Besides the Morillons, or rough emeralds, a major class of French tulips
(which was adopted also in English nomenclature) was the Agate. By mid-
century there was an abundance of names such as Agate Morin, Agate
Guerin, or Agate Picot, generally adding the name of the cultivator to the
designation of Agate; La Chesnee Monstereul mentioned fifty-five Agate
tulips in his list of 1654.46 John Rea gives us a good picture of the association
of the two objects; tulip petals, "warmed by the Sun, open and change into
divers several glorious colours, variously mixed, edged, striped, feathered,
garded, agotted, marbled, flaked, or specled, even to admiration."47 Like
tulips, stones such as agates and marbles were prized for their variation in
color and their attractive veining. Polished and worked, they formed an
ornament to buildings, were transformed into pietra dura furniture, or
formed objects for the Kunstkammer or cabinet. We find special collections of
agates in the period, such as that belonging to the goldsmith Antoine Agard
of Arles, not to mention the agates and marbles in the cabinets of tulip lovers
such as Christiaan Porret in Leiden or Bernardus Paludanus in Enkhuizen. 48
The designation of tulips as ghemarmerde, marbre, or "marbled"-
phrased as though they had actually undergone a process-raises, moreover,
the possibility that the connection was actually with marbled paper rather
than marble itself. Paper with the swirling patterns of marble had arrived in
Europe around the same time as the tulip and from the same country of ori-
gin, and, like tulips, it became associated with the Dutch in the seventeenth
century. Like real agates, marbled paper was valued by collectors and can be
found in collections, such as that of Paludanus; Pierre de l'Estoile records
instances in 1608-09 of giving a fellow Parisian collector sheets of marbled
paper because "I know he is a collector like myself."49 The resemblance
between the paper and stone was remarked upon early, and one of the old-
est patterns was known, like the tulip, as Agate. 5o Interestingly, Evelyn uses
the word "Pennaches (as the French call it)" to describe the paper's design;5l
panache was the usual word for the similar patterns on tulip petals, with a
broken tulip being known as panachie. Most suggestively, Anna Maria von
Heusenstain, who frequently wrote Clusius from Vienna asking for flowers,
reported in 1591 that a tulip she still lacked was called "Turkish paper" ("das
tirckisch papir").52 Although we cannot be conclusive about the connections
between marbled paper and tulips, it is clear, at least, that all these objects-
flower, paper, and stone-occupied a similar space of aesthetic appreciation
and social and intellectual exchange.
The praise of the tulip, indeed, is reminiscent of the kind of aesthetic dis-
cussion we find about paintings in the same period. Although the Dutch

Anne Goldgar
were, as Seymour Slive has put it, "unusually inarticulate" about their art in
the seventeenth century,53 with only a handful of critical writings to help us
enter the minds of artists, the same sorts of words to describe beauty (aerdigh,
fraey, schoon) and the same values of harmony of form and clear, bright col-
ors were applied both by a writer on painting like Karel van Mander in 1604
and commentators on the beauty of tulips.54 Both annotations to Dodonaeus'
chapter on tulips and Karel van Mander on paintings, for example, spoke of
netticheyt, the engagement of the eye, in the case of painting in a smooth ren-
dering of surface; the tulip was said to be "much honored in all lands for the
netticheyt of its petals."55 But perhaps the most important aesthetic comment
common both to tulips and art was the value of verscheydenheydt: variety.
The excitement of diversity dated back at least to Pliny,56 and formed an
important principle for a range of arts in the Renaissance, from painting to
music to literature. 57 For Karel van Mander it was a crucial concept; art
should found itself on nature, and "nature is beautiful through variety," both
the variety of colors and of forms and attitudes. 58 He compared a painting
with a field of flowers which draws the eye, like the honeybee, to dart from
one point to another, eager not to miss any of the sweetness. 59 And as we
have already seen, it was the variation of tulips, from year to year, from
mother bulb to offset, from bloom to bloom, which made tulips so much
more compelling than other flowers: "the more various, the more beauti-
ful." 60 Again, the appreciation of tulips falls within the same aesthetic uni-
verse as the appreciation of art.
While these similarities might be put down to the paucity of critical lan-
guage available to commentators, the very fact that a critical vocabulary was
being applied to tulips at all demonstrates another facet of their relationship
to works of art. In a period when critical discussion and connoisseurship of
painting among laymen was becoming more common, alongside the rise
of the virtuoso collector of artificialia, we also find an increasing application
of specific standards to tulips.61 Whereas in the late sixteenth and early sev-
enteenth century writers were relatively undiscriminating about tulips, find-
ing all a source of wonder, it did not take long for writers and enthusiasts to
develop a hierarchy of varieties and a set of characteristics of a good flower.
In a typical case, Peiresc complained in 1626 of seven or eight boxes of tulips
he had received from his brother's estate: "I found none which were perfect
of their sort. The drap d'or and drap or taille d'argent had so much red in
them that the rest seemed nothing of any worth."62 In guides for gardeners,
some of whom might not know "either what to choose, or what to desire,"
as Parkinson put it, we find descriptions of the best tulips; La Chesnee Mon-
stereul provided a chapter with the title "What the Tulip must be, both in its
Colors, Panaches, and in its Form," and similarly Val nay and Van Kampen
each spent some pages outlining such qualities. 63 Valnay instructed his read-

Nature as Art 333


ers that the most beautiful tulips were the "bigeares" (stripes of brown, red,
violet, or other colors on a yellow background), the more nuanced the better,
and the stripes or panaches had to be distinct; the colors should be as far
from red as possible, although "red ones on a white ground are not to be
rejected." He defined the proper form of the flower, a description of the best
color for the base, and the appropriate size and color of the stamens. 64 Some
of these standards might be thought to be fixed, as they resulted from obser-
vation of the best traits for breeding. Since the purpose of breeding was
beauty, however, the apparent "science" of such definitions in fact demon-
strates how the changing vision of floral excellence was an arbitrary matter
of aesthetics. And that the standards changed over time links tulips not only
to painting but also back to clothing and fashion. Rea, in 1665, remarked
that the flowers described by Parkinson in 1629 had "by Time grown stale,
and for Unworthiness turned out of every good Garden"; the popular flow-
ers in his day, tellingly, were known as Modes. 65 But for a certain core of
tulip lovers whom we can consider the connoisseurs, the standards hardened
into what would ultimately be called "florists' flowers," with the specific
requirements for shape, texture, color, and markings enforced by the devel-
opment of flower shows in the eighteenth and nineteenth centuries. 66
That the resemblance of tulips to art could extend to connoisseurship has
important implications. As we have seen, the idea that art and nature were
potentially oppositional is something of a cliche. So, indeed, is the suggestion
that the line between these two could become blurred in the early modern
period. Scholars who have written on these subjects have also approached
them by commenting on artistic objects that look like nature (Palissy's
ceramics) or natural objects that look like art (figured stones with apparent
landscapes or animals in their markings). But with tulips, as I have argued,
the tulip itself is objectified. Although the tulip resembled works of art, it
did not have to, since it was in fact art itself, an item that, it was thought,
could be created in its entirety by man. Moreover, once made, the tulip was
the subject of criticism and discussion. Thus not only was the tulip a real art
form, but it was an art that had further ramifications. The tulip as art pro-
moted sociability.
From the beginning, the increasing interest in exotic flowers, of which
tulips were the chief, fit into existing exchange relationships and engendered
new ones. Dodonaeus made it clear in his Cruydt-Boeck that in the Low
Countries the tulip was "only to be found in the gardens of the connoisseurs
[liefoebbers J,"67 and it is evident that many of these connoisseurs knew each
other and exchanged flowers. The botanist Carolus Clusius is an obvious
example; his correspondence includes much evidence of the trading of bulbs
and seeds, even with people he had never met. 68 Less professional plant-
lovers, such as Petrus Hondius, also clearly were involved both in the breed-

334 Anne Goldgar


ing of tulips and their exchange, and similar relationships are also obvious
among collectors in other countries. Peiresc, for example, reported sending
tulips to Antelmy, conseiller at the Parlement at Aix; Sir Thomas Hanmer's
correspondence shows him sending "rootes" to friends such as John Evelyn
and the parliamentary general Lord Lambert; while the anonymous accounts
of an English garden from 1638-39 included both bulbs sold by the owner to
"Mrs Rous" and a large number exchanged with a "M' Blackley" and a "M'
La candle": "Amanicis de Pari for Olivandesburge/Generall Gouda for Noris
morilio[n]/Satanee for Gulamp van Rhine/M' Jullien for Semp[er] Lond-
inu[s]."69 Such relationships could become institutionalized in clubs such as
the Society of Florists of Norwich, known to have existed since at least 1631,
when Ralph Knevet's play Rhodron and Iris: A Pastorall was presented at their
feast on the third of May.7D Such florists were not professional growers-the
word came to that meaning only later-but rather were lovers of flowers
who grew them for their beauty rather than any usefulness. 71
Whether or not tulip lovers were involved in formal societies, tulips
became a route of sociability for them. Because of their bulbs, John Rea wrote
in 1665, tulips could be regarded as "transferable favours from one Florist to
another, aptly conveyable (the seasons considered) many miles distant."72 The
fellow feeling that was engendered by these exchange relationships was an
important part of the interest in tulips. La Chesnee Monstereul devoted a
chapter of his treatise on the flowers to pleading that tulips should not be
allowed into the hands of the many. "If Tulips were made common, that
would remove the most praiseworthy interaction to be found among Men, &
would deprive them of the most sweet society which has ever existed among
gens d'honneur. How much company does their rarity give to curious wits?
how many agreeable visits? how many sweet conversations? & how many
well-founded discussions? Certainly this is the sweetest life in the world."73
Such interchange could be the foundation both of society and of a good col-
lection, so that the anonymous Valnay could write of himself that "through
purchases, exchanges, and accommodations, the principal stock of the only
beautiful Tulips are at present in the hands of M. de Valnay" and two other
persons, all officers at the Louis XIV's court. 74 In his pursuit of the tulip,
someone like Valnay, breeder and curieux, was both artist and collector.
Valnay listed only a few curieux and breeders in Paris who had experi-
mented on tulips, attempting to get more beautiful varieties as they studied
each other's productions. But the best flowers remained rare, and, according
to his account, one seventeenth-century collector, Lombard, who had the
best stock, would not exchange or sell any of his bulbs or seed, "which
increased the desire [for them] and the price."75 La Chesnee Monstereul's
argument that tulips should remain confined to a few is important to the sta-
tus of these flowers as collectibles or as art. The sociability of connoisseur-

Nature as Art 335


ship is based on expertise about the rare and costly; if everyone could com-
ment, or everyone could own these objects, there would be no point to the
connoisseurship. Thus although he declared that it was possible to change
tulips by art-in his case, essentially alchemy-he would not reveal the for-
mula. "I do not want the secrets of Divinity to be known other than by
Sages, so that they are not profaned by the vulgar; it being a certain thing,
that according to the sentiment of the learned [Roger1Bacon, he who reveals
the Mystical diminishes & reduces its majesty." He provided a mysterious
and cryptic recipe, saying "those who have eyes & ears will see & hear"; no
one without the requisite learning would be able to understand his formula
and thus join the exalted company of the artists and the connoisseurs. 76
But the similarity between tulip lovers and collectors of art and curiosities
does not end with the behaviors of connoisseurship and exchange. In many
cases, these groups did not simply resemble each other, but were actually the
same. Tulip buyers were themselves collectors of art. Pierre Morin, one of
the great professional florists of the first half of the century, was also "one of
the most intelligent men in the world in all these rarities," including objects
of art as well as shells, butterflies, and other naturalia. 77 But ordinary buyers
of bulbs, or owners of a few tulips, prove on investigation to have been active
in the purchase of artworks. A comparison of the names of tulip buyers in
Amsterdam up to r640 with those known to have bought art at auction sug-
gests that, in Amsterdam at least,78 among the main groups to take an inter-
est in tulips in the period were members of overlapping networks of art
collectors. The merchant Abraham de Casteleyn, for example, was involved
in both breeding and purchasing tulips, but also bought art at auction and
was a considerable collector of naturalia and artificialia. At his death in 1644
he left not only his cabinet of curiosities, but a carefully arranged collection
of tulip bulbs whose inventory ran to thirty-nine pages. 79 Similarly, men
such as Lambert Massa, Adam Bessels, Jacob Abrahamsz. van Halmael, and
Jan Hendricksz. Admirael, who were mainly merchants, purchased paint-
ings and prints at auction and even (in the case of Massa and Bessels) were
connected to important families of art patrons, but were also involved in the
purchase of tulip bulbs. sO These are only some of the examples of the
crossover between tulip and art collectors in Amsterdam in the first part of
the seventeenth century.
The fact that many of these transactions took place during the financial
craze of 1634-37 of course raises questions about the role of commerce in
this relationship between tulips and art. It is hard to determine exactly why
people bought tulips, even if they were collectors of art; and of course there
is a commercial element even to collecting, however much the values of
exchange and civility also entered into it. For tulips or for paintings, the
question remains whether the reason for purchase was decoration, collect-

Anne Goldgar
ing, or profit. 81 The remark of Val nay's cited above, pointing to the increased
desirability of rare tulips in Paris precisely because they were rare, reminds us
that commerce was a factor in the love of "artistic" tulips from the beginning.
Clusius, involved in exchange relationships with botanists, apothecaries, and
scholars across Europe, disapproved of the commercial activities of the "rhi-
zotomi" (root cutters), his name for the (mainly French) traveling sellers of
bulbs and exotic plants, at the same time as he made purchases from them.
One of his arguments was that of connoisseurship. Whereas he had always
maintained a garden to supply his friends, now, in 1594, "merchants, even tai-
lors and shoemakers and other petty people are involved, through the hope of
profit. For they see that rich people will sometimes give out handfuls of
money to buy some plant or other which is sought because of its rarity, so that
they can preen themselves in front of their friends because they own it."82 The
implication is that exchange, not purchase, is the proper way to obtain bulbs,
and that those appropriate to own such bulbs should remain select. But it is
evident that, well before the speculative craze of the 1630s, beauty was in the
eye of the purchaser. Commentators on Dodonaeus wrote in 16r8: "In this
country men love most the flamed, winged, speckled, jagged, or snipped and
the most strongly variegated: and they will pay the most, not for the most
beautiful or the finest, but for the rarest to be found, or those owned by only
one master; these can fetch high prices."83 And Rea's remarks in r665, quoted
above, that alternating soil can make "vulgar" tulips appear "better flowers
than they are, especially if a new name be put on them," reminded the reader
of a commercial advantage to the application of art. 84
These comments point us both to the tulipmania of the 1630S and beyond.
As trade in tulips heated up, some buyers, at least, wanted whatever would
command the highest price. The fact that by this time some tulips were sold
as futures and could change hands several times before they came to flower
makes it plain that at this juncture tulips were valued by some only for their
profitability. But the emphasis of the Cruydt-Boeck on beautiful features as
well as on the price that rarity brings reminds us that beauty and profit can
go hand in hand. After the crash in February 1637, tulips continued to be a
favorite flower and continued to command good prices. Indeed, the consol-
idation of the bulb-growing industry around Haarlem dates from the period
after 1637, with dynasties of commercial growers settling into what until
then had been sometimes a temporary or part-time occupation. The scion of
one such dynasty, George Voorhelm, was in 1752 still making arguments
from connoisseurship to justify his high prices for hyacinths, by then the
flower of mode.

Is not a unique Hyacinth, which twenty or thirty people have been trying
in vain to cultivate, a wonderful thing? Should not he who possesses it be

Nature as Art 337


pleased with himselP Is it not very satisfying to be able to say: there are
several people in my town who have magnificent Diamonds, but no one in
the world who has a flower as beautiful as mine? Does not such a Flower
have a real value? Is one not obliged to make more of it than of a thousand
other Flowers? Could one be such an idiot as to offer it for nothing? ...
Why should one make a fuss if it is sold for a thousand Florins? 85

Rarity, beauty, and profit thus go together; what is rare is beautiful, and
what is beautiful is profitable. This is the reason for the constant changes in
the fashion of flowers, for once a particular tulip is cultivated too widely, it
becomes "obsolete and overdated" and must be replaced by others now
claimed to be more beautifu1. 86 The ultimate result, for the tulip, was the rise
of the hyacinth.
The rhetoric of florists thus has considerable resonance for the commer-
cial world. If supplies were too great, then tulips would lose their value;
tulips must therefore not become common. New techniques had to be tried,
and new flowers grown, if tulips were to remain profitable for those invest-
ing their time and money. As in a craft guild, the "mysteries" of the craft had
to be protected, although, ironically, of course, in this case there was no
proven art of creating new tulips. Such techniques as were developed were
aimed to cut the cost and labor involved in production; to turn out new vari-
eties, breeders did not want to have to wait eight or ten years to see the
uncertain results of their culled seed. The interest in tulips, their praise, their
production, thus all owed at least some of their impetus to commerce. But,
as students of the art market have noted, such arguments can apply equally
well to painting.S?
Nicolas de Valnay, by arguing that tulips were superior to other works of
art, chose to situate the flower in the milieu of aesthetics, artistic creation,
collecting, and connoisseurship. In the long-standing argument over the
superiority of art or nature, he chose nature; but this nature, for him and his
colleagues, in fact was art. The apparent creation of their artwork, their
knowledge of it, their criticism, all, Valnay would say, gave them rights to
challenge the Academie de Peinture. And, as the comparison with the acad-
emy suggests, there were social ramifications for this identification of nature
with art. Connoisseurship and collecting established social networks, to be
institutionalized ultimately in florists' societies and flower shows. The com-
mercial craze for tulips in the 163os, with its concentration solely on profit,
is something of a diversion from these trends, but not entirely. For tulips, as
for other collectibles, commerce was always a part of the equation. La Ches-
nee Monstereul might protest that "well bred spirits have beauty rather than
mercenary advantage as their object,"S8 but with tulips so expensive, the line
between beauty and profit could blur. Perhaps Paul Contant, apothecary of

Anne Goldgar
Poitiers, was more honest with himself. His cabinet and his garden were
open to visitors; but he charged them admission. 89

I would like to express my thanks to the National Endowment for the Humanities for sup-
porting my work on collecting. I am grateful for the advice and suggestions of the editors,
Paula Findlen and Pamela Smith, and of Carrie Alyea, Bertrand Goldgar, Doug Hildebrecht,
Elizabeth Honig, Jennifer Kilian, Machteld U:iwensteyn, Susan Merriam, Stephanie
Schrader, and Betsy Wieseman. I would also like to thank Michael Montias for allowing me
to compare the Montias/RKD Databank of art auctions in Amsterdam I600-40 with my
archival materials on tulip sales.

I. [Nicolas de Valnay], Connoissance et culture paifaite des tulippes rares, des anemones extra-
ordinaires, des oeillets fins, et des belles oreilles d'ours panachies (Paris: Laurent d'Houry, I688;
first published I669), "Avertissement." The book is attributed to Valnay by E. H. Krelage,
Drie Eeuwen Bloembollenexport (The Hague: Rijksuitgeverij, I946), 538, and Antoine
Schnapper, Le Giant, la licorne, la tulipe: Collections et collectionneurs dans la France du XVIIe
siecle (Paris: Flammarion, I988), 44.
2. [Valnay], Connoissance et culture paifaite, "Avertissement."

3. Lorraine Daston and Katharine Park, Wonders and the Order of Nature 1150-1750 (New
York: Zone Books, I998), chap. 7.
4- For Zeuxis, see Pliny, Natural History xxxv. xxxvi. 64-66.
5. These views are much discussed in the literature, but see, among others, Erwin Pan of-
sky, Idea: A Concept in Art Theory trans. Joseph J. S. Peake (New York: Harper & Row, I968,
first published I924), 48; Ernst Kantorowicz, "The Sovereignty of the Artist: A Note on
Legal Maxims and Renaissance Theories of Art," in De Artibus Opuscula XL: Essays in Honor
of Erwin Panofsky, ed. Milliard Meiss (New York: NYU Press, I96I) I, 268, 27I; Joy Kenseth,
"The Age of the Marvelous: An Introduction," in The Age of the Marvelous, ed. Joy Kenseth
(ex. cat. Hood Museum of Art, Dartmouth College, I99I), 38; Walter Melion, Shaping the
Netherlandish Canon: Karel van Mander's Schilder-Boeck (Chicago: University of Chicago
Press, I99I), 20-2I.
6. Beatrijs Brenninkmeijer-De Rooij, "For the Love of Flora: A Brief Look at Seven-
teenth-Century Flower Painters," in Brenninkmeijer-De Rooij et aI., Boeketten uit de Gouden
EeuwlBouquets from the Golden Age: The Mauritshuis in Bloom (ex. cat. The Hague, Maurit-
shuis, I992), I4; Paul Taylor, Dutch Flower Painting 1600-1720 (New Haven, Conn.: Yale
University Press, I995), 82- 83.
7. Constantijn Huygens, "In praestantissimi pictoris Dan. Segheri rosas," in A Selection of
the Poems of Sir Constantijn Huygem (1596-1687) ed. and trans. Peter Davidson and Adriaan
van der Weel, (Amsterdam: Amsterdam University Press, I996), I29.
8. John Rea, "Flora, To the Ladies." Flora, seu, De Florum Cultura (London: for Richard
Marriott, I665). See also Samuel Gilbert, The Florists Vade-Mecum (London: for Thomas Sim-
mons, I682), 87, who versified on tulips: "Presuming Painters find their skil out-done\At
sight of these, so Pensil'd by the Sun, That Paterzeger, doth himself confess\He colours wants
their glories to express."
9. Daston and Park, Wonders, 255-260, 276-277; Martin Kemp, "'Wrought by No Artist's
Hand': The Natural, the Artificial, the Exotic, and the Scientific in Some Artifacts from the

Nature as Art 339


Renaissance," in Reframing the Rellaissance: Visual Culture in Europe and Latin America
1450-1650, cd. Claire Farago (New Haven, Conn.: Yale University Press, 1995), pp. 177-196.
TO. On Palissy, see Daston and Park, Wonders, 285-286, and Kemp, "'Wrought,'"

191 - 193. On naturalistic automata in gardens, see John Dixon Hunt, "'Curiosities to Adorn
Cabinets and Gardells,'" in The Origms of Museums, ed. Oliver Impey and Arthur MacGregor,
(Oxford: Clarendon Press, 1985), 198-200.
1 I. Kemp's article, "'Wrought,'" focuses entirely on such objects. On coconuts, see espe-

cially Rolf Fritz, Die Gefasse aus Kokosnuss in Aiitteleuropa 1250-1800 (Mainz am Rhein: Ver-
lag Philipp von Zabern, 1983). On ostrich eggs, see [sa Ragusa, "The Egg Reopened," Art
Bulletin 53 (1971): 435-443, and Creighton Gilbert, '''The Egg Reopened' Again," Art Bul-
letill 56 (1974): 252-258. Numerous such objects are pictured in Prag um 1600: Kunst und Kul-
tur am Hofe Rudolfs II. (ex. cat. Essen, 1988); see, e.g., color plates 68, 69, 71, 72.
12. John Gerard, The Herball OJ' Generall Historie of Plantes (London: John Norton, 1597),
I19; Rea, Flora (1665),5°.
13. Petrus Hondius, Dapes indemptae, of de Moufe-.,chans/dat is, De soeticheyd Des Buyten-
Levens, Vetgheselschapt met de Boucken (1614), (Leiden: Daniel Goels, 1621; orig. ed. 1614),
88-89,93·
14. Jean Franeau, Iardin d'Hyver ou Cabinet des Fleurs (Douai: Pierre Borremans, 1616),
Elegie XXVI, 2. Pagination starts over with this elegy.
15. Gilbert, Florists Vade-Mecum (1682), 12.
16. Gilbert's concern was not only with man's arrogance, but also with those who divided
nature from God; he wished "that men may make each Clod/Speak God of Nature, make not
Nature God" (ibid., 13). Such comments, besides worrying about the role of man, point to the
sort of concerns with the relationship between God and nature in the later seventeenth cen-
tury discussed by Daston and Park (Wonders, 296- 3(1).
17. Sieur de La Chesnee Monstereul, Le floriste fran(ois, Traittant de l'origine des Tulipes
(Caen: Eleazar Mangeant, 1654),72; Ruth Duthie, Flonst,.' Flowers and SocieticJ (Haverford-
west: c.r. Thomas and Sons, 1988),7-8; Krelage, D"ie Eeuwell Bloembol!enexport, 458; Elisa-
beth Blair MacDougall, "A Cardinal's Bulb Garden: A Giardino Segreto at the Palazzo
Barberini in Rome," in MacDougall, Fountaills, Statues, and Flowers: Studies in Italian Galdens
of the Sixteenth and Seventeenth Centuries (Washington D.C.: Dumbarton Oaks Research
Library and Collection, 1994),241 n. 58.
18. Gerard, Herbal! (1597),117.
19. In Rembertus Dodonaeus, Cruydt-Boeck (Leiden: Fran~ois yan Ravelingen, 1618),367,
"Biivoegsel."
20. Frank Tyrer, ed., The Great Diurnal of Nicholas Blundell of Little Crosby, Lancashire II,
Record Society of Lancashire and Cheshire 1 '4 (1972), 55, 28 April 1726.
21. Nicolas van Kampen, 'Ti-aiti des Flew'S aOignons (Haarlcm: C. Bohn, r760), 71.
22. Rea, Flora (1665),70-72; Gilbert, Florists Vade-Mecum (1682), 8r-82, 85, and Van
Kampen, Traiti des Fleun a Oignon (1760),58-59, also recommend alternating soil or using
poor soil to weaken bulbs.
23. Rea, Flora (1665),71.
24. See for discussions of such methods Hanmer, Garden Book (1659), 17; Van Kampen,
Traiti des Fleurs a Oignons (1760), 58-59; La Chesnee Monstereul, Le floriste fran,'ois (1654),
175- 176.
25. John Evelyn, Kalendarium Hortense, ed. Rosemary Verey (London: Stourton Press,
1983; first published 1664); Parkinson, Paradlsl (1629),23.
26. Giovanni Battista Ferrari, Flora, seu de florum cultura lib!'i IV (Rome, r633), 457- 503,
cited in MacDougall, "A Cardinal's Bulb Garden," 241.
27· La Chesnee Monstereul, Lefloriste jimlf'OlS (1654), ,63-164,

Anne Goldgar
28. Ibid., 164.
29. Nouvelle instruction pour la culture des fleurs (Amsterdam: Henri Desbordes, 1697),
I20; Gilbert, Florists Vade-Mecum (1682), "To the Reader."
30. Jan van der Groen, Den Nederlantsen Hovenier (Amsterdam: Weduwe van Gijsbert de
Groot, 1721; orig. ed. 1669), "Inleydingh."
3I. Gilbert, Florists Vade-Mecum (1682), 25I.
32. Brenninkmeijer-De Rooij, "For Love of Flora," 14.
33. Nouvelle Instruction pour la Culture des Fleurs (Amsterdam: Henri Desbordes, 1697), 120.
34. See Hessel Miedema, "De St. Lucasgilden van Haarlem en Delft in de zestiende
eeuw," Dud Holland 99, no. 2 (1985): 77-I09; Miedema, "Kunstschilders, gilde en academie:
Over het probleem van de emancipatie van de kunstschilders in de Noordelijke Nederlanden
van de 16de en 17de eeuw," Dud Holland 101, no. I (1987): 1-34; Svetlana Alpers, The Art of
Describing: Dutch Art in the Seventeenth Century (Chicago: University of Chicago Press, 1983),
100, 102; Celeste Brusati, "Stilled Lives: Self-Portraiture and Self-Reflection in Seventeenth-
Century Netherlandish Still-Life Painting," Simiolus 20, nos. 2/3 (I99o!r99I): 17I. On
embroiderers in the St. Lucasgilde, see Saskia de Bodt, . . . op de Raempte off mette Brodse:
Nederlandse Borduurwerk uit de Zeventiende Eeuw (Haarlem: H. Becht, 1987),22-36. The
comment on Frans Pietersz. de Grebber appears in Karel van Mander's Schilder-boeck; the
best edition is Karel van Mander, The Lives of the Illustrious Netherlandish and German
Painters, from the first edition of the Schilder-boeck (1603-4), ed. and trans. Hessel Miedema
(Doornspijk: Davaco, 1994), Vol. I, f. 300 of facsimile.
35. University of Leiden (UBL), Ms. Vule. IOI, Marie de Brimeu, princesse de Chimay, to
Carolus Clusius, letter 2, Leiden, 18 September 159I. Tapisseries, the word used here, seems to
have been a usual metaphor of hers for gardens; in other letters she describes the garden she
lost by moving to the north as "des belles tapisseries que jay perdu pendant ces troubles"
(UBL, Ms. Vule. 101, Marie de Brimeu to Clusius, letter 3, Leiden, 24 January 1592) and
remarks that "com bien q[uel naves moien de dresser vostre tapiserye tellem[entl que
desireries ie ne doute que nonobstant ce e1le serat tres belle" (UBL, Ms. BPL 885, Marie de
Brimeu to Clusius, The Hague, 5 November 1593)' That it was a more general comparison is
suggested by the verb se tapisser used by Christiaan Porret to describe his garden: "Mon iardin
commence a se tapisse[r] de petites fleurs" (UBL, Ms. BPL 2724d, Christiaan Porret to Mat-
teo Caccini, 25 February 161 I). This comparison with carpets is also discussed in Erik de Jong
and Marleen Dominicus-van Soet, Aardse Paradijzen: De tuin in de Nederlandse kunst, I5de tot
I8de eeuw (ex. cat. Haarlem, Frans Halsmuseum, 1996), ro3, and MacDougall, "A Cardinal's
Bulb Garden," 233.
36. Parkinson, Paradisi (1629), 14.
37. Pierre Vallet, Le Iardin du Roy Tres Chrestien Henry IV Roy de France et de Navare
([Paris): 1608). See also Penelope Hobhouse, Plants in Garden History (London: Pavilion
Books, 1992), roB.
38. Carolus Clusius, Rariorum plantarum historia (Antwerp: Plantin, 1601), II, cap. IX, sec.
III, pt. 2,146; cap. VII, sec. VII, pt. 8,142.
39. The earliest use of these names I have found is 1592, when the tulip-lover Jan van
Hoghelande in Leiden reported the flowering of "une [tulipel de drap d'or fort belle" in his
garden (UBL, Vule. 101, Jan van Hoghelande to Clusius, letter 7, 20 May 1592 N.S.). The ear-
liest I have found in Dutch is in Emmanuel Sweerts, Florilegium (Frankfurt am Main:
Anthonius Kempner, 16I2), "Catalogus den ersten Boeck," in which the "Tulipa geel met
fleyne roode strepenlgenaempt goude Laecken" stands out among a host of mere descriptions
of flowers; in the French section of the tetralingual Florilegium it is merely called "d'oree," as
a description, and the name also does not appear in the German and Latin sections. The min-
ister Walter Stonehouse had both "Cloth of golde" and "Cloth of sylver" in his garden in

Nature as Art 34 1
Yorkshire in 1640 ("The Garden of the Rev. Walter Stonehouse at Darfield Rectory in York-
shire," The Gardeners' Chronicle, 29 May 1920, 268), and La Chesnee Monstereul's long cata-
log of tulips in 1654 includes four types of "Drap d'argent" (Floriste fi·an,ois, 224). Annotators
of Dodonaeus's Cruydt-Boeck (1618 ed.) made the same comparisons: "sometimes one color
shines above the other: that is the white and the yellow have something red shining through
it: one like gold cloth/one like silver cloth" (Cruydt-Boeck [1618], 365).
40. Irene Groeneweg, "Court and City: Dress in the Age of Frederik Hendrik and
Amalia," in Princely Display: The Court of Frederik Hendrik of Orange and Amalia van Solms in
The Hague, ed. Marika Keblusek and Jori Zijlmans, (ex. cat, The Hague, Haags Historisch
Museum, 1997),203.
41. Linda A. Stone-Ferrier, Images of Textiles: The Weave of Seventeenth-Century Dutch Art
and Society (Ann Arbor: UMI Research Press, 1985), 172,215-216; Groeneweg, "Court and
City," 201 -203, 205-206; Valerie Cumming, '''Great vanity and excesse in Apparell': Some
Clothing and Furs of Tudor and Stuart Royalty," in The Late King's Goods: Collections, Pos-
sessions and Patronage of Charles I in the Light of the Commonwealth Sale Inventories, ed. Arthur
MacGregor (London and Oxford: Alistair McAlpine/Oxford University Press, 1989), 322,
326; Donald King, "Textile Furnishings," in MacGregor, The Late King's Goods, 307-308;
Alison McNeil Kettering, "Ter Borch's Ladies in Satin," in Looking at Seventeenth-Century
Dutch Art: Realism Reconsidered, ed. Wayne Franits (Cambridge: Cambridge University
Press, 1997),103-4- On the Leiden say industry, see Leo Noordegraaf, "The New Draperies
in the Northern Netherlands, 1500- 1800," in The New Draperies in the Low Countries and
England, ed. Negley B. Harte (Oxford: Oxford University Press, 1997), 173-195, and Stone-
Ferrier, Images of Textiles, 23-29; on the Amsterdam silk and silver cloth industries, see
Leonie van Nierop, "De zijdenijverheid van Amsterdam, historisch geschetst," Tijdschrift
voor Geschiedenis 45 (1930): 18-40, 151-172; and 46 (1931): 28-55,113-143, and H. Brug-
mans, Geschiedenis van Amsterdam (Utrecht and Antwerp: Het Spectrum, 1973) 3: 31. The
bright colors associated with tulips were not as divorced from the clothing of the Dutch elite
as somber but misleading paintings might suggest, as Irene Groeneweg argues in "Regenten
in het zwart: vroom en deftig?" Nederlands Kunsthistorisch Jaarboek 46 (1995): 199-251.
42. Lorenz Seelig, "The Munich Kunstkammer 1565-1807," in Impey and MacGregor,
Origins of Museums, 84-5.
43. De Nederlandsen Bloem-Hof, of de Nauwkeurige Bloemist (copy consulted has no title
page; publication attributed by University of Amsterdam Library to Amsterdam: Harmen
Machielsz and Nicolaas ten Hoorn, 1699), 7.
44. Franeau, lardin d'hyver (1616), 96- 104,106,108,110, 114,118,122,124,125; on the Due
van Tholl, 97; on "Ies grans," 125. The book is filled with names of those cultivating tulips.
45. For example, Franeau, 95, 102, 104, 114, 121, 125; Rea, similarly, refers to banks of
flowers as "Enamel'd" (Flora, "Flora. To the Ladies").
46. La Chesnee Monstereul, Le jloriste ji-an,ois, 208 - 2 I 3.
47. Rea, Flora (1665),51.
48. Antoine Agard, Discours et fOole des medailles & autres antiquitez ... Ii present rangees
dans Ie Cabinet du Sieur Antoine Agard . .. (Paris, 161 I), 14 - 17, 26, 27, 3l; Catalogus oft Regis-
ter vande Sonderling-Heden oft Rariteyten ende Wtgelesen Sinnelickheden ... Die Christiaen Por-
rett, wijlen Apotekel; in zljn Cunstcamer vergatert had (auction catalogue, Leiden: Jan Claesz.
van Dorp, 1628); UBL, ms. BPL 2596-9, Collectie Hunger, transcription of catalogue of cab-
inet of Paludanus from ms. original in KB Copenhagen, ff. 190-203 (in original ms., ff.
130-140). On the collection of agates, see Schnapper, Geant, 191 - 192; onpietra dura, see C. w.
Fock, "Pietre Dure work at the court of Prague and Florence: Some Relations" in Prag um
1600,51-59; J. F. M. Sterck, "Dirck van Rijswijck. Fen Amsterdamsch Goudsmid en Moza-
"iekwerker," Jaarverslag Konink"jk Oudheidkundig Genootschap (1908-09), 35 - 54; Seelig,

Anne Goldgar
"Kunstkammer," 79. The colors of agate and marble are praised in Thomas Nicols, A Lap-
idary; Or, the History of Pretious Stones (Cambridge: Thomas Buck, 1652).
49. On the introduction of marbled paper to Europe, see Phoebe Jane Easton, Marbling: A
History and a Bibliography (Los Angeles: Dawson's Book Shop, 1983), 33ff; Rosamond B. Lor-
ing, Decorated Book Papers (Cambridge, Mass.: Harvard College Library, 1942), 12- 13;
Richard J. Wolfe, Marbled Paper: Its History, Techniques, and Patterns (Philadelphia: Univer-
sity of Pennsylvania Press, 1990),3-14; Graham Pollard, "Changes in the Style of Bookbind-
ing, 1550-1830," The Library 5th seT. II, no. 12 (June 1956): 79. On the role of the
Netherlands, see Easton, Marbling, 63-66, and J. F. Heijebroek and T. C. Greven, Sierpapier:
marmer-, brocaat- en sitspapier in Nederland (Amsterdam: De Buitenkant, 1994), 15-17; the
latter authors disagree with Wolfe, who claims a more important role for Germany and
France (Wolfe, Marbled Paper, 13)' On Paludanus's early possession of marbled paper, see
Heijebroek and Greven, Sierpapier, 14. The example of I'Estoile is from Loring, Decorated
Book Papers, 24.
50. Easton, Marbling, 109, I I I.
51. John Evelyn ms. in BL, "An Exact Account of the Making of Marbled Paper," quoted
in Charles M. Adams, Some Notes on the Art of Marbling Paper in the Seventeenth Century
(New York: New York Public Library, 1947), II. Adams does not give the exact reference to
the Evelyn ms.
52. UBL, Ms. Vule. 101, Anna Maria von Heusenstain to Clusius, letter 9, Vienna, 7 May
1591. Another correspondent of Clusius's referred to a "plante ... du papier Persien et
Turquesque," but it is not clear if this was a tulip. UBL, Ms. Vule. 101, Jacques Plateau to
Clusius, letter 9, Tournay, 8 September 1592.
53. Seymour Slive, Rembrandt and His Critics 1630-1730 (The Hague: Martinus Nijhoff,
1953), I.
5+ The chief text for the aesthetics of early seventeenth-century Dutch painting is Karel
van Mander, Den grondt der edel vry schilder-const, the first section of his Het Schilder-Boeck
(Haarlem: Paschier van Wesbusch, 1604). The Grandt, or Groundwork, has been edited in a
separate modern version by Hessel Miedema (Utrecht: Haentjens Dekker and Gumbert,
1973), 2 vols. Miedema has also written several analyses of the chief descriptive terms in the
work: Hessel Miedema, Fraey en Aerdigh, Schoon en Moy in Karel van Manders Schilder-Boeck
(Amsterdam: Kunsthistorisch Instituut, 1984); Hessel Miedema, Kunst, Kunstenaar en Kunst-
werk bij Karel van Mander (Alpen aan den Rijn: Canaletto, 1981), see especially 146-152,
156-159. A challenge to some of Miedema's views on Van Mander is Walter S. Melion, Shap-
ing the Netherlandish Canon: Karel van Mander's Schilder-Boeck.
55. Van Mander, Grondt, chap. 12, stanza 21; Dodonaeus, Cruydt-Boeck (1618 ed'),365,
"Biivoegsel." On Van Mander's use of netticheyt, see Melion, Shaping the Netherlandish
Canon, 60-63. This is a concept also applied to flower painting; see Taylor, Dutch Flower
Painting, 96, 99.
56. E.g., on shells: Pliny, Natural History Ix.lii.
57. John Shearman, Mannerism (Harmondsworth: Penguin, 1967), 75, 86, 92, 100- 101,
105,139,140- 151.
58. "Door verscheydenheyt is Natuere schoone": Van Mander, Grondt, chap. 5, stanza 20.
On this verse, see Melion, Shaping, 8-9, 21. A canonic example for Van Mander is the Jan van
Eyck Adoration of the Lamb by the Elders in Ghent, which he valued for its clear,
new-seeming colors and the variety of its composition, encompassing as many as 330 differ-
ent faces. See Karel van Mander, The Lives of the Illustrious Netherlandish and German
Painters, ed. Miedema, Vol. 1,61; f. 220 of facsimile.
59. Van Mander, Grondt, chap. V, stanzas 32-3. Melion comments on this passage in Shap-
mg,9·

Nature as Art 343


60. Rembertus Dodonaeus, Cruydt-Boeck (Leiden: Fran<;ois van Ravelingen, 1608),389,
later annotations.
61. The subject of connoisseurship awaits full treatment, but see Zirka Zaremba Fil-
ipczak, Picturing Art in Antwerp 1550-1700 (Princeton, N.J.: Princeton University Press,
1987); Elizabeth Honig, "The Beholder as a Work of Art: A Study in the Location of Value
in Seventeenth-Century Flemish Painting," Nederlands Kunsthistorisch Jaarboek 46 (1995)
253 -297; Elizabeth Honig, "Making Sense of Things: On the Motives of Dutch Still Life,"
Res 34 (autumn 1998): 167- 183; Antoine Schnapper, Curieux du Grand Siecle (Paris: Flam-
marion, 1994). The literature on collectors is enormous, but see especially Paula Findlen, Pos-
sessing Nature: Museums, Collecting, and Scientific Culture in Early Modern Italy (Berkeley:
University of California Press, 1994); Schnapper, Giant; Ellinoor Bergvelt and Renee
Kistemaker, eds., De Wereld binnen handbereik: Nederlandse kunst- en rariteitenverzamelingen
1585-1735 (ex. cat. Amsterdams Historisch Museum, 1992). Alongside connoisseurship, we
can also place the setting of standards for tulips back within the context of craft guilds, whose
task it generally was to maintain a level of quality in their members' product. This was true
of the St. Lucasgilde, although less is said in its regulations about painting than about the
other crafts included in the guild. See Hessel Miedema, "Over kwaliteitsvoorschriften in het
St. Lucasgilde; over 'doodverf,'" Oud Holland IOI, no. 3 (1987): 141-147.
62. Nicolas-Claude de Fabri de Peiresc to his brother Palamede de Valavez, Aix, 3 May
1626, in Lettres de Peiresc, ed. Philippe Tamizey de Larroque (Paris: Imprimerie Nationale,
1888-1898) Vol. 6: SOL
63. Parkinson, Paradisi, "Epistle to the Reader"; La Chesnee Monstereul, Le floriste
fran,ois, chap. 5, "Quelle doit estre la Tulipe, tant en ses Couleurs, Panaches, qu'en sa Forme";
Van Kampen, Traite des Fleurs Ii Oignons (1760), 60-61, "Les qualites requises pour former
une belle Tulipe ... "; [Valnay], Connoissance et culture paifaite, I2ff.
64. [Valnay], Connoissance et culture paifaite, 12-21.
65. Rea, Flora, "To the Reader."
66. Duthie, Florists' Flowers and Societies, 33, 71; Sam Segal, "Exotische bollen als sta-
tussymbolen," Kunstschrift 31, no. 3 (1987), 96; D. Tarver and B. Elliott, "Des Fleuristes aux
Societes Horticoles. Histoire des Expositions Florales," in L'Empire de Flore, ed. Sabine van
Sprang (Brussels: La Renaissance du Livre, 1996), 117-II8. A possible precursor to such
shows might be the use of a "theater" of tulips, arranged on racks for other collectors to exam-
ine, an invention attributed to Valnay by the Nouvelle instruction pour la culture des fleUl"S
(1697), 12T "Monsieur de Valnay invented a completely lovely sort of theater, to be able to see
easily and all together a whole mass of panachees mixed together according to their different
colors & arranged near each other, so that, sitting in the shade & with one glance you can
divert your sight with all the rarities that a large garden can produce."
67. Dodonaeus, Cruydt-Boeck, (1608 edition), 388.
68. Examples are numerous in the Clusius correspondence at the University of Leiden, but
for several printed documents demonstrating this, see F. W. J. Hunger, Acht Brieven van Mid-
delburgers aan Carolus Clusius (Middelburg: Zeeuwsch Genootschap der Wetenschappen, 1925)'
69. Peiresc to Palamede de Valavez, Aix, April IO, 1626, in Lettres de Peiresc Vol. 6: 443; Sir
Thomas Hanmer to John Evelyn, Bettisfield, 21 August 1671, quoted in introduction to Han-
mer, Garden Book, xv; Hanmer's pocket book, quoted in introduction to Garden Book, xx,
gives accounts of Hanmer's gifts to Lambert; BL Sloane Ms. 95, ff. 153-154, "Accounts for
my Garden 1638." It should be noted that the anonymous English estate owner was clearly
indulging not only in exchange but in trade of flowers. The manuscript includes entries such
as "M" Rous hath fro[m] me 200 small Anemones to halves, and of my part m" Elizabeth
Budding must have our half of the profitt. ffreeman oweth me 3£ upon bill W'h is in my box,
in lue whereofI have of him 2 tulips called Aggot Goblins wch I have set in my little square

344 Anne Goldgar


north east, the wch I have agreed to have ifI like them when I see their flowers, or els returne
to him and have fro[m] him my three pounds." BL Sloane ms. 95, f. 148, "Accounts for my
Garden 1638."
70. [Ralph Knevet], Rhodron and Iris: A Pastoral! as it was presented at the Florists Feast in
Norwich, May 3, 1631 (London: for Michael Sparke, 1631). On florists' feasts, see Duthie,
Florists' Flowers and Societies, 9-14; and Tarver and Elliott, "Des Fleuristes," 124. It is not
clear in what fashion flowers were involved in the Norwich society's activities, although
Knevet's play is full of floral references.
7I. Duthie, Florists' Flowers, 5, 8. Duthie reports the earliest known use of the term as
1623, in a letter by Sir Henry Wotton, but Jean de Maes used "floriste" to describe amateur
growers in 1599, reporting as well that "Je commen~oie aussi a floriser ... " (UBL, ms. Vule.
lOI, Jean de Maes to Clusius, letter 5, Brussels, 8 April 1599).
72. Rea, Flora (1665),50.
73. La Chesnee Monstereul, Le jloriste franfois (1654), 180- I8r. That these sentiments
were not only French is perhaps indicated by the plagiarism of this passage (along with the
rest of the book) by Hendrik van Oosten in his Nieuwe Nederlandsche Bloemhof(I700); see the
English translation, Henry van Oosten, The Dutch Gardener: Or, the Compleat Florist, 2nd ed.
(London: D. Modwinter, 171 I), 160. The quotation here is my translation of La Chesnee
Monstereul rather than from the English edition of Van Oosten.
74. [Valnay], Connoissance et culture paifaite (I68S), 7-8.
75· Ibid., 7·
76. La Chesnee Monstereul, Lejloristefranr;ois, 173-180; quotation on 173 - 174' Schnap-
per, Giant, 218-219, also discusses ambivalent attitudes of gardening treatises toward reveal-
ing their secrets.
77. Michel de Marolles, Suitte de Memoires de Michel de Marolles Abbe de Ville- loin (Paris:
Antoine de Somma ville, 1657),268.
78. My preliminary research on Haarlem suggests that some involved in tulip sales there
may have been of a lower social station than in Amsterdam and thus less likely to be pur-
chasers of art, although some were also from leading Haarlem families. Unfortunately auc-
tion records from Haarlem appear to have been lost, making this topic more difficult to
investigate.
79. Casteleyn's inventory is in the Gemeentearchief, Amsterdam (GAA): GAA NA
939AIr87ff., 2S August 1644. A tulip sale involving Casteleyn is recorded at GAA NA
3411r55, 26 September 16II; he was obviously involved in tulip collecting from early on. On
Casteleyn, see also Bergvelt and Kistemaker, eds., De Wereld binnen Handbereik, 316.
So. My list of Amsterdam purchasers of tulips is compiled from notarial transactions in
the Gemeentearchief, Amsterdam (GAA), concerning sales or quarrels over sales. I am grate-
ful to Michael Montias for allowing me to compare this list with the Montias/RKD Databank
of buyers of art at auction in Amsterdam from 1600 to 1640' Notarial records about tulips and
concerning those collectors listed include: on Bessels, GAA NA 951, 14 May 1637; on Admi-
rae!, GAA NA 9171272" 16 October 1635; GAA NA 9171310, 1 December 1635; GAA, NA
9IS1r45" 19 May 1636; GAA NA 91SIrSO, I2 June I636; GAA NA 9IS1228, 6 July 1636; GAA
NA 9IS/519" S December I636; GAA NA 919/64, 13 February 1637; GAA NA 9191r95, I3
June 1637; GAA NA 920/334, 2 December I63S; on Halmael, GAA NA 9IS/554v, 3I Decem-
ber 1636; GAA NA 9I9120" 12 January 1637; GAA NA 9I9121, 15 January 1637; GAA NA
919/61, 10 February 1637; GAA NA 9I9/61, 10 February 1637; on Massa, GAA NA
II5SIr44-14f, 27 December 1636. Adam Bessels was closely connected to the major collect-
ing family, the Reynsts, as well as to well-known collectors such as Jacques Nicquet and Jean
Renialme; Lambert Massa was connected to Renialme; his brother, Isaac, was painted by Hals
three times, and his tulip purchase was from Susanna Spranger, sister of the important col-

Nature as Art 345


lector Gommer Spranger and widow of Outger Cluyt. Cluyt was a son of Dirck Outgertz
Cluyt, Clusius's assistant at the Leiden hortus, and had written a treatise on the tulip. There
are many other similar examples.
8I. On this question for Dutch paintings, see Marten Jan Bok, "Art-Lovers and their
Paintings: Van Mander's Schilder-boeck as a Source for the History of the Art Market in the
Northern Netherlands," in Dawn of the Golden Age, ed. Ger Luijten et al. (ex. cat. Amster-
dam, Rijksmuseum, r993), r 36- r66; Marten Jan Bok, Vraag en aanbod op de Nederlandse kun-
stmarkt, 1580-1700 (Proefschrift, University of Utrecht, r994).
82. Clusius to Lipsius, 22 October r594, quoted in F. W. Hunger, Charles de I'Escluse. Car-
olus Clusius. Nederlandsche Kruidkundige 1526-1609 (The Hague: Martinus Nijhoff, r927,
r943), Vol. 2,250. On the rhizotomi, see also Hunger, Clusius Vol. r: 303-304, and Vol. 2: 25r;
Krelage, Drie eeuwen, 3.
83. In Dodonaeus, Cruydt-Boeck (r6r8 edition), 365. "Biivoegsel." This passage does not
occur in the r608 edition.
84. Rea, Flora, 7I.
85. George Voorhelm, Traite sur la Jacinte, 3rd ed. (Haarlem: N. Beets, 1773; rst publ.
r752),21-22.
86. Gilbert, The Florists Vade-Mecum (r683), "To the Reader."
87. See especially J. Michael Montias, "Cost and Value in Seventeenth-Century Dutch
Art," Art History 10, no. 4 (December 1987): 455 - 466.
88. La Chesnee Monstereul, Le Jloriste franr;ois 1654), r94-
89. Paul Contant, "Le Iardin et Cabinet Poetique," in Jacques and Paul Contant, Les
Divers Exercices de Iacques et Paul Conlant Pere et Fils Maistres Apoticaires de la Ville de Poictiers
(Poitiers: Julian Thoreau and la Veuve d'Antoine Mesnier, 1628),55-56. Contant defends his
decision to take money- the charge was that he was too "prompt & diligent\A prendre d'un
chascun de l'or & de l'argent\Pour voir mon Cabinet" -by asking whether, if someone seeing
his cabinet wanted to give him money for seeing it, he really had to refuse?

Anne Goldgar
Inventing Exoticism
The Project of Dutch Geography and the Marketing
of the World, circa 1700

BENJAMIN SCHMIDT

W hat do we talk about when we talk about the exotic? This


depends, in part, on when we do the talking, though there may
be no better time to listen in on the conversation than the late
seventeenth and early eighteenth centuries, when the talk was relatively
thick. Consider: Georg Rumphius-also known as Plinius Indicus for his
monumental contribution to the natural history of the East Indies-spoke
of the "imperfect chaos" of the exotic world that he assembled in his mas-
sive studies of Asian marine and plant life, The Amboinese Curiosity Cabinet
and The Amboinese Herbal, both composed at the close of the seventeenth
century. This and other remarks by Rumphius on the unwieldy business of
ordering tropical naturalia reflect perhaps the finicky anxieties of a scholar.
By contrast, Engelbert Kaempfer-the "Humboldt" of the Indies, as he
came retrospectively to be known - praised the "pleasures of the exotic."
Here the emphasis lay on the readers' reception of those tomes chock-full of
Near and Far Eastern exotica compiled by Kaempfer in the late seventeenth
century, and on the effect that these stunning descriptions, visual no less
than verbal, presumably had upon their audience. And then there is the
commentary of Arnoldus Montanus. Neither a Pliny nor a Humboldt was
this modest schoolteacher turned geographer who delighted in the "nov-
elty," "variety," and "strangeness" of those exotica collected in his vast, ram-
bling books of "wonders" that explored the terrain of Asia, Africa, and
America at the twilight of the seventeenth century. I
All of these observations derive from a particular context, and it is this
context-time and place, purpose and perspective-that lends recogniz-
able shape and valuable definition to that all-too-vaguely articulated con-
cept "the exotic." All can be associated, that is, with the project of
geography in the Dutch Republic, circa 1700: a singular burst of printing,
painting, mapping, publishing, producing, and otherwise promoting

347
images of the non-European world undertaken chiefly in the province of
Holland around the turn of th e eighteenth century. Rumphius and
Kaempfer both gained their knowledge of the world while serving in the
Dutch East India Company; Montanus, an accomplished armchair traveler,
belonged to the stable of geographers, historians, and draftsmen who
worked for the Amsterdam publisher Jacob van Meurs. Finally, all of these
comments, despite their identifiably Dutch provenance, circulated through-
out late Baroque Europe in various editions and multiple languages, thereby
conveying to a very broad audience what indeed was meant when the exotic
was talked about.
To be sure, discussions focused on the idea of the exotic, both in and of the
late seventeenth century, tend to be rare and still more rarely satisfying. In
the first instance, few of the contemporary commentators on matters of
geography chose to elaborate on their theories of the exotic; the word itself
appears only infrequently in Baroque letters. 2 In the second instance-in
the, by contrast, very many critical discussions of Europe's mimetic engage-
ment with the expanding early modern world - the exotic too often gets
tucked away, neatly and nonchalantly, under the broad heading of "Europe
and its Other." This permits a certain laxness concerning the Other, under-
stood simply as the object of European imperial desire. It also allows-
equally problematically - considerable leeway for that grand, elastic figure
of the early modern European, who habitually goes undistinguished by gen-
der, class, culture, and so forth. It strips the exotic and its makers of context.
The comments of Rumphius, Kaempfer, Montanus, and others gain their
value precisely by pointing to a more cogent idea of exoticism, which per-
tains more particularly to late Baroque Europe-a crucial period of geo-
graphic production and European expansion that heralded the age of
empire. They indicate, too, how this idea was formulated, propagated, and
marketed by the Dutch, most conspicuously around I7oo.3
The Dutch project of geography at the turn of the eighteenth century
provokes a number of observations. First, the sheer quantity of materials
produced in the Republic is impressive. A veritable flood of georgraphic
goods-of literary works, such as travelogues, learned geographies, natural
histories, and books of "wonders"; of cartographic resources, including dec-
orated maps, multivolumed atlases, and luxurious globes; of visual artifacts,
comprising tropical painting, inexpensive prints, and coveted curiosa, (the
latter often sold with the cabinets that housed them)-streamed off the
presses and out of the ateliers of the Netherlands, suppliers in the province
of Holland manufacturing most assiduously. Second, many of the works
produced in the Netherlands share a certain bric-a-brac quality that suggests
a self-conscious strategy of-for lack of a better word-exoticism. Books
on the extra-European world move briskly and even programmatically

Benjamin Schmidt
among countries, customs, and creatures; the preface to one volume claims
to offer only quick "morsels" of exotica, variety being the favorite spice of
geographic life. 4 Paintings and prints likewise mix and match peoples and
continents, embracing a lush aesthetic of eclecticism. Third, the timing of
this burst of geography is significant for what it is not: coincident with any
period of Dutch imperial expansion. On the contrary, the later decades of
the seventeenth century witnessed a reduction of the Republic's activities
abroad: the Netherlands' American colonies (in Brazil and New York) fell
in the 1650S and 1660s, and the Dutch East India Company started losing
relative market share at about the same time. The Republic, in other
words, chose to market a world which it had a contracting stake in govern-
ing. Which brings up a final, crucial point: that the Dutch in fact "mar-
keted" and sold that version of the world which their geographers so
meticulously fashioned. This worked in a variety of ways. Rumphius's text
very plainly promoted the valuable shells of his own cabinet of curiosities
(a previous collection had been sold to Grand Duke Cosimo III of Tus-
cany), and Montanus-or his publisher, van Meurs-ambitiously adver-
tised in his Asian volume the merits of another deluxe edition, America, by
the same author-publisher team. 5 Somewhat more subtly, Kaempfer's tex-
tual descriptions of Oriental riches sold the reader on a veritable cornu-
copia of imports lately available in Europe, as did, in their own ways, any
number of Dutch texts, which sensuously described the wondrous prod-
ucts of the expanding globe. Still more to the point, Dutch geography sold
an idea of the world that appealed to readers, viewers, and consumers
across Europe, and this idea marketed a world that was identifiably
"exotic."
This essay investigates the development of geography in late Baroque
Europe and the manner in which various modes of describing the world
functioned in the decades surrounding 1700. It proposes that, in the
Netherlands at this time, a category and a strategy of discourse emerged,
that might best come under the rubric of "exotic." The term exotic has been
crucial to any number of discussions of Europe's post-Columbian expan-
sion, its establishment of commercial markets overseas, and its develop-
ment of imperialist ideologies. Exoticism has served as a touchstone for
analyses of colonial (and postcolonial) discourse, of European representa-
tions of the Other, and of early modern forms of geography - the evolving
art and science of empire. Despite this evident centrality, however, few
excavations of the term and its implications have informed the study of
these topics, and certainly not the study of geography in Europe's great age
of expansion. Few attempts have been made to locate the production and
consumption of exotica, especially in that period bridging the Old World's
initial thrusts overseas in the sixteenth and early seventeenth centuries, and

Inventing Exoticism 349


the establishment of vast European empires by the later eighteenth and
nineteenth centuries.

Where do we look when we look for the exotic? To track the exotic, one need
look no further than the Dutch Republic, geographers nonpareil of early
modern Europe, and Dutch strategies of representing the world circa 1700.
Dutch geography, prominent enough in the early years of Renaissance
exploration, expanded dramatically in the final decades of the seventeenth
century. In an astonishing assortment of media, the Dutch described, delin-
eated, reproduced, and otherwise propagated images of the world beyond
Europe. This applies not only in terms of the great quantity of works issuing
from the Republic. It characterizes also the phenomenal variety and daz-
zling quality of texts, images, and objects pertaining to geography (the prod-
ucts) and the extraordinary extent of their dispersal (their consumption). The
Republic, quite simply, dominated the field. By consequence, it framed the
way most Europeans viewed the globe at the turn of the eighteenth century.
Dutch geography came in various shapes and forms, genres and media.
There were, to begin with, "traditional" geographies (sometimes published
under the name "cosmographies"), which, in the humanist mode, sought to
detail the lay of the world. These were printed texts, which became, in many
instances, authoritative textbooks: Philip Cluverius's Introductionis in univer-
sam geographiam, which appeared in a staggering sixty-seven editions by
1725; Bernard Varenius's widely cited Geographia generalis; and the vast,
omnibus works of Georg Hornius-"social" studies on the empires of the
world, the origins of the races, the nature of world polities, and the like-
that numbered forty editions in the final third of the century.6 Next came the
regional studies-fabulous, often folio works, sometimes called "atlases"
and generally brimming with engraved prints, foldout maps, and panoramic
views-that filled a market for "local" geographies of the non-European
world: Schouten on Siam (nearly twenty editions), Nieuhof on Brazil (yet
another Jacob van Meurs title), Baldaeus on Malabar (reprinted in
Churchill's acclaimed Travels) (fig. 14.1).7 The magnificent natural histories
done in the Republic might also strive to be global, yet this genre tended to
encourage studies that were site-specific: Maria Sibylla Merian's glorious
Metamorphosis insectorum Surinamensium, Hendrik van Reede tot
Drakestein's twelve-volume Hortus Malabaricus, and Rumphius's Amboin-
sche rariteitkamer, all of which appeared in the decennium mirabilius of exotic
natural history, 1695-1705.8 The Dutch also produced a notably large share
of travel narratives: well over fifty editions of Willem Bontekoe's adventures
in South Asia; multiple versions of Ogier Busbecq's life among the Turks;
thick accounts of Cornelis de Bruyn's journeys to the Levant and Muscovy.
And they were behind some of the most important travel anthologies:

Benjamin Schmidt
.:,,.,\
L W . I
'-r,

a t+ /I./ Frontispiece from Philippus Baldaeus, Wahrhaftige ausfiirliche


beschreibung der beruhmten ostindischen &sten Malabar und Coromandel, als
auch der insel Zeylon (Amsterdam, 1672). Courtesy of the Rosenfeld University.
University of California, Los Angeles.
sprawling, multivolume collections, including Hartgers' "Voyages" and
Commelin's "Travels" (which formed the basis of Renneville's and
Churchill's similarly gargantuan collections); and the twenty-eight volume
series of Pieter van der Aa, which served as a standard geographic reference
work for half a century.9 A final category that comes under the rubric of "lit-
erary" text is the Dutch books of wonders and "things-of-the-world." These
described, in some cases, a single set of attributes or customs surveyed in a
global setting: Balthasar Bekker on the universality of witchcraft or Bernard
Picard on world religions, for example. In other cases, they jumbled
together, Kunstkammer-like, the miscellaneous mores and disparate marvels
of the world; arranged these quirky attractions with purposeful disregard
for place; and finally framed them with enticing titles, such as "The Great
Cabinet of Curiosities," "The Warehouse of Wonders," or, most bluntly,
"The Wonder-Filled World."ID
Also from the Netherlands came a stupendous outpouring of cartographic
texts, which placed the Republic in the enviable position of mapmaker to
Europe. From the workshops of Blaeu, Janssonius, and de Wit came the pre-
eminent "grand" atlases of the later seventeenth century. Under the signs of
Colom, Donker, and van Keulen one could obtain the most sought-after
"water worlds," or sea atlases, on the market. Alongside these sumptuous,
often folio works were simpler "minor" atlases and plain sheet maps for the
cost-conscious. Available for the truly profligate, though, were the high-end
wall maps, which decorated the finest homes of Europe. Elaborately deco-
rated and often hand-colored, these could cost more even than an atlas; they
also came in made-to-order "floor models": mosaics of the world mapped
onto a patron's floor. Deluxe Dutch globes spun in the best studies and
princely corridors of Europe; miniature "pocket" globes, meanwhile, could
be conveniently carried across the continent. And, just as Dutch printers and
engravers produced city views, topical news maps, and topographical panora-
mas (some up to 2 meters in length), so did Dutch draftsmen and painters,
often working from the same original sketches, furnish more durable scenes
of the identical exotic terrains done in watercolor and oil. 11
From Dutch painters, too, came diverse visual versions of the non-Euro-
pean world. Tropical landscapes-a genre wholly invented by the Dutch in
the second half of the seventeenth century - presented the far-flung corners
of the globe in lush, opulent color (fig. 14.2). Here the very competitive mar-
ket for paintings in the Republic induced innovation by encouraging artists
to select, and then specialize in, a particular region or style. Frans Post, per-
haps most famously, painted verdant variations of Brazilian life-though
other regions had their masters. Dirk Valkenburg cornered the market for
Suriname scenes; Andries Beeckman did the same for the East Indies; and
Gerard van Edema applied his brush to that growing urban jungle known

Benjamin Schmidt
i4.2 Frans Post, View of Olinda, 1662 (canvas, 107.5x 172.5 cm). Courtesy
of the Rijksmuseum, Amsterdam.

as Manhattan, which he rendered in vaguely pastoral, capriccio views. Other


genres also developed. Painters of the Republic reproduced on canvas the
products and peoples of the world-as in Albert Eckhout's studies of the
non-European inhabitants of Brazil-and the flora and fauna of the trop-
ics. Many of these images were subsequently recycled in still lifes and other
exotic compositions-Jacob van Campen's Triumph, with Treasures of the
Indies combines elements of still-life painting with human figures bearing
the rich harvest from overseas-or in delftware, tapestries, and other deco-
rative arts (fig. I ~ . ~ ) . ' ~
The actual objects depicted on the canvas had their market as well, which
the Dutch skillfully supplied. The Republic, that is, did a brisk business pro-
visioning the Kunst- and Wundert(ammern of Europe. Rumphius's shells
went to the ducal cabinet of Cosimo, who beat out Peter the Great of Russia,
one of the many (and most extravagant) shoppers to visit the Netherlands in
search of collectibles. Shells were certainly the best preserved exotica avail-
able and therefore the most popular items for sale. Yet a wide variety of nat-
uralia, artzjicialia, and hybrid items bridging the two-finely painted
animal horns, artfully crafted fossils, ingeniously worked coral -could be
had for a price from Dutch collectors, who readily dispensed with their
holdings in precisely these years. Tourists in the Republic also made sure to
visit the splendid botanical and zoological gardens in Leiden, Amsterdam,
and Rotterdam, which were well stocked with imports from the East and

In venting Exoticirm 353


j4.3 Polychrome
tile panel, Delft (workshop
unknown), ca. 1700 (170 x
79 cm). Courtesy of the
Rijksmuseum, Amsterdam.

West Indies; and they called on the renowned physicians of Holland, who
possessed among the finest collections of simples and specimens in Europe.
Finally, the Dutch plied a trade in the actual cabinets that contained these
collectibles. Commonly constructed out of tropical woods or other imported
materials, and elaborately decorated with scenes meant to evoke the distant
lands from which the treasures within had derived, these cabinets offered
consumers yet one more opportunity to imagine the shape of the exotic
world -or at least the Dutch version thereof.I3

354 Benjamin Schmidt


The products of Dutch geography, which spanned so many genres,
forms, and styles, also spanned multiple communities of readers, collectors,
and consumers, reaching markets far and wide. This is easiest to see in the
category of printed sources, which circulated exceptionally broadly at this
time. Texts on the extra-European world-a seaman's narrative, for exam-
ple-might appear originally in inexpensive octavo or duodecimo, with
crude woodblock prints and the vaguest of maps. Yet these images might be
redone by the same or another publisher with expertly engraved plates,
detailed maps, and text printed in folio. Or the reverse: a text first offered in
massive, luxury editions might be pirated by a publisher working for the
lower-end market, who would offer an abridged edition at a favorable price.
Plates were commonly recycled as well, with or without an accompanying
text. Romeyn de Hooghe's Les Indes Orientales et Occidentales (1710), a tour
de force of engraved exotica executed in a striking, oblong format, furnished
images for any number of other texts that had little or nothing to do with de
Hooghe's original commission. 14 More basically, Dutch geographic texts
were recycled in translation. What is particularly impressive about this
process, certainly not uncommon in early modern publishing, was the
degree of control exercised by Dutch publishers over their product. Transla-
tions usually derived/rom the original publisher in the Republic, whose edi-
tions would then compete with, and generally prevail over, translations
undertaken by foreign publishers. German, French, and Latin works of
geography commonly came from Holland, where printers employed a core
of translators. The Royal Geographer of Restoration England, John Ogilby,
may have departed from this pattern somewhat by hiring his own translator,
yet he commonly ordered his prints from Amsterdam and produced a final
product that was (Ogilby's boastful claims of authorship notwithstanding)
undeniably Dutch. IS
In the case of other sources of Dutch geography, patterns of dissemination
are perhaps less plain-yet the fact of their dispersal throughout Europe is
abundantly clear. Brazilian landscapes by Frans Post shipped to Versailles,
part of a spectacular trove of Dutch Americana purchased by Louis XlV.
Tropical stilllifes by Albert Eckhout went to Frederick III of Denmark, gain-
ing a prime position in the royal collection in Copenhagen. And cartographic
watercolors from the venerable Vingboons atelier reached Rome by order of
Christina of Sweden, who abdicated her throne in Stockholm en route to
becoming a leading patron of Baroque arts and letters. Dutch images, like
Dutch texts, spread across the continent. They provided, in short, rich and
plentiful models for those who would contemplate the exotic world. 16
The sum of these sources and their dissemination constitutes what might
be called a geographic moment, which can be traced to the final decades of
the seventeenth century and early years of the eighteenth. This poses a sim-

Inventing Exoticism 355


pie pair of related questions: Why then-around 1700? And why there-
the Dutch Republic?l! That the Netherlands triumphantly concluded an
epic Eighty Years' War against Spain in 1648, and that their success in for-
eign policy had boded well for domestic affairs, cultural no less than eco-
nomic-the Dutch Golden Age-may explain in part the vibrancy of
Dutch geography, which was, in essence, a cultural export with economic
dimensions. Yet 1648 denotes not so much the dawn as the dusk-or per-
haps, pre-dusk dimming-of that Golden Age, which was, by most mea-
sures, in a state of decline by the wonder years of Dutch geography. What
had changed by the end of the war was the harshly polemical representations
of Habsburg "tyranny," as the Dutch chose to characterize Spanish govern-
ment both at home and abroad. By the second half of the century, with no
need to demonize the enemy, descriptions of Spain and its empire had
become more mild; Dutch geography lost its habitual Hispanophobic edge. IS
This shift accounts for questions of quality rather than quantity, however. It
fails to explain the Dutch dedication to worlds overseas in the first place, and
the assiduous concern of the new Republic with matters imperial.
The conclusion of the Netherlands' struggle against Spain did free up
ships and soldiers, which might suggest an expansion of colonial efforts-
and a parallel extension of colonial discourse in the form of geography. Yet
the very opposite would seem to be the case, at least as far as Dutch overseas
expansion is concerned. Dutch settlements in America collapsed, in fact, in
the years immediately following the Treaty of Munster (r648)-New Hol-
land (Brazil) succumbed to a joint Spanish-Portuguese armada in 1654,
while New Netherland fell to the duke of York only a decade later-and
the Dutch West India Company declared bankruptcy by 1664. In Asia,
though there were no such dramatic setbacks, the Republic experienced a
number of reversals and contractions and a loss of relative market share to
its European competitors. By the later seventeenth century especially, the
Dutch East India Company had given ground to the English and French;
and if the VOC (as the Company was called) still turned a profit, it did so
increasingly as middlemen in inter-Asian trade rather than masters of pan-
Asian domains. Rather than land-based, imperial ambitions, the Republic
cultivated more profit-minded, commercial strategies. In both the East and
West Indies, the Dutch peddled the products-the stuff-of the world.
And increasingly, they also plied a trade in the image of the world, which
took the form of geography.

The image of the world marketed by the Dutch came in many forms. Yet
this image, in virtually all of its permutations, shared certain qualities and
followed discernible patterns that may be said to characterize the Dutch
brand of geography flourishing circa 1700. These qualities and patterns,

Benjamin Schmidt
moreover, functioned to render this image-the Dutch production of the
world - agreeable and therefore salable to the broad, Europe-wide audi-
ence for Dutch geography. Taken together, they constitute an identifiable
and effective strategy of representation, which might be termed "exoticism."
The world according to Dutch geography was a rich, jumbled hodge-
podge of peoples and places, the creatures of one land and customs of
another slipping easily across borders and even continents. It was a res-
olutely disorderly world, not infrequently described as "strange," somewhat
paradoxically as "novel," or more prosaically as "marvelous." It burst with
social, cultural, and natural bric-a-brac, arranged with seemingly careless
disregard for order or provenance. Dutch geography was eclectic and
notably catholic: all manner of scenes and snippets of exotica, described by
all manner of European observer, were made welcome. Yet there was a
method to this madness, or at least a pattern to the chaotic clutter of Dutch
geography.
Dutch geography emphasized, firstly, variety. The mix-and-match qual-
ity of texts and images was often by design, and difference was accentu-
ated - not so much to distinguish the cultures and landscapes of the world
as to conflate them. Distinctions, like borders, were studiously erased. Here
a strategic shift becomes apparent, when works of geography from the later
seventeenth century are compared with those dating from just a few decades
earlier. Johan Nieuhof's masterly description of Brazil, penned originally in
the middle of the century-at the height, namely, of the Dutch West India
Company's (WIC) struggle against Portugal-appeared finally in 1682,
combined with the author's reports on the East as well as West Indies. In
place of the original, carefully focused narrative of the Republic's tenure,
and then decline, in South America, the reader now contends with a swirl of
exotic settings, ranging "from China to Peru." Orient and Occident com-
fortably commingle, at times with haphazard abandon. Brazilian flora segue
to Chinese fauna, while Jewish merchants of Recife bump up against
tobacco-addled natives of Malaysia. 19 In much the same spirit, an edition of
Willem Piso's Historiae naturalis Brasiliae-first introduced in 1648 by a
director of the WIC, who unabashedly endorsed the Netherlands' empire in
Brazil- was reissued after the fall of New Holland, carrying in this later
version a preface in praise of wonders and a newly appended, yet otherwise
incongruous, study of Asian naturalia. A fresh frontispiece engraved for this
second edition casually blends a heraldic Brazilian figure with a vaguely
Persian one-neither, to be sure, having much to do with the natural histo-
ries within. 20 This sort of inspired eclecticism is particularly embraced on the
frontispieces of (printed) Dutch geographies. The magisterial Ceremonies et
coutumes religieuses des tous les peuples du monde, published initially in Ams-
terdam in 1723, opens with a fabulous melting pot of intercontinental reli-

Inventing Exoticism 357


gion; Simon de Vries's Curieuse aenmerckingen der bysonderste Oost- en West
Indische verwonderens-waerdige dingen (Curious observations of the most
exceptional East and West Indian wonders, r682) entices the reader with a
baroque blend of global goods; and Joan Blaeu's monumental Atlas maior
(r662 and following) announces itself with an allegory of all the world's
parts, offering an iconographic image and a cartographic style that would
become standard for the next half century.21
Dutch geography collapsed distinctions, and, while efforts were certainly
made to compare and contrast, this was done in a peculiar "analogistic" style
that suggests, once again, a shift in discursive method. Whereas Hernan
Cortes, in the early sixteenth century, found the towers of Mexico evocative of
Seville and Granada-crucial loci, it should be stressed, of the Spanish
Reconquista-Amoldus Montanus by the r670s associated Mexican temples
with those of China and Japan. Montanus, that is, matched exotic with exotic;
Cortes, meanwhile, compared newly won Spanish domains with the soon-to-
be colonized province of New SpainY Dutch geography, more generally, cel-
ebrated the vastness of the world and conflated global space - which offers
yet another point of contrast, in this case with contempory geographic agen-
das. Thus, where an Athanasius Kircher or Joseph-Fran~ois Lafitau infused
the study of geography with a pronounced historical dimension-compar-
ing Native Americans or Ch'ing Chinese with the Ancients-an Amoldus
Montanus or Olfert Dapper pursued the "savage" (or "civilized") American
by looking to examples in Asia and Africa. Dutch cross-references tended to
be spatial, not temporal; Dutch geography expanded the globe rather than
excavated the past. Dapper (who, like Montanus, collaborated frequently
with the indefatigable Jacob van Meurs) exemplifies this trend in his very
impressive and extensive oeuvre: two tomes on Africa, an immense volume
dedicated to China, three lavish works on the Near East and South Asia, two
more sprawling accounts of the Mediterranean. All were done in the space of
two decades, and it is no wonder that engravings of "races" tend naturally to
collapse the different continents; that the term "Indian" is applied to Chinese
and American indigenes alike; that the "biggest," "richest," and "most fabu-
lous" marvels of the world seem to resurface in multiple exotic locales. How
could there not be a fair amount of geographic bleeding from text to text,
region to region, race to race-in some instances creating a stunning kalei-
doscope of global color - for this author of thousands of wonder-filled pages,
whose stream of publications would finally crest with the aptly titled, two-
volume anthology Dapperus exoticus curiosus? 23
The world fashioned by Dapper and his colleagues was unusually fluid and
supple. It was a decontextualized space, with no obvious center and with min-
imal orientation. Camelis de Bruyn, a superbly capable draftsman from The
Hague, took his sketch pad on a tour of the globe and recorded, in a single

Benjamin Schmidt
whirlwind volume, his impressions of Russia, Persia, Sri Lanka, and the East
Indies. In another, equally remarkable volume of travels, de Bruyn trans-
ported his readers to the Levant and North Africa, with stops en route in
Habsburg Austria and Italy. In both works, de Bruyn insisted on taking the
most indirect narrative routes imaginable, pausing occasionally to tidy things
up with chapters devoted to "Odd Matters": miscellany that merit otherwise
haphazard attention. 24 Arnoldus Montanus, who hardly ever left the bosom of
Holland, almost made a virtue of discursiveness-which provoked fierce
attack from competing (non-Dutch) geographers.25 Montanus had no reason
to worry: his very popular Atlas Japannensis spun the reader from inquests into
native flora to reflections on Buddhist dogma, from volcanic topography to
Tokugawa history, from snippets on local reptiles to the wisdom of Shinto
priests. It went through nine editions in French, German, English, and Dutch,
exasperating Montanus's critics - though not his faithful readers - for more
than half a century. (His America, if slightly less digressive in form, turned out
to be slightly less successful as well: a mere three editions in three languages.)
These and other examples of Dutch geography cultivated a meandering style
and flaunted a purposeful indeterminacy. They eschewed-once again, rela-
tive to other traditions of geography - the sort of "national" perspective that
had characterized earlier works by Hakluyt, Purchas, Ramusio, and Herrera.
Dutch geography lacked a sharp focus and avoided a straightforward narra-
tive. Many of the texts it produced were (and still are) all but unreadable. 26
Or rather, many of the texts resisted the sort of linear reading to which
their critics may have subjected them-though this may be the point. For if
Dutch geography discouraged easy, systematic study (this apropos of
printed, literary materials), it did encourage a more flexible, almost casual
style of perusal. Volumes of Dutch geography are, fundamentally, immense
and cumbersome objects. This pertains both to the length of the text-
many of the works produced in this period run nearly, and sometimes over,
a thousand pages -and to its format, which often was delivered in an
imposing folio. They do have outstanding indices, most include helpful
tables of contents, and they often have summary chapter headings. If readers
could only uncomfortably clasp one of the enormous products of Dutch
geography, then, they could easily browse the volume and readily taste the
diverse "morsels" of exotica (as one editor referred to them) offered within.
Prefatory materials sometimes advocated just such a cursory manner of
scanning the text. One author sheepishly confessed at the outset of his 938-
page survey of world religions that his book was wholly wanting of an orga-
nizing principle. Another cheerfully noted his role in gathering, and then
scattering, data for the soon-to-be overwhelmed reader, pointing out that no
single person could possibly master the harvest of exotica that he had reaped:
"Non omnia possumus omnes."27 Many of these editions sought to make mat-

Inventing Exoticism 359


ters easier on the reader by their generous inclusion of figures, engraved
plates, and pull-out maps, adding a visual component to the text that further
invited skimming. Indeed, the extensive illustration programs of printed
Dutch geographies also lent a fluidity, and even interchangeability, to these
sources. Plates produced for one volume typically reappeared in others, and
a reader could hardly be faulted for eliciting only the broadest and most
basic themes from the resulting, perhaps recycled product. Dutch geography
in its printed form-immense, attractive, elegantly illustrated, and pro-
grammatically superficial-may represent a first generation of coffeetable
books at the dawn of the age of coffee. 2H
Specimens of Dutch geography, whether prank or not-the Dutch word,
not easily translated, well conveys the quality of sumptuousness and ostenta-
tion inherent in so many of these objects-often drew attention to them-
selves. Books, images, and artifacts all had an importance in and of
themselves, which came at the expense of their subject - the swirling
melange of places to which they were ostensibly dedicated-and their nom-
inal authors. This, in turn, had the effect of effacing authorship, if not
authority. Which is not to suggest that the original voyagers, draftsmen, and
naturalists who produced Dutch geographies went unnamed; some did, if
many did not. Yet even when original authors were cited, the text had a way
of taking attention away from specific places and persons-Johan Nieuhof
in the East Indies, say-and directing it on the work itself. Again, a useful
contrast can be drawn with other samples of geographic literature. Dutch
texts of this period rarely have recourse to what Anthony Pagden has called
the "autoptic imagination": the device, so prevalent in early modern travel
narratives, of privileging eyewitness evidence and first-person presence
when describing new worlds. 29 Dutch texts, on the contrary, commonly
obscure the author, following his (and it nearly always is a he) perfunctory
opening remarks. This made the Indies the subject, rather than Nieuhofs
personal experience there. More generally, it allowed products of Dutch
geography to be more widely accessible-less rooted in a particular Dutch
experience. Broad dissemination came easily to these products, if piracy,
likewise, naturally followed. Thus could John Ogilby brashly claim author-
ship of Montanus's and Dapper's major works, which he repackaged and
incorporated into his multivolume "great atlases" of the exotic world. 30
Which brings up one more vital quality of Dutch geography: its sem-
blance of neutrality. Dutch descriptions of the world endeavored to be disin-
terested. They dispensed as far as feasible with the built-in biases of
geography, and they disguised where they could the Dutchness of their
product-yet another strategy meant to foster Europe-wide consumption.
Painters, accordingly, hid the fact of the Republic's prior presence in their
renderings of exotic landscapes-as Frans Post did in his Brazilian scenes,

Benjamin Schmidt
which as often as not highlighted the Portuguese inhabitants of this formerly
Dutch domain; and as Gerard van Edema did in his views of New York,
which he undertook almost exclusively for an English clientele. Texts, too,
went out of their way to be evenhanded. When Philippus Baldaeus com-
posed a geography of Malabar (India), he opened his account with a strik-
ingly inclusive survey of traffic to the region by "the Dutch, English, Moors,
and Portuguese." Evert Ysbrand Ides frankly acknowledged the political
purposes of his voyage from Moscow to Beijing (he traveled at the behest of
the Amsterdam regent, Nicolaes Witsen, in an embassy of the czar), yet
emphasized his chief "Obligation ... to impart to the curious World what I
saw and observ'd in my Journey"-fabulous exotica-rather than to relate
the political and commercial horse-trading done for his patrons. 3 l Dutch
geography chose not to narrate the story of the Republic's rise overseas, and
it conspicuously lacked an aggressively "national" focus. The late seven-
teenth-century Republic, it merits repeating, produced no Hakluyt or Her-
rera to sing its imperial glories. Indeed, Dutch geography in this period
avoided any taint of politics or hint of polemic-at least as far as European
rivalries were concerned. Studies of Japan gave reasonably sympathetic
accounts of the persecution of the Catholics - when penned by Calvinist
authors, no less - and of the expulsion of the Portuguese. Maps of the
Americas erred on the side of neutrality when it came to charting colonial
control. Even old rivals could be reconciled in the world of Dutch geogra-
phy. A remarkable vignette in one widely translated travelogue describes
how a Hollander and a "fine" Castilian met on a Moluccan island, where the
two discussed the benefits of "amity" among Christians. They agreed that
their nations, ancient and fierce rivals, might now fruitfully cooperate in the
conversion of the heathen. This friendly tete-a.-tete pointedly gives way to a
jaunty discourse on birds of paradise and other exotica of the bountiful, and
presumably mutually profitable, tropics. 32
Such soft "internationalism," wholly foreign to earlier Dutch descriptions
of Habsburg "tyranny" in the Indies, was not uncommon to the products of
Dutch geography manufactured around 1700. By this time, memories of the
epic struggle against Spain (concluded in 1648), of the highly charged
Anglo-Dutch Wars (waged chiefly in the 1650S and 1660s), and of the
French invasion of the Netherlands (1672) were quickly fading-or at least
mellowing enough to permit a depiction of the world that appeared, to most
European consumers, mostly inoffensive. Works of Dutch geography, despite
their abundance of "color," despite their tendency toward strangeness, and
despite their embracement of difference, often had a neutral, almost generic
quality to them. The potpourri of countries and customs, of peoples and
products, could come across finally as bland. Diffuse, digressive, often disori-
enting, sometimes recycled, purposely decontextualized: Dutch geography

Inventing Exoticism
ended up being specific to none and thus palatable to all. The exotic world
designed by the Dutch was a brand, ultimately, of very wide appeal.

Geography, whether local or global, has a double context: the place and time of
production, and the place and time of consumption. Comprehending both
of these contexts establishes a firmer foundation for understanding the nature of
the imagined world (geo) and its description (graphia), a step, in turn, toward
understanding that common substance of geography, the exotic. The Dutch
Republic emerged as the leading geographer of Europe around the turn of the
eighteenth century, a critical moment in the construction - imaginative as well
as real-of the modern world. At the advent of the age of empire, as England
and France especially intensified their expansion east and west, the Netherlands
assumed the role of chief purveyor of words, images, and ideas concerning the
globe. Europeans devoured the products of geography at the very moment that
they began to gobble up the colonial world. The Republic, meanwhile, set to the
task of manufacturing representations of that world. It is not that the Nether-
lands had no part in the expansion of overseas commerce at this time: the Dutch
continued to trade in Asia (if less so in America and Africa) and certainly traf-
ficked in imported non-European goods. Yet, by the late seventeenth and early
eighteenth century, there was a marked shift away from expansionist, imperial,
colonial projects-and an upsurge in the production of geography. The Repub-
lic, in other words, became less and less engaged in land-based empire, while
becoming more and more involved in an empire of images: geography. Increas-
ingly, the Dutch marketed, rather than colonized, the world.
The Dutch plied their trade in geography in multiple profitable ways. They
sold, most basically, the actual wares of geography: books, maps, paintings,
prints, and artifacts, which retailed at a considerable profit. Folio volumes - to
cite but one example-fetched spectacular prices in the late seventeenth cen-
tury, especially when sold in large orders. Six hundred readers signed on for
Frans;ois Valentijn's Oud en Nieuw Oost-Indiifn at a cost of 38 guilders per five-
volume set, netting the publisher a phenomenal gross of 20,000 guilders-this,
when the average VOC sailor earned 120 guilders per annum. Servants of the
Company, nevertheless, might supplement their wages by dealing in exotica on
the side-such as the birds of paradise abundant in the Moluccas, or the lucra-
tive shell collections of Rumphius, which the author sold, replenished, then
sold again. Geography was good business. 33 Yet, more than the mere objects of
geography, the Dutch sold the very things of the world in mimetic form - rep-
resentations of actual, exotic merchandise - in their breathless descriptions
and enticing promotions of those foreign products lately available in Europe.
Dapper lavished loving attention on the Damascus bazaar, where the "choicest
and dearest" things-silk garments, gold and silver jewels, sables, pearls, "and
also slaves of both sexes"-could be had. One van Meurs-produced print

Benjamin Schmidt
&J&M i b . 4 "Diverse Sorts of Bonnets / Verscheiden slagen van Bonnetten."
Engraving from John Ogilby, Atlas Chinensis (London, 1671), based on Olfert Dap-
per, Gedenckwaerdig bedryf der Nederlansche Oost-Indische Maetschappye, op de kwte
en in het keixerrzjk van Taising of Sina (Amsterdam, 1670). Courtesy of the William
Andrews Clark Memorial Library, University of California, Los Angeles.
arranged a selection of Oriental goods catalog-style, the adjacent "copy" (as it
were) supplying the reader with nearly every commercial detail save the shop
address where the commodities could be purchased (fig. 14.4). Other prints
advertised other costly imports with monumental reproductions: the "Clove
Tree" in Nieuhof's Indien, depicted literally on a marble pedestal; or
Kaempfer's homage to the tea plant and its elaborate preparation, engraved as
an impressive pull-out feature to his study of Japan. Frontispieces (and paint-
ings) exhibit almost without fail the most stupendous riches of the globe-
Japanese lacquerwood, Brazilian sugarcane, African ivory, Indian incense,
Chinese porcelain, Persian carpets - which virtually spill out of the picture
frame and onto the viewer's lap (see previous fig. 14.1).34
Most of all, though, the Dutch sold an idea of the globe, and that idea
made the non-European world seem immensely, alluringly, and ineluctably
desirable. For the world formulated by Dutch geography was not simply full
of commodities, open to trade and potential profit. It was, more merely full
of wonder, engagingly disheveled and charmingly chaotic. It was also can-
nily decentered and politically decontextualized, which transformed it, as a
result, into a realm cleansed of commercial rivalries and refreshingly clear of
colonial polemics. Instead of a hotly contested space of exploding imperial
antagonisms, the non-European world created by the Dutch abounded with
curiosities, diversions, and delight. It was, in the end, not so much the prod-
ucts of the world as the world as product that captivated the armchair trav-
eler. The world of Dutch geography beckoned the consumer with what one
magnificent tome of Asiana succinctly pronounced "the pleasures of the
exotic": Amoenitates exoticae.
All of this was by design. At the turn of the eighteenth century, as Euro-
pean expansion and imperial jostling began significantly to intensify, the
architects of Dutch geography constructed a world - pursued a marketing
strategy, as it were-of extraordinarily wide appeal. The world fashioned
by the Dutch was not only enticing; it was broadly inviting and readily
agreeable to the avid consumers of late Baroque and early Enlightenment
geography. The Dutch represented the world circa 1700 as a supremely
seductive and wonderfully accessible space. To Europeans, at least, this
world appeared enchantingly, amenably, and reassuringly exotic.

1. Georgius Everhardus Rumphius, D'Amboinsche rariteitkamer (Amsterdam: Fran~ois

Halma, 1705); idem, Het Amboinsche Kruid-boek (Amsterdam: Fran~ois Changuion, 1741),
from which th e quotation derives (n.p.); Engelbert Kaempfer, Amoenitatum exoticarum
politico-physico-medicarum fasciculi V, quibus continentur variae relationes, observationes &
descriptiones rerum Persicarum & Ulterioris Asiae (Lemgo: H . W. Meyer, 1712); and Arnoldus

Benjamin Schmidt
Montanus, Gedenkwaerdige gesantschappen der Oost-Indische Maetschappy in 't Vereenigde Ned-
erland, aen de Kaisaren van Japan (Amsterdam: Jacob van Meurs, I669), for which see the Eng-
lish edition, Atlas Japannensis (London: John Ogilby, I670), 488. For Montanus's predilection
for wonders, see also idem, De wonderen van 't Oosten ofte de beschrijving en oorlogsdaden van
Oud en Nieuw Oost-Indien (Amsterdam: Cornel is Jansz, I655); and idem, Oud en Nieuw Oost-
Indien (Amsterdam: Cornel is Jans[z] Zwol, [I680]).
2. The term first comes into usage in the literature of natural history, a notably precocious
example being Carolus Clusius's [Charles de L'Ecluse] Exoticorum libri decem (Leiden: Plan-
tin Office [Raphelengius], I605). And when "exotic" does appear in print, it tends to be in
Latin: rare is an instance of the word in Dutch, German, or English vernacular letters.
3. The literature on early modern "exoticism" per se is virtually nonexistent-though the
subject is skirted in the expanding field of "wonder" studies. Of the many recent titles on the
latter topic, see Stephen J. Greenblatt, Marvelous Possessions: The Wonder of the New World
(Chicago: University of Chicago Press, I99I); Lorraine Daston and Katharine Park, Wonders
and the Order of Nature, 1150-1750 (New York: Zone Books, 1998); and Mary Baine Camp-
bell, Wonder and Science: Imagining Worlds in Early Modern Europe (Ithaca, N.Y.: Cornell Uni-
versity Press, 2000). Two very useful and provocative studies, focused on texts somewhat
earlier and slightly later, respectively, than those cited in this essay, are Mary Baine Campbell,
The Witness and the Other World: Exotic European Travel Writing, 400-1600 (Ithaca, N.Y.: Cor-
nell University Press, I988), and Peter Mason, Infelicities: Representations of the Exotic (Balti-
more: Johns Hopkins University Press, I998). For more generalized overviews, see Bernard
Smith, Imagining the Pacific: In the Wake of the Cook Voyages (New Haven, Conn.: Yale Uni-
versity Press, I992), I-39 ("Art in the Service of Science and Travel"); and G. S. Rousseau and
Roy Porter, eds. Exoticism in the Enlightenment (Manchester: Manchester University Press,
I990), especially the editors' introduction, I -22.
4. P[etrus] de Lange, Wonderen des werelds (Amsterdam: Marcus Willemsz Doornick,
I67I), sig. A2v. Cf. also the translator's preface to Thevenot's Travels: "He [Thevenot] therein
gives you a succinct account of all that is curious in every place, and a character of the several
people. In short, he says enough to give one a reasonable information of those countreys, and
not too much, to cloy the reader with the repetition of what hath seen before" (Jean de
Thevenot, The Travels of Monsieur de Thevenot into the Levant, trans. Archibald Lovell [Lon-
don: Henry Clark, I687], n.p.).
5. Arnoldus Montanus, De nieuwe en onbekende weereld: of Beschryving van America en 't
Zuid-land (Amsterdam: Jacob van Meurs, I67I). On Rumphius and his shells, see the superb
introduction by E. M. Beekman to Georgius Everhardus Rumphius, The Ambonese Curiosity
Cabinet, trans. and ed. E. M. Beekman (New Haven, Conn.: Yale University Press, I999).
6. The classic bibliographies of Dutch geography-Po A. Tiele, Nederlandsche bibliogra-
phie van land- en volkenkunde (Amsterdam: Frederik Muller, I884), and idem, Memoire bibli-
ographique sur les journaux des navigateurs Neerlandais riimprimes dans les collections de De Bry
et de Hulsius, et dan les collections hollandais du XVIIe siecle (Amsterdam: Frederick Muller,
I867)- tend to adopt a less generous view of the range of "geography" than do I in this essay,
and they offer, therefore, only moderate guidance. For works with any relevance to the
Dutch experience in Asia, however, see John Landwehr, VOC: A Bibliography of Publications
Relating to the Dutch East India Company, 1602-1800 (Utrecht: HES, I99I); and, for books
that make mention of the Americas-which applies to the geographies of Cluverius, Vare-
nius, and Hornius-see John Alden and Dennis Landis, eds., European Americana: A
Chronological Guide to Works Printed in Europe Relating to the Americas, 1493-1750,6 vols.
(New York : Readex Books, I980-I997). More broadly useful is John Landwehr, Studies in
Dutch Books with Coloured Plates Published 1662-1875: Natural History, Topography and Travel
Costumes and Uniforms (The Hague: Junk, I976).

Inventing Exoticism
7. Joost Schouten, Notitie van de situatie, regeeringe, macht, religie, costuymen, traffijcquen,
ende andere remercquable saecken, des Coninghrtjcks Siam (The Hague: Aert Meuris, 1638),
which appeared in English, German, French, Latin, and Swedish, mostly in the second half
of the century; Johan Nieuhof, Gedenkweerdige Brasiliaense zee- en lantreize (Amsterdam:
Widow of Jacob van Meurs, 1682); Philippus Baldaeus, Naauwkeurige beschryvinge van Mal-
abar en Choromandel, der zelver aan grenzende ryken, en het machtige eyland Ceylon (Amster-
dam: Johannes Janssonius van Waasberge and Johannes van Someren, 1672); and cf.
Awnsham and John Churchill, eds., A collection of voyages and travels, 4 vols. (London:
Awnsham and John Churchill, 1704).
8. On all three-and on the decennium mirabllius more generally-see David Freedberg,
"Science, Commerce, and Art: Neglected Topics at the Junction of History and Art History,"
in Art in History/History in Art: Studies in Seventeenth-Century Dutch Culture, ed. David Freed-
berg and Jan de Vries (Santa Monica, Calif.: Getty Center for the History of Art and the
Humanities, 1991),376-428. The Hortus Malabaricus first appeared in 1678, with complete
editions following in 1683 and 1703.
9. Garrelt Verhoeven and Piet Verkruijsse, eds., Iovmael ofie gedenkwaerdige beschrijvinghe
vande Oost-Indishe reyse van Willem Ysbrantsz. Bontekoe van Hoom: Descriptieve bibliographie
1646-1996 (Zutphen: Walburg, 1996); Ogier Ghislain de Busbecq, Itinera constantinopoli-
tanum (Antwerp: Christopher Plantin, 1581), which came out in a much expanded version
published by Elsevier (Leiden) in 1633 that served, in turn, as the model for many later sev-
enteenth- and eighteenth-century editions; Cornelis de Bruyn, Reizen van Comelis de Bruyn,
door de vermaardste deelen van Klein Asia, de eylanden Scio, Rhodus, Cyprus, Metelino, Stanchio,
&tc mitsgaders de voomaamste steden van Egypten, Syrien en Palestina (Delft: H. van Kroon-
eveld, 1698); and idem, Comelis de Bruins Reizen over Moskovie, door Persie en Indie (Amster-
dam: R. and G. Wetstein, }. Oosterwyk, H. van de Gaete, 1714)' For the travel anthologies,
see Landwehr, VOC, esp. 99 - 133.
10. Balthasar Bekker, De betoverde weereld, 4 vols. (Amsterdam: Daniel van den Dalen,

169 I - 93); and J. F. Bernard et aI., eds., Ceremonies et coutumes religieuses des tous les peuples du
monde, 8 vols. (Amsterdam: J. F. Bernard, 1723-43)' For examples of the sort of "literary
Kunstkammem" that I ha ve in mind, see Simon de Vries, D'edelste tijdkortingh der weet-geerige
verstanden: of De groote hlstonsche rariteit-kamer der sonderlinghste natuerlijcke en boven natuer-
lijcke saecken, geschiedenissen en voorvallen van allerley sfagh, 3 vols. (Amsterdam: Jan Bouman,
1682-95); idem, Wonderen soo aen als in, en wonder-gevallen soo op als ontrent de zeeen, rivieren,
meiren, poelen en fonteynen (Amsterdam, 1687); and de Lange, Wonderen.
I I. Dutch-produced atlases are excellently cataloged in Cornelis Koeman, Atlantes Neer-

landici: Bibliography of Terrestrial, Maritime and Celestial Atlases and Pilot Books, Published in
the Netherlands up to 1880,6 vols. (Amsterdam: Theatrum Orbis Terrarum, 1967-85); for
globes, see P. C. J. van der Krogt, Globi Neerlandici: The Production of Globes in the Low Coun-
tries (Utrecht: HES, 1993). On Dutch cartographic materials more generally, see Kees Zand-
vliet, De groote waereld in 't kleen geschildert: Nederlandse kartografie tussen de middeleeuwen en
de industriele revolutie (Alphen aan den Rijn: Canaletto, 1985); and on the exotic world more
particularly, see idem, Mapping for Money: Maps, Plans and Topographic Paintings and Their
Role in Dutch Overseas Expansion during the Sixteenth and Seventeenth Centuries (Amsterdam:
Batavian Lion International, 1998), which discusses "floor" maps on 21 I.
12. The literature on visual exotica is uneven. For Post, see Joaquim de Sousa-Leao, Frans
Post, 1612-1680 (Amsterdam: A. L. van Gendt, 1973); and P. J. P. Whitehead and M. Boese-
man, A Portrait of Dutch Seventeenth-Century Brazil: Animals, Plants and People by the ArtIsts of
John Maurits of Nassau, Royal Dutch Academy of Sciences, Natural History Monographs, 2nd
ser., vol. 87 (Amsterdam: North Holland, 1989), which covers Brazilian iconography in toto
and, in doing so, details the oeuvre of Albert Eckhout, Post's colleague who specialized in

Benjamin Schmidt
exotic still lifes and portraits. Valkenburg, Beeckman, and van Edema are still in need of
biographers and catalogers, but see the brief treatment of Valkenburg in C . P. van Eeghen,
"Dirk Valkenburg: Boekhouder-schrijver-kunstschilder voor Jonas Witsen," Oud Holland 61
(1946): 58-69. On exotic themes in decorative arts, see the relevant chapters of Hugh Honour,
The European Vision of America (Cleveland: Cleveland Museum of Art, 1975).
13. Patterns of Dutch collecting are surveyed in Ellinoor Bergvelt et aI., De wereld binnen
handbereik: Nederlandse kunst- en rariteitenverzamelingen, 1585-1735, 2 vols. (Zwolle: Waan-
ders, 1992), and see especially the essays of Jaap van der Veen, "'Dit klain Vertrek bevat een
Weereld vol gewoel': Negentig Amsterdammers en hun kabinetten," 232-258; K. van
Berkel, "Citaten uit het boek der natuur: Zeventiende-eeuwse Nederlandse naturalien kabin-
netter en de ontwikkeling van de natuurwetenschap," 169-191; and Roelof van Gelder, "De
wereld binnen handbereik: Ned erlandse kunst- en rariteitenverzamelingen, 1585-1735,"
15-38, which notes the unique Dutch habit of selling collections to foreign buyers (this in the
final decades of the seventeenth century). On "hybrid" exotica-and for a rich inquiry into
collecting and exotica more generally-see Martin Kemp, "'Wrought by No Artist's Hand':
The Natural, the Artificial, the Exotic, and the Scientific in Some Artifacts from the Renais-
sance," in Reframing the Renaissance: Visual Culture in Europe and Latin America, 1450-1650,
ed. Claire Farago (New Haven, Conn.: Yale University Press, 1995), 177-196.
14- Romeyn de Hooghe, Les Indes Orientales et Occidentales et autres lieux (Leiden: Pieter
van der Aa, 1710), and see also (to cite but one example) Wouter Schouten, Oost-Indische voy-
agie, vervattende veel voorname voorvallen en ongemeene vreemde geschiedenissen, bloedige zee- en
landt-gevechten tegen de Portugeesen en Makassaren; belegering, en verovering van veel voorname
steden en kasteelen (Amsterdam: Jacob van Meurs and Johannes van Someren, 1676), which
contains selections of de Hooghe's (in this case, wholly irrelevant) images in some editions.
15. Ogilby claimed authorship of a number of Dutch works-most notoriously, perhaps,
Montanus's America-and, even when he did relinquish claims of authority, he sometimes
cited original authors inaccurately (ironically giving credit to Montanus for Dapper's Atlas
Chinensis; see below). For a case study of the intricacies of publishing early modern geogra-
phy, see Isabella H. van Eeghen, "Arnold us Montanus's book on Japan," Quaerendo 2 (1972):
250-272; and compare Katherine S. Van Eerde,John Ogilby and the Taste of His Times (Folke-
stone: Dawson & Sons, 1976),95-122, which seems to have overlooked the Dutch sources of
Ogilby's "great atlases."
It may also be worth pointing out how well printers in the Republic themselves poached
and appropriated works of geography not originally written in Dutch-as was the case with
Charles de Rochefort's Histoire naturelle et morale des iles Antilles de l'Amerique (Rotterdam:
Arnout Leers, 1658), the Rotterdam editions of which became signal texts for later printers.
See Everett C. Wilkie Jr., "The Authorship and Purpose of the 'Histoire naturelle et morale
des iles Antilles,'" Harvard Library Bulletin, 2nd ser., no. 3 (1991): 26-84-
16. On Post and Eckhout, see Whitehead, Dutch Seventeenth-Century Brazil, 162- 193 and
passim; for the so-called "Christina atlas," see Zandvliet, Mappingfor Money, 179 - 180.
17. Freedberg poses this question implicitly, in identifying a surge of Dutch natural-his-
tory writing in the later seventeenth century ("Science, Commerce, and Art"). This is also a
theme broached by V. D. Roeper and G. 1. D. Wildeman, Reizen op papier: Journalen en
reisverslagen van Nederlandse ontdekkingsreizigers, kooplieden en avonturiers, Jaarboek van het
Nederlands Scheepvaart Museum (Zutphen: Walburg, 1996), which pays well deserved
attention to the remarkable work of van Meurs and other Dutch printers of travel literature.
18. This pertains particularly to representations of America. See Benjamin Schmidt,
"Tyranny Abroad: The Dutch Revolt and the Invention of America," De Zeventiende Eeuw
I I (1995): 161-174; and idem, "Exotic Allies: The Dutch-Chilean Encounter and the (Failed)

Conquest of America," Renaissance Quarterly 52 (1999): 440-473.

Inventing Exoticism
19. Johan Nieuhof, Gedenkwaerdige zee en lantreize door de voornaemste landschappen van
West en Oostindien, 2 pts. (Amsterdam: Widow of Jacob van Meurs, 1682), and see I: 20- 38
and 195-21 I for Nieuhofs (or his publisher's) conflation of exotic naturalia, and I: 2II -226
for his survey (and confusion) of exotic "races."
20. Willem Piso et aI., Historia naturalis Brasiliae (Leiden: F. Haack, and Amsterdam: L.
Elsevier, 1648); and idem, De Indiae utriusque re naturali et medica libri quatuordecim (Ams-
terdam: L. & D. Elsevier, 1658).
21. Bernard, Ceremonies et coutumes (and note how the first Paris imprint of 1741 actually
scotched this engraved frontispiece in favor of a more "Catholic" image of the Church);
Simon de Vries, Curieuse aenmerckingen del' bysonderste Oost- en West Indische verwonderens-
waerdige dingen, 4 vols. (Utrecht: Johannes Ribbius, 1682); and Joan Blaeu, Atlas maior, sive
Cosmographia Balviana, I I vols. (Amsterdam: Joan Blaeu, 1662).
22. Cf. Hernan Cortes, Letters from Mexico, trans. and ed. Anthony Pagden, 2nd ed . (New
Haven, Conn.: Yale University Press, 1986),67, 102, 105; and Montanus,fapan, 91.
23. Olfert Dapper's magnificent oeuvre began, it should be added, with the first-ever
Dutch translation of Herodotus (1665). See further Olfert Dapper, Naukeurige beschrijvinge
der Afrikaensche gewesten, van Egypten, Barbaryen, Libyen, Biledulgerid, Negroslant, Guinea,
Ethiopien, Abyssinie (Amsterdam: Jacob van Meurs, 1668); idem, Naukeurige beschryvinge der
AJi'ikaensche eylanden (Amsterdam: Jacob van Meurs, 1668); idem, Gedenckwaerdig bedryf dey
Nederlansche Oost-Indische Maetschappye, op de kuste en in het keizemjk van Taising of Sina
(Amsterdam: Jacob van Meurs, 1670); idem, Asia: of Naukeurige beschryving van het rijk des
Graoten Mogols, en een groote gedeelte van Indien (Amsterdam: Jacob van Meurs, 1672); idem,
Naukeurige beschryving van gantsch Syrie, en Palestyn of Heilige Lant (Amsterdam: Jacob van
Meurs, 1677); idem, Naukeurige beschryving van Asie [sic] behelsende de gewesten van
Mesopatamie, Babylonie, Assyrie, Anotolie of Klein Asie (Amsterdam: Jacob van Meurs, 1680);
idem, Naukeurige beschnjving del' eilanden, in de archipel del' Middel/andsche zee, in en ontrent
dezelv, gelegen (Amsterdam: Wolfgang, Waesberge, Boom, van Someren, and Goethals, r688);
idem, Naukeurige beschrijvinge van Morea, eertljds Peloponnesus en de eilanden gelegen onder de
kusten van Morea en binnen de Golf van Venetien (Amsterdam: Wolfgang, Waesberge, Boom,
van Someren, and Goethals, 1688); and J. C. Mannling, ed., Dapperus exoticus curiosus, 2 vols.
(Frankfurt and Leipzig: M. Rohrlachs, 1717-18).
24. BTUyn, Moskovie; and idem, Klein Asia, in which Chapter 25 is among those expressly
dedicated to miscellany. On BTUyn, see Jan Willem Drijvers, Jan de Hond, and Heleen San-
cisi- Weerdenburg, eds., 'Ik hadde de nieusgiengheid': De reizen van Cornelis de Bruljn (ca.
1652-1727) (Leiden: Ex Oriente Lux, and Leuven: Peeters, 1997).
25. See, for example, the introduction by the translator to Kaempfer's history of Japan,
which complains of the "large digressions" of Montanus and of the lavish plates, which are
dismissed as "the greatest embellishments": Engelbert Kaempfer, The history of Japan, trans.
J[ohann] G[aspar] Scheahzer (London: J. G. Scheuchzer, 1727), xliij-iv.
26. Smith (Imagining the Pacific, 23 - 24) makes a similar point regarding narrative order-
ing, though in reference to visual sources, specifically, of Dutch geography. Much the same
can be said, I wish to argue, for the literary (no less than visual) texts produced at this time.
27. De Lange, Wonderen, sig. A2v, on bite-size exotica; HoannesJ A[ysma], Spiegel der
Sibyl/en, van vierderley vertooningen (Amsterdam: J. Aysma, 1685), sig. "'3r, on disorderly orga-
nization ("in weynig regulen"); and Johan Nyenborgh, Tooneel der Ambachten: of Den winckel
der handtwercken en konsten (Groningen: Jacob Sipkes, 1659), sig. "'2, where the author quips,
in Dutch as well as Latin, "Een eenig man, niet alles kan" (No one can do it all).
28. A number of authors were explicit about the importance of a visual component in
geography. Kaempfer noted in his preface that "copper engravings also needed to be made
[for this book], since exotica are very difficult to comprehend without the help of clarifying

Benjamin Schmidt
illustrations": Kaempfer, Amoenitatum exoticarum, n.p., and see the English translation
(which I have altered slightly) in Engelbert Kaempfer, Exotic Pleasures: Fascicle III, Curious
Scientific and Medical Observations, trans. Robert W. Carrubba, Library of Renaissance
Humanism (Carbondale: Southern Illinois University Press, 1996), xix. The English intro-
duction to Bernard's Ceremonies likewise remarks on the essential place of pictures in geogra-
phy: "no subjects stand more in need of illustration than these [geography and travel
literature]; so hardly any have been ever set off with such truth and advantage." See J. F.
Bernard, The ceremonies and religious customs of the various nations of the known world, 6 vols.
(London: William Jackson for Claude du Bosc, 1733-37) vol. 4, viii.
29. Anthony Pagden, European Encounters with. the New World: From Renaissance to
Romanticism (New Haven, Conn.: Yale University Press, 1993),51-87. Meanwhile, the "I-
witness" (to adopt Clifford Geertz's pun) was making great headway in contemporary fic-
tional prose -Swift's Gulliver, Defoe's Crusoe, and Montesquieu's Uzbek all claim
first-person authority-just as Dutch editors were deemphasizing the traveler's position in
the text. On Nieuhofs Asian narrative, which was much worked over by his Amsterdam
publisher (van Meurs), see Roeper and Wildeman, Reizen op papier, 98.
30. Cf. Van Eerde,John Ogilby, 95- 122, and, more generally, Adrian Johns, The Nature of
the Book: Print and Knowledge in the Making (Chicago: University of Chicago Press, 1998),
which, inter alia, seems also to misread Ogilby's title pages.
31. Baldaeus, Malabar en Choromandel (and cf. Churchill, Voyages and travels, 572); Evert
Ysbrants Ides, Drie1'arige reize naar China (Amsterdam: Fran~ois Halma, 1704) (quotation
from English edition: Three years travels from Moscow over-land to China [London: W. Free-
man et aI., 1706], sig. A2v).
32. Schouten, Oost-Indische voyagie, 44-45. If one had to point to an "enemy" in Dutch
geography of ca. 1700, it would be the Portuguese, an antagonist widely acceptable to the
"new" imperialists of Enlightenment Europe. Virtually no text produced in the Republic was
ever translated into Portuguese, in any event, nor were the products of Dutch geography
apparently marketed for Portuguese consumption.
33. Fran~ois Valentijn, Oud en Nieuw Oost-Indien, 5 vols. (Amsterdam: Gerard Onder de
Linden and Dordrecht: Joannes van Braam, 1724-26), which includes a subsciption list of
some six hundred buyers. On prices of luxury books, see Landwehr, Studies in Dutch Books
(81 for Valentijn's geography). Data on wages-from 9 to I I Dutch guilders per month for a
VOC sailor (or soldier) and about twice that amount for an officer-are taken from Femme
S. Gaastra, De geschiedenis van de VOC (Zutphen: Walburg, 1991),91; and see also Jan de
Vries and Ad van der Woude, The First Modern Economy: Success, Failure and Perseverance of
the Dutch Economy, 1500-1815 (Cambridge: Cambridge University Press, 1997),607-32'
34. Dapper, Syrie en Palestyn, 19; idem, Gedenckwaerdig bedryf, plate facing 459 ("Ver-
scheiden slagen van bonetten"); Nieuhof, West en Oostindien, 2: plate facing 31 ("Nagel-
boom"); Kaempfer, De beschryving van Japan: behelsende een verhaal van den ouden en
tegenwoordigne staat en regeering van dat rYk (The Hague: P. Gosse and J. Neaulme, and Ams-
terdam: Balthasar Lakeman, 1729), plate no. 39.

Inventing Exoticism
Shopping for Instruments
in Paris and London

JAMES A. BENNETT

On 8 December 1768, Jean Bernoulli wrote from London:

I've been here in London for eight days. I am still not able to tell you
about astronomers or observatories, but I will share the pleasant surprise
which strikes an astronomer walking through the streets of this capital.
You have surely heard talk of the richness and brilliance of the shops of
London, but I doubt whether you can imagine how much astronomy
contributes to the beauty of the spectacle: London has a great many opti-
cians; the shops of these artists are full of refracting and reflecting tele-
scopes, octants, &c. All these instruments, ranged and set out with care,
strike the eye at the same time as imposing reflections on the mind. 1

He was astonished, and was confident that his correspondent would share
his astonishment, that he could continue his report on the state of astron-
omy in Europe without yet having visited such conventional sites as obser-
vatories, academies, and the homes of astronomers. He was not surprised
that instrument making flourished in London: that was well known to any-
one active in practical astronomy in the eighteenth century, when London
instruments were being used in observatories all over Europe. His corre-
spondent also was familiar with this, as Bernoulli introduces the famous
names as though he had met in the flesh some shared object of distant
curiosity and admiration ("You know very well the name of M. DOL-
LOND, the celebrated artist").2 Instead what surprised him was the com-
mercial context for all this astronomy: the vulgar but dynamic and
exhilarating world of the London shops -of Fleet Street, the Strand, Hay-
market, and Piccadilly. What business did astronomy have to display its
apparatus in such a context? And the display was not confined to the
famous names he had known from their publications and exported instru-
ments: there were evidently a great many more makers of unknown reputa-
tions, but whose shops seemed similarly to be filled with organized presen-
tations of telescopes and octants. It was clear from the very first impression
that the experience of an astronomer shopping in London would be very dif-
ferent from one in Paris.
We have become familiar with the idea that over the course of a century
or so, a "public" science had been established by the later eighteenth century,
and the ingredients commonly cited in its formation include societies, cof-
feehouses, lecture courses, museums, polite or rational entertainments, and
popular books. This chapter will look at shopping for science in the seven-
teenth and eighteenth centuries, at the nature of the interaction between
buyer and retailer, and at how and where their transaction was conducted.
We will use the experiences of foreign visitors: they notice and remark on
what strikes them as unexpected and distinctive, drawing attention to things
that locals pass over as unremarkable and taken for granted. 3

MONCONYS, VON UFFENBACH, AND LALANDE IN LONDON

Our first shopping visitor to London arrived in May 1663. He was Balthazar
de Monconys, who was accompanying the duc de Chevreuse on a tour of
Europe at the behest of his father, the duc de Luynes. 4 Monconys was a
member of the Montmor Academy in Paris and was keen to meet fellows of
the Royal Society: he tried to find Henry Oldenburg as soon as he arrived in
London. He was an active, enthusiastic, and perceptive observer of the Eng-
lish scene, who deliberately set out to experience as much as he could man-
age, filling his days with the rich and colorful life of London - the manners
of the court, the bookshops of St. Paul's, the workshops of instrument mak-
ers, the public prize fighting and bear baiting. More than anything he was
taken by all kinds of "ingenuity" in London, by the activities of the Royal
Society, and the inventions, experiments, and "secrets" of its fellows. His
chief guides were Oldenburg, Robert Moray and William Brouncker.
On his first day in London, before he had met any of these guides, he
bought a telescope from a maker close to the Royal Exchange. In the days
that followed he visited more shops, bought more telescopes, and on his
third day in London, before he had found Oldenburg, he made his way to
Richard Reeve's workshop in Longacre, to which he had been particularly
recommended. No telescopes were in stock, such was the demand, but Mon-
conys was shown something that astonished him. He attended a demonstra-
tion of a magic lantern. It is clear from his careful description that he had
never seen such a thing before-indeed his description became one of the
very first published accounts of the magic lantern. He has no shortcut vocab-

Shopping for Instruments in Paris and London 37 1


ulary to deploy, but must describe the process with deliberation; the slide, for
example, is "a glass 'leaf on which the objects are painted."5 Not only does
Monconys give us what seems to be the first recorded English instance of a
magic lantern, but more important for our purposes, his account shows that
more things could take place in the shops of instrument makers than buying
and selling: already they could be the sites of public demonstrations in nat-
ural philosophy.
Monconys continued to visit shops and to buy instruments, while becom-
ing caught up in the life of the London virtuosi. He began to attend meetings
of the Royal Society on 13 May-an important occasion for the new society,
since it was the day when the royal charter was read to the council- and
returned to his lodgings by way of Reeve's shop, to see his range of micro-
scopes. He was back at the society the following week. He met Robert Boyle
then, and visited and admired Boyle's laboratory, noting his very good tele-
scopes and two excellent microscopes. He attended the society again on 27
May, and this time the Journal Book records him contributing to the discus-
sion. He visits other laboratories and workshops, and ventures as far as
Stratford Bow to visit the inventor Johann Kiiffeler, where he saw a self-reg-
ulating "philosophical furnace" for experiments in natural philosophy.
Monconys records all these encounters with evident relish and enthusi-
asm. He was on an odyssey of discovery. Boyle tells him about dissecting the
frozen eye of a bull and describes in detail an experiment of his demonstrat-
ing the weight of the air. Monconys buys sixty-six examples of what he calls
"glass teardrops," known in England as "Prince Rupert's drops"-small
glass vials with unusual mechanical properties that the Royal Society had
investigated at the king's request and of which Robert Hooke was to publish
an explanation in his Micrographia of 1665. The following day Monconys
sees a special design of pendulum clock at Moray's house, goes with him to
the laboratory of Nicaise Ie Fevre, former chemist at the Jardin du Roi and
now apothecary to Charles II, and from there to Oldenburg to discuss the
last meeting of the Royal Society. A few days later he buys eighty-four more
of the glass teardrops.
Monconys was delighted to be involved in this activity. When Samuel
Sorbiere, secretary of the Montmor Academy, visited London at the same
time and sought out Monconys, he says, "I found him in his element,
immersed in the conversation of natural philosophers and breathing nothing
but machines and new experiments"6 The shops of certain makers were an
integral part of this world. When Monconys returns to Reeve's to see his
microscopes, he also engages him in discussion and learns, for example,
something of the technique of working gems.
Other instrument makers visited by Monconys were Ralph Greatorex, to
whom he entrusted a loadstone for "arming"- that is, fitting with iron pole

372 James A. Bennett


pieces to enhance its magnetic properties-and Anthony Thompson. He
wanted Thompson to make him an example of Christopher Wren's per-
spectograph, even though he could not have it ready in less than fifteen days,
by which time Monconys would have left England. Monconys arranged for
Oldenburg to send it on to him, which he did in November 1664, but it
arrived in Paris shortly after Monconys died.?
There are a number of things to be learned from the record of Monconys's
experience in London. He was keen to acquire London instruments, and
there were a number of makers to choose from. Among those Monconys vis-
its, some-in particular Reeve, Greatorex, and Thompson-are clearly
linked to members of the Royal Society and are an integral part of the virtuosi
culture of London. Monconys is aware that Reeve's collaboration is central to
the efforts of Goddard, Paul Neile, Moray, and Wren to develop the long
refractor, efforts that had attracted the personal interest of Charles II. When
Oldenburg reported observations of a transit of Mercury to Boyle in October
1664, he said that "Our Virtuosi did observe both at Gresham [College] and
Mr Reeve's".8 We know that on the occasion of an earlier transit, in 1661,
Christiaan Huygens was among the observers at Reeve's shop.9 There is evi-
dence that Thompson's shop in Hosier Lane, Smithfield, was also a recog-
nized meeting place, and Monconys knew that he had made the original
example of Wren's perspectograph. lO Greatorex too was one of the makers
closest to the Royal Society and had worked for Wilkins, Goddard, and Boyle.
We have seen that Monconys does not only transact business with Reeve:
he records things he learned from his discussions with him, and it is partic-
ularly striking that Reeve has a demonstration to offer his visitors. It is
important in this context that Reeve has a shop-a business address that is
more than simply a workshop. His address is known; Monconys can go
there and be received before he has met his Royal Society contacts. This is a
commercial space that is in the public domain, but it is a site also for the
exchange of intelligence in natural philosophy and even for observation and
demonstration of the kind normally associated with the Royal Society.
In these London shops there were off-the-shelf items that could be had
immediately, but for more unusual instruments from the best makers, the
customer would wait while his order was attended to. This was not always
successful: Monconys was promised by a glassworker that he could have an
example of Wren's circular thermometer-an instrument that could be
made self-registering- but on returning several days later, he was disap-
pointed. This is not surprising: it was a very unusual instrument, requiring
critical shaping and balancing. Even this phenomenon of delayed orders and
returning visits is linked to the different uses of the shop space. The visitor
might come back not only to see or receive his completed instruments, but
also or alternatively to see finished work for other customers or for stock,

Shopping for Instruments in Paris and London 373


and a more advantageous time for a visit could be predicted on the basis of
progress in the workshop. This reinforced the idea of the shop as a place of
resort, where instruments and production techniques could be viewed and
discussed, while the visitor's growing familiarity with the maker and his
premises added to the impression of a semipublic space where he and other
visitors had an interest shared with the maker himself. Monconys saw Reeve
several times, and records being given specimens of materials, but so far as
we know he bought nothing.
These characteristics of the London trade had developed significantly by
the time the collector, connoisseur, and traveler Zacharias Conrad von
Uffenbach of Frankfurt-am-Main came to London with his brother in
I7IO.II He visited the shops of John Marshall, Edmund Culpeper, John
Patrick, John Rowley, and Francis Hauksbee, all of whom are remembered
today as talented and original makers. In each case he seems to have gone by
coach deliberately to the premises of a known maker at a known address.
On one occasion they "intended to wait on" Rowley, but he was not at
home. 12 On another they failed to find "the shop of the famous mechanician
Moxon."13 The Moxon who could most reasonably be described as "famous"
was Joseph, who had died in 1691, but could well have had a continuing
European reputation through his publications. His son James succeeded to
the business, but the last record of him trading dates from 1703.14
Von Uffenbach's brother persuaded Marshall, on payment of 7 guineas, to
teach him the art of glass cutting, not because he thought Marshall was par-
ticularly skilled in this direction-he did not-but because he was curious
to learn the English practice. As a result the brothers made frequent visits to
Marshall's shop, usually in the afternoons, for these practical lessons. They
bought some small telescopes, and were taken by Marshall's method of
demonstrating their quality: he "had had his name placed in white on the
roof of a house some twenty houses from his own," which could be read eas-
ily with one of his telescopesY From Culpeper, von Uffenbach bought a
number of fairly routine items-an architectonic sector, a magnet, a micro-
scope, a spyglass, a drawing pen, and an oval reading glass - but Culpeper
could not help the brothers in their quest for an example of Samuel Mor-
land's calculating machine. At Patrick's they saw many different kinds of
barometer, which von Uffenbach describes in detail, while mentioning
which are new inventions and what were their particular advantages, so
they must have spent some time viewing and discussing them, but they seem
to have bought nothing. 16
They did buy a telescope at "an optician's called James Praun," and they
asked him to repolish the lenses of instruments bought from Culpeper. It is
curious that no James Praun or anyone with a name that might be so con-
strued is known to have been trading at the time. In von Uffenbach's view

374 James A. Bennett


"This man is very cheap and good and, if he takes the trouble, makes a fine
polish,"l7 but we must remember that he had judged Marshall's talents with
glass inferior, even though he knew that "everyone makes a great to-do over
him and his work."l8 Von Uffenbach's judgment reflects on his engagement
with instruments, and in any case the whole account of his visit to London
reminds us that he is a general connoisseur, collector, and dilettante. The
telescope he buys from Praun is in keeping with this. It is a refractor with
five lenses, a shagreen covered body and a number of draw-tubes. The lenses
and tubes can be configured in four different ways, to yield different tele-
scopes, which von Uffenbach describes. This delight in ingenuity, often tak-
ing the form of a multipurpose instrument that can be variously configured,
was a familiar ploy to catch the dilettante shopper.
When von Uffenbach did find Rowley in, he was willing to make Mor-
land's calculating machine for 5 guineas, but none seems to have been
ordered. Rowley, who was a very resourceful and original maker, showed
the brothers some long object glasses between 15 and 18 feet focal length,
that he was not willing to sell. One interesting thing we learn from von
Uffenbach, which does not seem to be known otherwise in the history of the
telescope, is that Rowley was making a Newtonian reflector. It is generally
thought that the only such attempt between Newton himself and John
Hadley in 1721 was by Francis Hauksbee, who was closely associated with
Newton. Although the details are slight, it seems likely that the objective
mirrors were made of glass, to be silvered: von Uffenbach introduces them
in the context of the long focal length objective lenses, says they are "convex-
concave" and compares their polish with that of the lenses. Silvered glass
mirrors for telescopes came into general use only much later but the idea
was around in the early eighteenth century: John Pound mentions it as a
possibility in 1723, and James Short was making such mirrors in 1734. Both
Newton and Hadley used metal mirrors. Rowley valued his objective mir-
rors highly - at 7 or 8 guineas each. l9
So the pattern of transactions in the shops of the makers is becoming
familiar. Viewing of original pieces, discussion with the maker, instruction,
and demonstration are all part of practicing the trade on the one hand and
polite shopping on the other. Makers with substantial reputations were par-
ticular targets, even if such shoppers were prepared to buy elsewhere. Von
Uffenbach was especially fortunate to have seen a Newtonian telescope
under construction at Rowley's, and to have listened while "he praised this
invention of Newton very highly, as though it were quite matchless." But he
still recorded his disappointment that "there were no other instruments
completed at the moment, which is greatly to be regretted, since he is con-
sidered one of the best mechanicians in England."20 Rowley's name is found
on a great variety of surviving instruments, and it is difficult to imagine that

Shopping for Instruments in Paris and London 375


he had nothing in his shop beyond a few lenses that were not for sale. It
seems more likely that von Uffenbach is ignoring the standard and routine
instruments: this aspect of the shopping experience related to the original,
ingenious, or ornate piece, an instrument worth a special visit, one for which
the maker himself had a direct responsibility and that was not simply made
or bought in as an item of stock.
Von Uffenbach witnessed the most developed example that London had
to offer in this rise of polite shopping when he visited the premises of Fran-
cis Hauksbee. Hauksbee was the maker closest to the Royal Society, being
employed as a demonstrator to provide experiments at the meetings. He
took these experimental demonstrations from the society into the public
realm with his subscription series of lecture/ demonstrations, at first in col-
laboration with James Hodgson.21 By 1710 Hauksbee was offering the
courses on his own, and von Uffenbach's account suggests that experiments
would be performed for private visitors as well as within the lecture course;
we cannot be certain of that, but there is no suggestion that the audience was
larger than the two brothers.
The first time they visited Hauksbee's premises in Wine-Office Court,
Fleet Street, it was not by appointment, since "we did not find him in."
Clearly they treated Harksbee's address like that of a shop, where they could
expect to be received on demand. They were not entirely disappointed, as
Hauksbee's young "cousin"-probably his nephew, also Francis Hauksbee,
who was later involved in a rival lecturing enterprise-"took us up and
showed us some common experiments." They then made an appointment
and bought a copy of the elder Hauksbee's Physico-Mechanical Experiments
(London, 1709), "so that after studying it at home, we might see the experi-
ments again with great profit."22
When they returned, they were in the hands of a master philosophical
showman: "We saw with amazement his excellent demonstrations and
experiments, especially those relating to the nature of light, which were cer-
tainly very excellent and curious." Von Uffenbach followed the experiments
from Hauksbee's book and was enchanted by his skill in carrying off a
whole series of pneumatic demonstrations in a virtuoso display on the air-
pump. Hauksbee was practiced at drawing his audience into the experi-
ments, getting them to test the effect on the phenomena of rubbing the
evacuated vessel with their fingers, or breathing on it.2l Here is a telling
early example of the commercial talent and technique of a new breed of
philosophical entrepreneur: the combined use of the illustrated textbook and
the demonstration to engage the potential customer for examples of the
apparatus in use. Hauksbee derived income from courses and publications,
but his principal aim was to sell the products he manufactured. To do so he
had first to create a new market, and the policy he adopted was to transfer

James A. Bennett
to his shop-with only slight modification-the traditional experimental
performance demanded formerly of Robert Hooke and now of himself at
the Royal Society.
An unfortunate imbalance is already evident: we are much better sup-
plied with foreign instrument shoppers in London than in Paris. For the end
of the seventeenth century, we do have Martin Lister's visit to Paris: he is
interested in various trades, but he takes little notice of instrument makers.
In fact the only one who figures significantly is Michael Butterfield, who
was "a right hearty honest Englishman, who has resided in France 35
years."24 Butterfield is a further example of a maker conducting experimen-
tal demonstrations on his premises - in this case with his large collection of
lodestones - but it is clear that this is not typical of the Parisian scene.
The imbalance simply reflects the state of affairs in the period: informed
observers did not go to Paris to shop for instruments. Long ago Maurice Dau-
mas noted that the Parisian makers served "a small, local clientele," which he
identified as the royal court, the salons of a few wealthy individuals, and the
cabinets of several colleges. 25 There was practically no export trade, and the
local market would not witness the consumer expansion that in England
would give the hardware of natural philosophy a place in many well-to-do
homes. It is only toward the end of the eighteenth century that a few visitors
to Paris indicate the beginnings of a revival in the French industry.
But French visitors to London can be an indirect comment on Paris,
because of the unfamiliar things they note and the contrasts they draw.
Lalande came to London in 1763, the visit being in itself a comment on the
relation of English to French work in precision mechanics, for his mission
was in part to discover as much as he could about the chronometers of John
Harrison, and in part to investigate the acquisition of astronomical instru-
ments.26 He was certainly struck by the vitality and confidence of the
mechanical scene in London, by the distinctive and visible presence of a
dynamic group of practitioners, and by the acceptance of some of them into
the Royal Society. For some, such as the leading makers of astronomical
instruments, involvement with the Royal Society was important, but it must
be remembered that the makers had other institutions and associations and
were more than able to take independent initiatives. Historians of science
tend to focus on the Royal Society as the source of intellectual and cultural
patronage, but it is far from clear that it had an exclusive prominence in
these respects for the makers. Nonetheless, Lalande found the easy relations
between savants and fabricants, between sciences abstraites and sciences
appliquees, different, refreshing, and positive.
Lalande spend a good deal of time with James Short, and also with John
Bird and Francis Watkins. He discussed polishing techniques several times
with Short, who was his principal link to Harrison. On one occasion Short

Shopping for Instruments in Paris and London 377


confided that giving a mirror a parabolic shape-and his reputation rested
particularly on this supposed skill- was a matter of touch, not science: "it's
only by feeling one's way."27 Lalande was entertained several times by
Watkins and described instruments he saw there. He also visited Jeremiah
Sisson more than once and viewed his instruments. It may have been from
Watkins that he learned that Sisson had gone to prison several times for not
paying his workmen, and that failure to complete instruments had led him
to pawn others, only to see them sold well below their market value.28
Bird, he discovered, would charge I,200 French livres (equivalent to 50
guineas) for a quadrant of I8 inches, half to be paid in advance, and 350
pounds sterling (8,000 livres) for an 8-foot quadrant-that is, one like the
mural instrument he had built for Greenwich. Packing would be an addi-
tional IO pounds, and the delivery time would be two years. 29 Lalande had
learned at Sisson's that it was Jonathan, rather than his infirm father, Jere-
miah, who had made several well-known quadrants, including that of his
mentor Pierre Charles Le Monnier; at Bird's, on the other hand, Bird told him
that he had seen complaints about Sisson's work. 30 No doubt, after the order
from Le Monnier and in view of Le Monnier's increasing promotion of Eng-
lish instruments, Lalande was seen as a serious possible source of a significant
commission, so the rivalry between Bird and Sisson was understandable. A
week later Bird was able to play an important card: Lalande had breakfast at
Bird's with Thomas Hornsby, professor of astronomy in Oxford, and they
went together by boat to Greenwich to examine the instruments there; both
Bird's guests were shopping for quadrants, and both would eventually buy
small and large quadrants from Bird. Lalande placed his first order two weeks
later, for an I8-inch quadrant, paying the stipulated 25 guineas in advance. 3l
Lalande introduces us to some new features of the London scene. No
ordinary shopper, he was favorably placed as a fellow of the Royal Society,
and he carried the promise of a significant commission. His shopping tech-
nique included spending a long period in the astronomical community of
London - makers and astronomers. At the same time, the situation of the
makers was changing. The success of Graham's instruments at Greenwich,
including the first empirical proof of the motion of the earth and the setting
of a new foundation of accuracy in positional astronomy through the discov-
eries of aberration and nutation, had set the elite makers of precision instru-
ments on a path that would progressively raise their status above that of the
rest of the trade. Sisson and Bird had begun to reap the benefits through the
beginnings of an export trade in astronomical instruments. 32 Bird would
became even better known through the Board of Longitude's endorsement
and publication of his construction methodsY But he was already able to
take distinguished shoppers to the Greenwich Observatory-the perfect
showroom for his quadrant and transit instrument.

James A. Bennett
Lalande's elevated position also reveals to us an aspect of the trade in
instruments of natural philosophy that we have not encountered before. He
went with Dr. John Pringle, physician to the queen and later to George III,
to see the royal collection. According to Pringle, George had told him that
he knew Lalande's works. The king was going to Richmond but had left
orders that the air pump designed by Smeaton was to be made ready for
Lalande, and it was George Adams, the maker of the pump and instrument
maker to the king, who had the considerable task, according to Lalande's
account, of getting it to work. Princely patronage for instrument makers
was not typical of the London trade, but it did exist. 34

BERNOULLI, BUGGE, AND VAN MARUM IN LONDON AND PARIS

Three of our travelers from the later eighteenth century go both to London
and to Paris. The first is Bernoulli, whose surprise and delight at the
vibrancy of the commercial trade in London in 1768 we have already seen.
Some of the practices we have encountered with Reeve and Haukbee are
now being pursued vigorously by Benjamin Martin. Bernoulli and his corre-
spondent already know Martin's name through his publications. His shop is
one of the best stocked, and he has a clientele eager for his courses on
mechanics, experimental physics, and astronomy, all made doubly interest-
ing by the beautiful instruments used for illustration. Bernoulli himself
attended "with pleasure" a lecture on the forthcoming transit of Venus,
where Martin used a large animated diagram on one wall of the room,
which represented the progress of the transit, as it might be visible in Lon-
don, up to the time of sunset. This was a "virtual" experience offered in a
room of Martin's shop, available daily with no disappointment from bad
weather. Bernoulli is very taken by this, and in general gives the impression
of a popular, thriving, and innovative business in natural philosophy.15
Although he judged that Martin's twenty or so books had been very well
received, not everyone had been impressed. It is perhaps an indication of the
circles Lalande had moved in that he had been told in 1763 that Martin
"gives courses in physics ridiculously."36
Opposite Martin, and also well known through his publications, was
Adams, who had a great number of instruments of physics, mathematics,
gnomonics, and astronomy. Henry Pyefinch also had a very well-stocked
shop, and Edward Nairne is mentioned as a well-known maker of tele-
scopes and other instruments. Short has died, and Bernoulli attends the sale
of his instruments. 37
Of the mathematical instrument makers, three rivals stand above the oth-
ers, and among them Bernoulli believes their reputations to stand in the

Shopping for Instruments in Paris and London 379


order: Bird, Sisson, and Jesse Ramsden. He has clearly spent time talking to
Bird and Sisson, but not Ramsden. He does describe in detail a telescope
mount by Ramsden, but he may have seen this in Dollond's workshop.
Among other things, he has discussed with Bird the application of achro-
matic lenses to divided instruments and with Sisson his method of dividing
scales-a point of interest because Bird's method had now been published
by the Board of Longitude, and Sisson had been trained by his own father,
who had also trained Bird. For the well-informed shopper the history and
workshop traditions of the London makers already fall within the orbit of
connoisseurship. Bernoulli is aware of Bird's latest foreign commissions-
clearly considered a mark of his distinction - and while his instruments are
judged the finest, they are also the most expensive. 38
One of the most useful aspects of Bernoulli's record is the negative report he
writes on Dollond, because it illustrates part of the increasing complexity of
commercial practice. Bernoulli has been to Peter Dollond's workshop: it is
large and its products cover a wide range-reflecting as well as refracting tele-
scopes, and almost all the types of optical and astronomical instruments used in
England. The more valuable instruments are available only to order. It is help-
ful that, while the word "shop" is ambiguous in English in this period, in
French Bernoulli has three words-"atelier," "boutique," and "magasin"-
that he can deploy more precisely, and at this point we know he is talking about
Dollond's workshop, "atelier." (Martin, Adams, and Pyefinch each have a
"magasin"-a shop of a larger size than a "boutique.") For the instruments
made in quantity, he cautions that people who buy English work imagining
that the signature "Dollond" is enough to ensure excellence are seriously mis-
taken. Indeed, if they are fortunate enough to get a good instrument, this
means that it was not made by Dollond's workmen at all, but that at least part
of the work has been subcontracted to his brother-in-law Ramsden. 39
Whatever the rights of Bernoulli's opinion of Dollond's workshop, we
have learned that by I769, at least in the larger enterprises, a signature does
not identify the work of the master; it does not even necessarily identify the
workshop where the piece was made. It has to be said that Bernoulli was not
impressed by Peter Dollond, because he did not have the theoretical knowl-
edge of his father, John. It was surprising, wrote Bernoulli, that he had man-
aged to achieve so much in achromatic lenses on the basis of trial and error.
He did not, of course, hear this from Dollond, but had been assured on good
authority that a great many lenses of crown and flint grass were manufac-
tured and successful combinations found simply on the basis of trying them
out. Bernoulli did, however, have his own experience to add to this. Extra-
ordinarily, he posed a series of questions to Dollond, questions that had been
set beforehand by Beguelin. It is not clear whether Dollond realized that he
was sitting an examination, but he definitely failed .40

James A. Bennett
Despite all this, the only purchases Bernoulli records making were of
three lenses from Peter Dollond. Dollond held the patent for achromatic
lenses, which is part of the explanation for widespread resentment among
other opticians, as well as for the questioning and probing from foreign vis-
itors, who believed that the theoretical advantage lay with Continental
mathematicians. Sometimes shoppers were obliged to buy from a certain
maker for reasons other than choice. Bernoulli could not be completely sure
of the significance of Dollond's examination failure: either he could not or he
would not answer, but either way he did not feel obliged to cooperate with
Bernoulli's stratagem. Perhaps the strength granted to him by the patent
meant that he did not need to cultivate shoppers in the fulsome manner
Bernoulli had enjoyed elsewhere.
Bernoulli left England for France in May I769, and he sees the mural
quadrants by Bird and Sisson in the observatory of Le Monnier. He says he
has little new to report concerning the makers of astronomical instruments.
They are not as good as the English makers, who particularly excel in
divided instruments, but he can at least mention Langlois and Canivet for
producing distinguished work. There are amateur lens makers producing
good achromatic objectives for telescopes, but no "Artists by profession."
The only shop he mentions is that of Passemant, where good reflectors and
refractors can still be found, though his report seems not to be firsthand.
Throughout, no maker is reported as having said anything to Bernoulli: at
least so far as his letters are concerned, the Parisian makers are silent. 4!
The visits to London and Paris made by the Danish astronomer Thomas
Bugge were separated by some twenty years, so we will look first at his expe-
riences in London in September and November I777 and separately at his
time in Paris. In London he visited the shops of many makers, buying books
and instruments, and enjoying conversation and experiments. 42 Adison
Smith "at the golden quadrant" in the Strand, who in I764 had been one of
the petitioners against the patent for achromatic lenses held by Peter Dol-
lond, told him the makers' account of the priority of Chester Moor Hall,
while taking his order for a triple prism to demonstrate Dollond's tech-
nique. 43 On his visit in I763, Lalande also had heard of the claims on Hall's
behalf from John Bevis and Short. Bugge may have caught the flavor of this
dispute, for he later noted, having visited the premises of both Peter Dollond
and his brother John, that "none of the Dollond brothers seems to have any
theoretical knowledge."44 It was precisely Peter's contention that any earlier
lens combinations that may have existed were made empirically and not on
the basis of any theoretical understanding.
Some instruments Bugge could buy directly from stock-a rule from
Smith, for example, comparing English, French, Dutch, and Antwerp mea-
sure, or Dollond's form of rolling parallel rule. 45 Martin's shop was "well-

Shopping for Instruments in Paris and London


supplied," and he bought a number of books there. 46 He was particularly
taken with Nairne & Blunt, finding Edward Nairne a "splendid man," who
could show him "many excellent objects"-a new kind of marine barome-
ter, a dip circle, different sorts of theodolites, and a new telescope mount. 47
He returned so that Nairne could show him a number of experiments with
his electrical machine, and he ordered one. 48
Bugge was very taken with the watchmaker Alexander Cumming, buying
his book on the Elements of Clock and Watch Work, admiring his barograph,
and hearing the details of the construction of different types of compensation
pendulum. Cumming showed him the observatory he had at the top of his
house, with a transit instrument and an equatorial, both by Ramsden. 49
Bugge later returned and "examined his barometric clock very closely,"
learning along with the design details that the price for such a clock was 500
pounds. 50 On his visit to John Arnold he was told of his improvements to
chronometers, was shown a demonstration of the effect of temperature on a
balance spring, and made an appointment to view experiments on different
types of compensation pendulum, which took place the following week.
Bugge discovered that Arnold had a transit instrument that had formerly
belonged to George Graham. 51 Sisson described to him an observatory
designed for Nathaniel Pigott, and showed him a transit instrument to be
installed there. 52 Here as elsewhere, details of the design are carefully
explained and recorded, especially where there are new ideas. These were
particularly evident at Ramsden's shop in Piccadilly, where he was shown an
8-foot quadrant being constructed and noted "Ramsden's ideas about some
new instruments."53 George Adams was able to show him a transit instru-
ment he was making for the king's observatory at Richmond. 54
Bugge spend some 88 pounds on instruments and associated apparatus, as
well as 34 pounds on books. The most expensive items were a compound
microscope from Dollond at 8 guineas, lenses from Smith at 8 pounds, 9
shillings, and a range of things from Nairne and Blunt: a hydrostatic balance
(7 pounds, 17 shillings, 6 pence), an electrical machine (13 guineas), a battery
of nine Leiden jars (4 guineas), a mechanical model (mechanical powers) (21
pounds), and an air pump (I I pounds, IS shillings). Nairne was able to show
Bugge the instruments made for him on his last day in London, and he
"found them all very pleasing."" He also bought a set of drawing instru-
ments from Adams at 2 guineas. 56 It is worth emphasizing that he returned
twice to see more extensive experiments - in Arnold's case certainly by
appointment, and probably in Narine's as well, to judge by the extent of the
experimentation and the careful notes taken by Bugge.
In July 1785 Martinus van Marum of Teyler's Museum in Haarlem was in
Paris, visiting physical cabinets, attending sessions of the Academie des sci-
ences, witnessing the therapeutic applications of electricity, and so on-

James A. Bennett
thoroughly occupied with the French practice of natural philosophy.57 He is
particularly interested in electricity and is curious to see the use of a machine
by Nairne. It is clear that there is plenty for him to do and see and that, while
there is a growing vitality that is less commercially based than in London,
there are also occasional similarities to London. Courses oflecture-demonstra-
tions continue in the tradition of Nollet, generally not conducted by makers
themselves, but using a wide range of instruments. Roulant was the successor
of the better-known Sigaud de la Fond and was in the middle of moving
when Van Marum visited him, so not much of the collection was visible, but
he did say that he had held thirteen courses in physics the previous year.
However Van Marum does visit one shop-that of Bianchi (or "Bianchy")
in the fashionable shopping street the rue Saint-Honore. As an Italian who
had previously plied his trade in Amsterdam, Bianchi was something of an
interloper in the Parisian scene, but he brought with him practices of the
Anglo-Dutch variety that linked experimental demonstration with making
and trading. Bianchi probably began as a glass worker and barometer maker
in the Italian tradition, and it was the use of blown glass in electrical machines
that occupied the conversations and demonstrations that engaged Van Marum
and himself, continued by arrangement into a second day. We know sepa-
rately that Bianchi advertised himself as a demonstrator in physics, conducted
courses on experimentation in electricity, and supplied a wide range of exper-
imental apparatus, including electrical machines, air pumps, barometers, ther-
mometers, microscopes, and telescopes. 58 This was a real shop in the London
sense-a space where the maker cultivated the customers' interests, engaged
them with impressive experiments, offered them courses to deepen their com-
mitment, and sold them philosophical instruments.
This was a single instance, the initiative of an enterprising immigrant.
Otherwise, Van Marum is very largely engaged with visiting the many pri-
vate cabinets; these give him his principal visual impression of natural phi-
losophy in Paris. When he comes to London in 1790, he seems to be less
motivated to make a full record of his activities, but in only a few pages of
notes covering three weeks, which trail off into mere jottings, instrument
makers and their shops are a prominent and recurring feature. 59 Not all are
named, but the better known among those who are are Nairne, Adams, and
Haas. Van Marum visits Tiberius Cavallo, who uses his extensive cabinet of
instruments for a series of demonstrations. Although not generally seen as
a maker, Cavallo incorporated a commercial dimension into his perfor-
mance, as he takes an order for four pieces from Van Marum and will have
them made. Also an Italian, his situation is not very unlike that of Bianchi
in Paris.
Otherwise, the conduct of Van Marum's shopping in London is familiar
to us. He returns by appointment to see more with Adams, Nairne, and Dol-

Shopping for Instruments in Paris and London


lond. Adams was rewarded by an order for a complete table planetarium
with lunarium and tellurium. Nairne "requested us to come again the next
Wednesday, when he expected to have some new instruments on hand,"
while at Dollond's "we received from him the promise that he would show
us another day what he had ready." Van Marum also went twice to Haas,
but by this stage his notes are very sketchy. These visits fall within the pat-
tern we know already of the London trade in relation to specialist shoppers.
Perhaps the only new fact of commercial interest we learn from Van Marum
is that Nairne is sufficiently prosperous to have a country house. 6o

CASSINI IN LONDON; COMMERCE AND NATURE

J.D. Cassini offered an analysis of the French situation in relation to the


English-in effect Paris in relation to London-as he saw it when he took
over the directorship of the Observatoire de Paris in 1784. He was deeply dis-
satisfied with its prospects, particularly on account of his instrumentation.
The famous expeditions of the past had created an enthusiasm for instru-
mental improvements and had encouraged the talents of French makers,
but they had now been left far behind by the English, whose technical and
commercial dominance could not be ignored.

The Observatory was then equipped with mural instruments, with large
movable quadrants, the work of Langlois, of Canivet, of Lennel, who
were in those days the most celebrated makers of astronomical instru-
ments. But at the time I took over the direction of the Observatory, these
old talents had been eclipsed by those of Bird and Ramsden, English
artists who had carried their art to the highest perfection, leaving the
French far behind them, from whom they had usurped the trade in opti-
cal and mathematical instruments almost entirely.61

Cassini's analysis is instructive, as is the proposal he offered to remedy the


situation. French makers, he was sure, were not inferior in talent or educa-
tion; the problem lay in commerce, not in mechanics. The French makers
could not get a fair price for their products: French customers would not pay
a fair price for French instruments. Yet at the same time he expected them
to pay as patriots, rather than as shoppers. Caught up in an "anti patriotic
mania," they behaved as though they were buying lengths of cloth:

How often have I been offended by people who, begging me to procure a


good instrument for them, haggle as if it were a length of cloth. What will
it cost? I was asked; not as expensive, surely, as an English instrument?

James A . Bennett
Why not? I replied; do you want to be fair? if the instrument is worth
nothing, don't take it: but if it is good, pay what the maker will ask, even
if he were to make you pay more than for an English one. 62

In fact it was impossible, he said, for a maker in Paris to produce the same
instrument for the same price as one in London, and for three reasons: they
did not have the means and the appropriate machinery, they did not have
the capital to fund the first stages of construction, and they were the victims
of the "Anglomania" that created a preference for English instruments.
Cassini's solution, however, was more French than English. He did not
argue for, say, the abolition of the restrictive practices of the corporations,
whose activities were particularly detrimental to a manufacturing trade
which, if it was to progress, had to be able to bring together different skills and
combine a variety of materials. He did not try to imagine how he might
encourage a broader-based consumerism with respect to mathematics and
natural philosophy, as his jibe about shopping for cloth makes clear. Rather he
proposed the establishment of a kind of national workshop at the Observatoire.
Here the machines that individual makers could not afford to buy would be
made available to them. Two or three promising mechanics would be sent for
training in the great workshops of London, to be employed in the national
workshop on their return, and to maintain the momentum of improvement
the workshop should be given an annual subvention from the government.
Cassini's analysis of the differences between makers in Paris and London
was largely accurate, but in London capital for the machines needed for effi-
cient production and capital for large, prestigious, and innovative instru-
ments were generated by commerce in smaller, everyday instruments, and
by the entrepreneurial development of forms of shopping that made buying
instruments not so very different from buying cloth. Of course, London
makers had this commercial opportunity because they were in the middle of
a broader consumer revolution among the English bourgeoisie, and they
could hitch themselves to this engine of capital production. But in doing so,
they had taken risks that Cassini did not feel could be expected of the
Parisian makers: John Cuff and Benjamin Martin were among a number
declared bankrupt. Martin even committed suicide.
The workshop and an associated foundry were established at the Obser-
vatoire but were not successful. Yet, just as the project seemed to be faltering,
Cassini had the opportunity to go to London on a plausible and legitimate
mission that would take him into the heart of the instrument world. The
meridians of Paris and Greenwich were to be linked by a collaborative geo-
detic enterprise, with Cassini in command of the French component. He
would have to go to London to make the necessary arrangements, as he
explained to the minister: "I will go to see these superb English instruments,

Shopping/or Instruments in Paris and London


to examine these masterpieces of the art, and these models which it is our first
ambition to imitate in France. I will have occasion to see and to cultivate Dol-
lond, Troughton, Ramsden."63 He was particulaly anxious to cultivate Rams-
den, for by now his reputation was unrivaled, and it was in his workshop that
he hoped to place the two or three chosen Frenchmen. How could Ramsden
be persuaded of such a plan? Cassini sought permission to order an instru-
ment from him, having no doubt that he would be suitably mollified by
receiving such a commission: "without doubt he will be flattered to work for
the Observatoire Royal de Paris, where there is still nothing by him."64 This
would be the device for softening Ramsden's attitude to training Frenchmen.
It was, admittedly, quite a step to buy an English instrument for the Paris
observatory, but the long-term outcome made it worth the loss of face, and
they would acquire a desperately needed modern instrument as well as a
model for future imitation. Cassini left in September 1787 with permission to
order a 7-foot transit instrument and to negotiate the placing of French
workers in London workshops. He would be a shopper with a very particu-
lar agenda, not least because he did not really approve of shopping.
Cassini was not disappointed in Ramsden; indeed he was enormously
impressed: "I recognised that despite all our efforts we would never have so
consummate an artist in France; all rivalry, all comparison seemed now to
me impossible faced with such a great talent."65 Ramsden was full of ideas
and conversation. The kinds of shop discussions we have noted, engaging
other makers and other visitors, had become in Ramsden's case an inspira-
tional audience for Cassini.

Leaving one day from one of these conversations in which I loved to take
part and be instructed by him, I said to a stranger, who was just as enthu-
siastic as I about the merit of M. Ramsden: In truth, this man is an electri-
cal machine which you only have to touch to draw a spark . Your
comparison could not be more appropriate, was the stranger's animated
reply, for you could very well get nothing more than sparks here. 66

Elsewhere the electrical machine was a star attraction in the polite enter-
tainment offered in the shops of instrument makers; here it was a metaphor
for the extraordinary maker himself. Whether at this stage Cassini appreci-
ated the meaning of the foreigner's response is not clear, but sparks were all
he could hope to draw from this machine: he would never acquire an instru-
ment from Ramsden.
Ramsden was utterly charming and plausible. Cassini spent close to two
years in London, and he visited the shop "ceaselessly." Ramsden spent a
great deal of time with him, showing him instruments, promising him
everything, and delivering nothing. Cassini attributed this to Ramsden's

James A. Bennett
perfectionism and his fascination for new and improved designs: "Don't
think that it is indifference or laziness on his part: quite the contrary. But a
new idea, a difficulty to overcome or an instrument of some new type that
someone proposes to him will attract all his attention and make him aban-
don any work he has started."67 In this context a transit instrument, one stip-
ulated to be modeled on the one already made for Palermo, would not excite
much interest for Ramsden. Whether or not he was flattered to receive an
order from the Observatoire, it did not represent the most exciting work he
had in hand. In fact the discussions resulted in two instruments being
ordered-the transit instrument and a rotatable wall with an 8-foot quad-
rant on one side and a complete circle on the other. The virtues of the circle
over the quadrant were beginning to exercise Ramsden's interest and inform
his later designs, something Cassini must have picked up from their many
conversations. The instrument would permit a comparison between the cir-
cle and the quadrant, he wrote to Ramsden on his return, confirming the
order: "I believe this deserves to excite your genius."68
Despite all Cassini's efforts to obtain his instruments, and despite Rams-
den's continuing promises, only the transit instrument was completed, and it
was made by Matthew Berge, Ramsden's foreman and successor in business.
I t was delivered in 1804, eleven years after Cassini's return and four years
after Ramsden's death. Meanwhile Cassini's greater projects had been over-
taken by the disruption of the Revolution. In truth Ramsden did not need
Cassini's commission. He had told him in replying to his written order that
he had forty or fifty men employed in his workshop. The regular business
provided income for the extraordinary, but this meant that Ramsden was
free to choose the extraordinary on his own terms. Cassini could have
noticed this danger when he read Ramsden's reassurance that he did not
need to worry about financing the work: "The regular business of my work-
shop provides me with the necessary income. My main objective is the per-
fection of instruments. This part of our profession is still in its infancy."69
Cassini offers more analysis than other visitors, but the general pattern of
their experiences has become familiar and to some extent repetitive. It is
time to stop adding examples and attempt some analysis of our own. It is
evident to our shoppers that makers in London are much more engaged in
entrepreneurial commerce than those in Paris; this leads both to a greater
visibility, and to a more flourishing manufacturing enterprise largely invisi-
ble to visitors. In the way things are managed in London, the growth of
demand for routine items enables innovation and design development for
more individual and special pieces. The more successful makers are less
dependent on the dictates of clients for these more ambitious designs, and in
extreme cases their success and reputation were such that they could afford
to ignore commissions that were not to their taste-not because they were

Shopping for Instruments in Paris and London


difficult or costly to undertake, but because their design ambitions lay else-
where. The astronomer Francis Wollaston failed to interest either Ramsden
or Troughton in his proposal for an instrument and was obliged to take his
custom to the less able and less prestigious maker William Cary, complain-
ing that, "observers know best what they want; and an instrument-maker
who will condescend to listen to them is a treasure."70
We have seen that the existence of a shop space was a valuable resource in
the makers' situation. They were, after all, presented with some problems.
On the one hand, they were trading in mathematics and natural philoso-
phy-their material products were supposed ultimately to deliver more than
materiality to their customers - so they needed to present a "polite" image to
the world. But at the same time they ran mechanical workshops, and the skill
and reliability of those who worked there was of legitimate concern to cus-
tomers. A shop-an intermediate space between the street and the work-
shop-could be used in imaginative ways to negotiate the distance between
the polite and the mechanical. The shop was a place to which visitors had
unannounced access, and to that extent it was part of the public realm, but the
maker controlled this space and what took place there. He could make him-
self and his special instruments available there in different ways, according to
the importance and the purchasing potential of the customer.
As we have seen from our shoppers in London, the ways in which this
access was managed included repeat visits, discussions, demonstrations, lec-
tures, experiments, and even mechanical instruction, as well as traditional
shop displays. The shop could to some extent be held ready for casual
demonstration and explanation, or could be made ready for a special session
arranged by appointment. Its relationship with the workshop meant that
such sessions could be scheduled in sequence with workshop output. Anita
McConnell's study of the London workshops cites examples of recorded
premises that included a shop, a workshop, living quarters, and the possibil-
ity of other workshops nearby. These resources were augmented by links to
more widespread workshops under different direction through the network
of subcontracting. 71 The control the maker or retailer exercised over the
shop space meant that his response could be tailored to the shopper, though
in the case of the most prestigious and independent makers, this could spell
frustration even for distinguished callers. Hornsby records several frustrat-
ing visits to Ramsden's premises to complain about delays, when he was kept
waiting in the shop without seeing the great man, even though he was con-
vinced that Ramsden was in a back room.72
This chapter has been mainly concerned with commerce; what can it say
about the representation of nature? First, it is clear that there is a link
between activities in the shop and those taking place at other locations in the
geography of contemporary natural philosophy, including the Royal Society.

James A. Bennett
We are inclined to think of the commercial manifestation as an opportunist
outgrowth from the academy-Hauksbee finding a business opportunity in
the model of illustrative demonstration developed by Hooke. But while that
may still be the dominant direction of influence, we need to be cautious in
simply assuming that this is the whole story, not least because the chronology
is not clear-cut, as we have seen through the cases of Reeve, and even earlier
of Allen. Monconys could listen to the Royal Society debates about the "glass
teardrops" and, had he lived, could have read about them in Hooke's Micro-
graphia, but he could also buy them by the score in the shops of instrument
makers. Whatever may be concluded about directionality from a more
detailed study, it is clear that similar kinds of things were taking place across
a range of locations, some under a stronger commercial agenda, others pri-
oritizing natural philosophy, but for contemporary observers some morpho-
logical similarity would have been obvious. At the very least, it was the shops
that disseminated a form of natural philosophical practice to a broad public;
is was therefore appropriate that the popular textbooks of the eighteenth
century arose largely from these commercial enterprises.
One important difference between the situations of makers in London
and Paris was the guild or company regulation that applied in the two cities.
In Paris, companies regulated particular trades, seeking to control the size of
workshops, the provision oflabor, the methods of production, and the choice
of material. 73 Further, these regulations became stricter and more stringently
enforced in the eighteenth century. Various companies sought to control
instrument making, as a relatively new trade, but it was not clear which was
most appropriate, and disputes arose that disrupted production, with mak-
ers even having tools and materials seized. As instruments were developing
in the period, makers in fact needed to use a variety of skills and materials,
and they did not fit into a particular category. One way of avoiding these dif-
ficulties was to achieve royal protection and work in the Galeries du
Louvre-a further step away from the kind of brash commercialism evident
in London.
The guild situation in London was very different. Here a maker needed
to belong to some company or other in order to trade in the city, but it did
not matter which he joined and he was not bound by any regulations gov-
erning his product. Once a successful maker belonged to a company, his
apprentices would be made free of that company after completing their
apprenticeship, so that dynasties of successive generations of master and
apprentice would become established in the Grocers, Stationers, Merchant
Taylors, or whatever.
If we consider the makers we have come across in this study, Allen was a
Grocer and Clockmaker, Greatorex a Clockmaker, Thompson a Stationer,
Culpeper and Adams were Grocers, Rowley a Broderer (embroiderer), Mar-

Shopping for Instruments in Paris and London


shall a Turner, Patrick a Joiner, Moxon a Weaver and Stationer, Martin a
Goldsmith, and Smith, Pyefinch, Nairne, and Watkins were Spectaclemak-
ers. Many other seemingly unlikely trade affiliations were represented
throughout the instrument makers. With the exception of the Specta-
clemakers, the instrument makers did not, of course, practice any of these
trades, but they did belong to the companies and assumed such obligations
as holding office. 74
Instrument historians have been unsure what to do with this information,
apart from being amused at the idea of makers being classified as grocers,
dyers, fishmongers, haberdashers, and so on, and using the surviving guild
records as a powerful tool for dating makers and their signed instruments.
However, the involvement of the makers in a variety of companies will have
brought them into contact with a range of trading practices and given them
a familiarity with the general commercial life of London.
One very important development in London, which is particularly evident
at the beginning of the eighteenth century, is trading across the traditional
distinction between mathematical and optical instruments. Mathematical
instruments were made by specialist engravers, while the relative newcomers
were the optical instrument makers, who had emerged from among the most
skilled makers of spectacles. In addition to these two categories, there were
now also instruments of natural philosophy, derived from the new enthusi-
asm for experiment. There is no prima facie reason why these three types of
instrument should be retailed together and why they should eventually be
accommodated by a single category, "scientific instrument"-that it seems
natural to us is simply our inheritance from a conjunction formed in the eigh-
teenth century-but the possibility was created by company regulation in
London, where there were no restrictions on what company a maker might
join or what he might make or sell.
Two of the makers we have encountered are particularly associated with
early trading across the mathematical/optical distinction, namely Culpeper
and Rowley, but it soon became widespread, and with the addition of exper-
imental instruments of natural philosophy, it was characteristic of the oper-
ations of many makers, among whom Adams and Martin were prominent.
By this account there is indeed a link between commerce and the represen-
tation of nature: the very formation of the category of scientific instrument
owed something to the regulation of trading in London, while the makers'
acumen in encouraging the extraordinary growth of consumer interest in
natural philosophy may well be related to their association with colleagues in
companies dealing with luxury consumer goods, such as the Broderers or the
Merchant Taylors.
A few makers managed to trade without joining companies, especially
those associated with the Royal Society, either as fellows or through their

390 James A. Bennett


close connections with senior astronomers and observatory commissions.
Among those we have mentioned, this would apply to the Sissons, Bird,
Short, and Ramsden. We have seen that there were links between experi-
mental demonstration as presented at the society and as organized in the
shops of makers. Might there be another relationship with the commercial
world? Might the Royal Society have acted in some respects in the role of a
company, by making it possible for some makers to operate outside the usual
obligations to the city? It might be said that this simply reflects the declining
power of the London companies, but makers could scarcely have begun to
flout their authority without some alternative form of support and patronage.
This was not a formal responsibility of the Royal Society, but by virtue of
its royal charter it had the status, rights, and privileges of a chartered corpo-
ration. George Graham, a Royal Society Fellow, who established the role of
the elite, specialist maker of astronomical instruments in London, was the
first to employ assistants in the instrument-making side of his business with-
out binding them as apprentices. This was how Sisson and Bird entered the
trade, and they never joined a company. (Graham was a Clockmaker, hav-
ing served an apprenticeship under Thomas Tompion, and took apprentices
in the company for that side of his activities.) Lalande noted the status
enjoyed by makers such as Sisson and Bird within the circles of the Royal
Society and among the important astronomers-a de facto if not de jure
measure of protection, which allowed them to avoid the attentions of the
regular companies. Short was introduced to the London mathematicians
through his patron in Edinburgh, Colin Maclaurin, and was a fellow before
he moved permanently to London. Ramsden entered the trade through his
employment by Jeremiah Sisson, and he became a fellow in his own right;
his best-known employees, Matthew Berge, John Stancliffe, and Thomas
Jones (a fellow), had no company affiliations.
Thus the entanglement between the Royal Society and the commercial
life of the instrument makers had at least two aspects: the makers' shops had
some of the functions we associate with the society in relation to natural
knowledge, and the society in effect had some of the functions of a livery
company.

EPILOGUE: BUGGE IN PARIS

By the time Thomas Bugge visited France at the end of the eighteenth cen-
tury, the old guild structure and its regulatory apparatus had been swept
away by the Revolution. 75 Already there are clear signs of a revival of the
Paris workshops. Bugge's account has something of the flavor of Bernoulli's
letters from London: there is plenty to say. Occasionally Bugge enters the

Shopping for Instruments in Paris and London 39 I


caveat that the work is not quite as good as in London, but in general he is
very positive, and some of the work he sees is excellent.
Lenoir had a large workshop, which Bugge was able to visit, and he made
a wide range of mathematical and astronomical instruments. The optical
instrument maker Laroche had good reflecting telescopes and achromatic
refractors, and he had lots of conversation. For the first time among the
records of these visitors, we hear of a French maker engaging in extended
discussions and demonstrations. Dumotiez made physical instruments,
which Bugge bought and recommended; he did not only offer the standard
designs, but would work from the customer's drawings. Fortin was very
good, though expensive, while Betalli and Perrical were excellent glasswork-
ers. Behind these individual stories, there were general signs of a consumer
market, with between twelve and fourteen shops selling instruments near the
Pont Neufin the Quai de tHorloge and several more in the Palats Royal.
In 1801 Lalande announced that the work of Lenoir had demonstrated
that "the French industry is no longer behind the English." This was pre-
mature, but certainly a change was under way that would bring the two tra-
ditions into much more equal competition by the midcentury. Comparison
between Paris and London is not so artificial a perspective for the historian
as it may appear. It was present, implicitly or explicitly, in most of the
accounts we have examined, and it occasionally surfaces in remarks made by
the makers themselves. As an old man, Lenoir designed a dividing
engine-the machine for which the English makers, particularly Ramsden
and Troughton, had been especially celebrated. It is not clear what he can
have meant in technical terms, but the emotional significance of his remarks
was clear when he explained, "I have not tried at all to imitate the English;
on the contrary I dare to assert that my method is completely French."76

I. "Me voici a Londres depuis 8 jours; je ne puis vous parler encore d'Astronomes ni d'Ob-

servatoires; mais je vous serai part de la surprise agreable OU est jette un Astronome en par-
courant les rues de cette Capitale. Vous aves surement oui parler de la richesse & de I'eclat des
boutiques de Londres, mais je doute qu e vous vous representies com bien l' Astronomie con-
tribue a la beaute du spectacle: Londres a un grand nombre d'Opticiens; les Magasins de ces
artistes sont remplis de Telescopes, de Lunettes, d'Octans &c. Tous ces instrumens, ranges &
tenus proprement, flattent l'oeil autant qu'ils imposent par les reflexions auxquelles ils don-
nent lieu." J. Bernoulli, Lettres astronomiques (Berlin, '77,),63'
2. "Vous connoisses tres bien Ie nom de M. DOLLOND; l' Artiste celebre" Bernoulli, Let-

tres, 65 .
3. For foreign accounts of London instrument makers, see G. L'E. Turner, "The London
Trade in Scientific Instrument Making in the Eighteenth Century," VIstas in Astronomy 20
('976): '73 - ,82.

392 James A. Bennett


4- Balthazar de Moncon ys, Journal des voyages de Monsieur de Monconys (Lyons, 1665 - 66).
5. "vne feiiille de verre sur laquelle ces obiects sont peints," Monconys,Journal, 17-18.
6. "ie Ie trouuay dans son element, enfonce dans Ie commerce des Physiciens, & ne respi-
rant que machines, & que nouelles experiences." Samuel Sorbiere, Relation d'un voyage en
Angleterre (Paris, 1664),66.
7. Henry Oldenburg, The Correspondence of Henry Oldenburg, ed. A. Rupert Hall and
Marie Boas Hall, 9 vols. (Madison: University of Wisconsin Press, 1965-73),2: 285-291.
8. Ibid., Vol. 2: 271.
9. A. D. C. Simpson, "Richard Reeve-the English Campani-and the Origins of the
London Telescope-Making Tradition," Vistas in Astronomy, 28 (1985),357-365, For more
detail on Reeve, see AD.C. Simpson, "Robert Hooke and Practical Optics: Technical Support
at a Scientific Frontier," in Robert Hooke: New Studies, ed. Michael Hunter and Simon Schaf-
fer (Woodbridge: Boydell Press, 1989),33-61. The much earlier shop of Elias Allen, in the
1620S, seems to have been a place of resort for practical mathematicians, a source of news and
a means of communication. See William Oughtred, The Circles of Proportion (London, 1633);
Hester Higton, Elias Allen and the Role of Instruments in Shaping the Mathematical Culture of
Seventeenth-Century England, Ph.D. dissertation, University of Cambridge, 1995.
10. E. G. R. Taylor, The Mathematical Practitioners of Tudor and Stuart England (Cam-
bridge: Cambridge University Press, 197°),220-221.
I I. Zacharias Conrad von U ffenbach, London in 1710 from the Travels of Zacharias Conrad

von Uffenbach, ed. W. H . Quarrell and Margaret Mare (London: Faber & Faber, 1934).
12. Ibid., 158.
13. Ibid., 146.
14. Gloria Clifton, British Scientific Instrument Makers 1550-1851 (London: Zwemmer,
1995), 194·
15· Uffenbach, London, 77.
16. Ibid., 145.
17· Ibid., 173.
18. Ibid., 77-
19. Ibid., 168. For the received early history of the reflector, see Henry C. King, The His-
tory of the Telescope (New York: Dover, 1979),67-92.
20. Uffenbach, London, 168.
21. A. Q. Morton and J. A. Wess, Public and Private Science: The King George III Collec-
tion (Oxford: Oxford University Press, 1993),39-65.
22. Ibid., 77-78.
23. Ibid., 168- 170.
24. M. Lister, A Journey to Paris in the Year 1698 (London, 1699),80.
25. Maurice Daumas, Les instruments scientifiques aux XVIIe et XVlIIe siicles (Paris:
Presses Universitaires de France, 1953), 97- I l 3, 339-385.
26. J. Lalande, Journal d'un Voyage en Angleterre 1763, trans. Helene Monod-Cassidy
(Oxford: Voltaire Foundation, 1980).
27. "Ce n'est que par tatonnement." Lalande,Journal, 25.
28. Lalande, Journal, 60.
29· Ibid., 33·
30. Ibid., 32-3·
31. Ibid., 38, 42, 47·
32. J. A. Bennett, "The English Quadrant in Europe: Instruments and the Growth of
Consensus in Practical Astronomy," Journalfor the History of Astronomy 23 (1992): 1-14.
33. J. Bird, The Method of Dividing Astronomical Instruments (London, 1767); J. Bird, The
Method of Constructing Mural Quadrants (London, 1768).

Shopping for Instruments in Paris and London 393


34. Lalande,fournal, 72 .
35. Bernoulli, Lettres, 72 - 74.
36. "fait des cours de phis[iqueJ ridiculement," Lalande, Journal, 44-
37· Bernoulli, Lettres, 70,74,96.
38. Ibid., 126- 132.
39. Ibid., 68- 69.
40. Ibid., 65-66.
4l. Ibid., 138-173-
42. Thomas Bugge, Journal of a Voyage through Holland and England, 1777, ed. K. M0ller
Pedersen (Aarhus: Department of History of Science, 1997)'
43. Ibid., 128- 13l.
44- Ibid., 154- 155 .
45. Ibid., 128 - 129.
46. Ibid., 130- 133.
47· Ibid., 130- 133.
48. Ibid., 140- 145.
49. Ibid., 162- 17l.
50. Ibid., 296-30l.
5l. Ibid., 156-161, 170- 177.
52. Ibid., 178-185, 191.
53. Ibid., 184- 189.
54- Ibid., 254- 255.
55· Ibid., 344-345·
56. Ibid., 198-208.
57. Martinus Van Marum, Martin Van Marum. Llfe and Work, ed. R. J. Forbes, 6 vols.
(Haarlem: H . D. Tjeenk & Zoon, 1969-76), Vol. 2: 220-239.
58. Maurice Daumas, Scientific Instruments of the Seventeenth and Eighteenth Centuries and
Their Makers, trans. M. Holbrook (London: Batsford, 1972), 148, 330.
59. Van Marum, Life and Work, 2: 266-272.
60. Ibid., 2: 270-272.
61. "L'Observatoire avait ete muni alors de muraux, de grands quarts de cercles mobiles,
ouvrages des Langlois, des Canivet, des Lennel, qui etaient en ces terns-Iii les plus celebres
constructeurs d'instrumens d'astronomie. Mais au moment OU je pris la direction de l'Obser-
vatoire, ces vieux talens etaient eclipses par les Bird et les Ramsden, artistes anglais qui
avaient porte leur art iI la plus haute perfection, laissant bien loin derriere eux les Fran<;:ais, iI
qui ils avaient enleve presqu'entierement Ie commerce des instrumens d'optique et de math-
ematiques." J. D. Cassini, Memoires pour servir a l'histoire des sciences et a celle de l'observatoire
royal de Paris (Paris, 1810),4.
62. "Com bien de fois je me suis indigne contre des personnes qui, me priant de leur pro-
curer un bon instrument, Ie marchandaient comme une aune de drap! Que! prix coiltera-t-il?
me demandait-on; pas aussi cher, sans doubte, qu'un instrument anglais? Pourquoi non?
repondais-je; voulez-vous etre juste? si I'instrument ne vaut rien, ne Ie prenez pas: mais s'il est
bon, payez-Ie ce que l'ouvrier demandera, vous Ie fit-il meme acheter plus cher que s'il etait
anglais." Cassini, Memoires, 5.
63. "j'irai voir ces superbes instrumens anglais, examiner ces chef-d'oeuvres de l'art, et ces
modeles que notre premiere ambition est d'imiter en France . .J'aurai occasion de voir et de
cultiver les Dollond, les Stroughton [Troughton]' les Ramsden." Cassini, Memoires, 20 .
64. "sans doubte il sera flaw: de travailler pour l'Observatoire royal de Paris, OU il n'y a
encore aucun de ses ouvrages." Cassini, Memoires, 20.

394 James A. Bennett


65. "je reconnus que malgre tous nos efforts nous n'aurions jamais en France un artiste
aussi consomme; toute rivalite, toute comparaison me parurent desormais impossibles vis-a-
vis d'un si grand talent." Cassini, Memoires, 23.
66. "Sortant un jour d'un de ces entretiens OU j'aimais tant a m'engager avec lui et a m'in-
struire, je dis a un etranger non moins enthousiasme que moi du merite de M. Ramsden: En
verite, cet hom me est une machine electrique qu'il suffit de toucher pour en tirer une etin-
celie. Rien de plus juste que votre comparaison, reprit vivement I'etranger, car vous pourriez
fort bien ici ne tirer que des etincelles." Cassini, Memoires, 23.
67. "Ne croyez pas que ce soit indifference ou paresse de sa part: bien au contraire. Mais
une idee nouvelle, une difficulte a vaincre ou un instrument d'un nouveau genre qu'on vien-
dra lui proposer vont attirer toute son attention et lui faire abandonner tout ouvrage com-
mence." Cassini, Memoires, 24.
68. "je la crois digne d'aiguillonner votre genie." Cassini, Memoires, 28.
69. "Le courant de mon atelier me fournit suffisamment l'argent necessaire. Mon princi-
pal objet en vue, c' est la perfection des instrumens. Cette partie de notre profession est encore
dans son enfance." Cassini, Memoires, 179.
70. See D. W. Dewhirst, "Meridian Astronomy in the Private and University Observato-
ries of the United Kingdom: Rise and Fall," Vistas in Astronomy 28 (1985): 147-158.
71. Anita McConnell, "From Craft Workshop to Big Business-The London Scientific
Instrument Trade's Response to Increasing Demand, I750-I820," London Journal, 19 (1994):
36-53.
72. J. A. Bennett, "Equipping the Radcliffe Observatory: Thomas Hornsby and his Instru-
ment-Makers," in Making Instruments Count: Essays on Historical Scientific Instruments Pre-
sented to Gerard L'Estrange Turner, ed. R. G. W. Anderson, 1. A. Bennett and W.P. Ryan
(Aldershot: Variorum, 1993), 232-241.
73. Anthony Turner, "Mathematical Instrument-Making in Early Modern Paris," in Lux-
ury Trades and Consumerism in Ancien Regime Paris, ed. Robert Fox and Anthony Turner
(Aldershot, I998), 63 - 96.
74. For the makers and the London guild companies, see Joyce Brown, Mathematical
Instrument-Makers in the Grocers' Company 1688-1800 (London: Science Museum, 1979); M.
A. Crawforth, "Instrument Makers in the London Guilds," Annals of Science, 44 (I987):
3 I9 - 377; Clifton, Directory.
75 . Thomas Bugge, Science in France in the Revolutionary Era. Described by Thomas Bugge,
ed. M. P. Crosland (Cambridge, Mass.: Society for the History of Technology, I969).
76. "l'industrie fran~aise ne Ie cede plus a celie des Anglais"; "je n'ai point cherche a imiter
les Anglais; j'ose au contra ire avancer que rna methode est toute fran~aise"; Daumas, Les
instruments scientifiques, 365.

Shopping for Instruments in Paris and London 395


This page intentionally left blank
EPILOGUES
This page intentionally left blank
A World of Wonders,
A World of One

LISSA ROBERTS

When nature is reborn in us, it calls itself art.


-Max Weber, American painter

W hen Copernicus placed the sun at the center of the cosmos, he


was driven by an aesthetic compulsion. The efforts of other
astronomers, as he told it, were "just like someone taking from
various places hands, feet, a head, and other pieces, very well depicted, it
may be, but not for the representation of a single person; since these frag-
ments would not belong to one another at all, a monster rather than a man
would be put together from them." If the world was indeed "created for our
sake by the best and most systematic Artisan ofall," as Copernicus believed, it
was bound to exhibit "symmetry of its parts."! The task and burden of the
philosophical astronomer was to uncover the cosmic order manifested in its
unified arrangement. Copernicus struggled to do just this, breathing his last
as De Revolutionibus was being printed in 1543.
This tragically heroic tale is the starting point for many renditions of the
"Scientific Revolution," though authors tend not to dwell on the aesthetic
dimension of its history. It is not my plan here to rehearse the generally told
story of the Scientific Revolution or even to refute it. I want instead to intro-
duce a cluster of issues that sit within the above snippet of Copernicus's text
and find resonance in this collection of essays. By doing so, I hope to get at
a tension that exists in our understanding of the history of early modern
European art and science, one that needs to be resolved if we are to achieve
the kind of historical synthesis that this book aspires to stimulate. At the
risk of anticipating what I set out below, this tension can be understood
briefly as existing on two related levels. Historiographically, an explanatory
cleft lies between the history of the mathematical sciences, on the one hand,

399
and of the descriptive and experimental sciences on the other. A tandem
divide separates historians who discuss the arts and sciences of early modern
Europe as having been undergirded by the aesthetic principles of unity and
order from those who emphasize the aesthetic principles of variety and
uniqueness. Once I spell this tension out more fully, I want to indicate what
I see as the potential of this book's collective vision for resolving it. And
finally, I want to argue that the tension between the principles of unity and
variety was already recognized and debated during the early modern period
itself. What ultimately served to resolve it historically was a focus on activ-
ity, both in terms of philosophical reflection and in terms of everyday life, as
increasingly dominant market activities came to shape the production of
goods and ideas in a way that reflective philosophy never had the power to
do on its own. This book's focus on practice, then, couldn't be more fitting.

THE ESSENTIAL TENSION?

A man cannot tell whether ApeHes, or Albrecht Durer, were more tri-
fler; whereof the one would make a personage by geometrical propor-
tions; the other, by taking the best parts out of divers faces to make one
excellent.
-Francis Bacon, "On Beauty"

I begin with Copernicus's carefully chosen analogy between the astronomer's


depiction of the cosmos and the anatomist's depiction of the human micro-
cosm. Without speaking of influence, it is certainly striking that Andreas
Vesalius's De Humani Corporis Fabrica was published in the same year as De
Revolutionibus. A fellow Paduan (Copernicus had attended the University of
Padua prior to Vesalius's arrival), Vesalius too presented an anatomized
image of the body-human microcosm rather than cosmic macrocosm -
that exhibited the ordered symmetry of its parts. Indeed, whether associated
with Plato, Vitruvius, Horace, or any other ancient antecedent, the ideal of
nature as an harmonious, unified whole that was constructed from a divinely
(quasi-)mathematical blueprint was a recurrent trope during the Renais-
sance. 2 Not only did this feed new representations of the world, its parts, and
inhabitants (and methods for investigating them), it inspired various artists
and architects to adopt new principles of perspective as a foundation for their
work as well.
That these were more than coincidental developments among a range of
natural philosophical and artistic endeavors is attested to by repeated claims
that reverberate across apparent divides. Architects drew inspiration for
their designs from contemporary depictions of the human body ("nowhere

Lissa Roberts
else than that sacred temple made in the image and likeness of God, which
is man-in whose make up all the other wonders of nature are com-
prised"),3 while descriptions of the human body as a divinely constructed
"temple" echoed Copernicus' (and others') explicit description of the cos-
mos. 4 From microcosm to macrocosm, from nature to products of human
design, the guiding watchwords were (divinely rooted) unity and order.
Historians have argued that this perspective was crucial for the subsequent
development of a unitary natural philosophy that sought to portray the uni-
verse as a system with a structure and motions that were bound by divinely
ordained mathematicallaws. 5 At the heart of what is traditionally presented
as the Scientific Revolution stood the urge to read the divinely authored
"book of nature" which, for men such as Galileo, was written in the lan-
guage of mathematics. Because nature was created and maintained by uni-
tary laws, according to this perspective, it must be possible to uncover them
by examining natural phenomena in terms of the measured and measurable
patterns they exhibited. Whether this was done through experimentally
based induction, the application of mechanical analogy or mathematical
analysis, the goal was the same-to reveal the order and unity of nature. 6
If combined historical and historiographical scrutiny has taught us any-
thing, it is that there are other ways to tell this story. In his classic study of
European art history, for example, Arnold Hauser referred the Renaissance
urge toward unity to a "spirit" very different from Neoplatonism.

The principles of unity which now become authoritative in art ... express
the same dislike for the incalculable and the uncontrollable as the economy
of the same period with its emphasis on planning, expediency, and calcula-
bility; they are creations of the same spirit which makes its way in the orga-
nization of labour, in trading methods, the credit system and double-entry
bookkeeping, in methods of government, in diplomacy and warfare. The
whole development of art becomes part of the total process of rationaliza-
tion .... The things that are now felt as "beautiful" are the logical conformity
of the individual parts of a whole, the arithmetically definable harmony of
the relationships and the calculable rhythm of a composition ... and the
mutual relations of the various parts of the space itself.?

Hauser's narrative fits perfectly with the minority voice of Marxist histori-
ans of science (and their fellow travelers) mentioned by Paula Findlen and
Pamela Smith in this book's introduction, who sought to root the develop-
ment of modern science in the rise of capitalism. According to this version of
history, it was the materially grounded march of economic development that
both brought unity to a range of human endeavors and posited unity as the
philosophical expression of an historic process of rationalization.

A World of Wonders, A World of One


It is important to note that, while I began with a saga of intellectual
struggle and then countered it with one built on the dialectic of material
conditions, both these accounts privilege the linked aesthetic principles of
unity and order. They portray this aesthetic as prescribing method, deter-
mining boundaries between science and craft, and giving shape to both sci-
entific and artistic production. Strikingly, the essays in this volume imply a
rather different tale of the relations between art and the study of nature.
While it also comes in two (though expressly not autonomous) versions-
we might call one a "history of representing" and the other a "history of
intervening" - what brings both these versions together, by and large, are
the inverse aesthetic principles of variety and uniqueness.8 According to
the authors in this volume, what seems to have marked and demarcated
both nature and its investigation on one side, and artistic production and
appreciation on the other, were the thirst for and apprehension of unique-
ness and wonder. 9 Ironically, while historians' focus on the aesthetics of
unity has led to an image in which a boundary separates the arts and sci-
ences, the present focus on variety paints a picture of continuity between
them.

SCIENCE: A MATRIX OF MARVELS

.. a marvelous museum ... a small theater in which all the wonders of


nature can be displayed in miniature.
-Giralmo Porro's description of Padua's botanic garden, 1591

As was the case with so many beliefs, Aristotle's authority helped legitimate
the traditional opposition between art and nature. 1O In contrast to this tradi-
tion that fostered a fascinating history of art's competition with natural
beauty and wonder, the dictum that "beauty is truth" became a highly
respectable one by the beginning of the eighteenth century, especially as for-
mulated in the third Earl of Shaftesbury's aesthetics. I I Even before his artic-
ulation of this view, however, many of the actors presented in this book
accepted and modified the equation by defining beauty in terms of the indi-
viduating aesthetic principles of variety and uniqueness. Hence, the con-
struction and increase of knowledge required, first, the collection of
extensive stores of singular phenomena. On one hand, this connects with
discussions of how these naturalia, artificialia, scientific instruments, and
models (the general categories of items that stocked these collections) were
identified, produced, and distributed, to which we will return. On the other
hand, it leads to the question of how the increasing range of phenomena
were stored, categorized, represented, and used.

Lissa Roberts
The most obvious collection sites were gardens and cabinets where aris-
tocrats, merchants and corporations (be they cities, universities or amateur
societies) alike amassed and displayed their holdings. To these Claudia Swan
adds mimetic illustrations and grids: representational collections in which
illustrated (either pictorially or verbally) elements were laid out in a way that
at least partially corresponded to their physical position in space and time.
The reference to Foucault is blatant.

The Classical age gives history a quite different meaning: that of under-
taking a meticulous examination of things themselves for the first time,
and then of transcribing what it has gathered in smooth, neutralized, and
faithful words . . .. The documents of this new history are not other words,
texts or records, but unencumbered spaces in which things are juxtaposed:
herbariums, collections, gardens; the locus of this history is a non-tempo-
ral rectangle in which, stripped of all commentary, or all enveloping lan-
guage, creatures present themselves one beside another, their surfaces
visible, grouped according to their common features, and thus already vir-
tually analyzed, and bearers of nothing but their own individual names.12

If the analogy between material sites and their representations seems only
partial- painters, for example, were free to portray bouquets of blooming
flowers that, in fact, blossomed at different times-the difference was actu-
ally only a matter of degree. For what all these venues had in common was
that they formed an intersection of art and the objects of nature that endeav-
ored to escape the strictures of natural time and space, and it was out of that
juncture that knowledge (not to mention value) was produced.
Since the Renaissance, gardens were spoken of as embodying a "third
nature"-a world borne of the marriage between the creative forces of pri-
mal nature and human artifice, dedicated simultaneously to utility and
pleasure, built to grow plants out of both their native spaces and times. 13
And, as a number of essays in this collection remind us, the Wunderkam-
mern of Europe were also filled with a range of items-some found in
nature, others embellished or created out of whole cloth by artists, artisans,
and instrument makers - drawn from around the world and across the
scope of the human imagination. If gardens were experimental sites for
transplanting exotics, creating new varieties and advancing "physick"
through the cultivation of medicinal plants (not to mention the architec-
tural fancies they housed), cabinets held collections that also stimulated
work as well as wonder. They fed the intellectually productive fires of
anatomy, natural history, and metallurgy, while stoking the skills of speci-
men production and preservation, of artisanal precision and presentation.
Opening their doors opened the world in miniature, rearranging the matrix

A World of Wonders, A World of One


of space and time for the benefit of human understanding, consumption,
and manipulation.
Representations of nature - be they in the form of still life paintings,
anatomy sheets whose flaps could be raised to reveal bodily organs, natural
histories or maps - helped to standardize portrayals of natural phenomena
by adopting a naturalist style that made them, in all their uniqueness,
amenable to surface comparison and categorization. l4 They also made them
transportable, extending the potential for their interpretive consumption
and exploitation. By domesticating the exotic and refusing to be bound by
the ordering strictures of natural space and time, these representations
afforded their viewers the opportunity to travel on their own terms-in the
comfort of their homes and at a pace set by their leisure - through worlds
laid out for their perusal and profit.

BEYOND THE GREAT DIVIDE: TOWARD A HISTORY OF DOING

L et physicians and confectioners and the servants of the great houses be


judged by what they have done.
-Isak Dineson, "The Dreamers"

Before turning to the question of how the contents of these collections came
to be there, I want to address another important set of contrasts between the
sagas with which I began and those bound in this volume. In addition to
revolving around mirror-image aesthetic principles-unity and order, on
the one hand, variety and uniqueness on the other - they also appeal to dif-
ferent forms of natural knowledge and art. The first privileges the mathe-
matical sciences-astronomy and physics-and unified perspective. The
second focuses largely on natural history, alchemy, and medicine, along with
representational description and handcrafted objects. Are we flirting here
with a coupling of traditional oppositions: mathematical versus experimental
sciences together with "Italianate" versus "Netherlandish" representation?
Traditionally, such a coupling worked at the expense of the "experimen-
tal sciences" and northern art. Consider Thomas Kuhn's claims, as referred
to in this volume's introduction, that tied Baconians to commercial interests
and thereby cut them off from the Scientific Revolution. Consider too
Hauser's condescending appraisal. He writes that rationalism, the burgeon-
ing hallmark of the Italian Renaissance, "does not remain restricted to Ital-
ian art; but in the North it assumes more trivial characteristics than in Italy,
it becomes more obvious, more naive."l;
The art historian Svetlana Alpers embraces these coupled dichotomies, but
with a different strategy in mind, that of championing what she considers the

Lissa Roberts
descriptive nature of Dutch seventeenth-century art. Not interested in resolv-
ing these claimed distinctions, she focuses instead on presenting a more positive
interpretation of experiment and description. Alpers explicitly posits a parallel
between the Baconian project and Dutch painting, with its purported penchant
for description and its "established alliance with those craftsmen - goldsmiths,
weavers, glassblowers and geographers-whose products became the crafted
objects in their representations." Bacon's world, Alpers tells us,

is stilled, as in Dutch paintings, to be subjected to observation. Detailed


descriptions, compiled almost without end and fitted into the table, dis-
place time, since each observation is separate from the next. Indeed,
despite its title Bacon's natural history displaces history.... It is, like the
Dutch art with which we have linked it, description, not narrative. 16

Interested to portray seventeenth-century art as essential to its contempo-


raries' understanding of the world, Alpers commends Bacon's dictum that
"the nature of things betrays itself more readily under the vexations of art
than in its natural freedom."17
Should we too leave these paired dichotomies unresolved, with or with-
out debating their (hierarchical) relation? To do so, I fear, fails to answer the
challenge of crafting an inclusive explanation for what has come to be called
the Scientific Revolution. Given the current unpopularity of grand narra-
tives, this is perhaps a good thing. It could well be that no overarching expla-
nation exists-or needs constructing-for the concatenation of activities
and ideas associated with the investigation and representation of nature in
early modern Europe. On the other hand, this volume houses a pregnant
suggestion that deserves to be fleshed out: that we replace a history of deeds
by a history of doing. Once homo faber takes center stage, apparent opposi-
tions begin to disappear. IS It is unclear whether their disappearance is a sim-
ply a function of our redirected attention, but we will discover this only
when such an active history is told in more detail.
Klaas van Berkel takes us a step along the way toward writing this kind
of history, though he still couches the introduction to his essay in this volume
in the binary terms of mathematical and experimental sciences. The major
actor of his story, like so many early modern practitioners of science, did not
respect the boundary that these two categories imply. Indeed, I doubt Meijer
recognized the existence of such a boundary; he practiced astronomy, was a
successful engineer, and published books that beautifully illustrated both his
own inventions and experiments done before the Roman Accademia Fisico-
matematica, of which he was a member.
While this broad range of activities is characteristic of many "scientific"
practitioners in early modern Europe, whether obscure or as famous as

A World of Wonders, A World of One


Galileo (consider also that Copernicus trained in both medicine and law, as
well as learning to paint and translate) and deserves attention, I want to focus
here on Meijer's work as an engineer. Then as now, (successful) engineering
combined mathematical ability with experimental and entrepreneurial skills.
Besides being honored in scientific academies across the continent, engineer-
ing was taught in courses housed (with varying degrees of official recogni-
tion) at major universities in the Netherlands during portions of the
seventeenth and eighteenth century. No less a figure than Willem 's
Gravesande (known generally, in caricatured fashion, for having brought
Newtonian science to the Continent) taught "Dutch mathematics," as it was
known, for a time at Leiden University.19 Institutionally as well as intellectu-
ally, then, engineering existed precisely at the point denied by a historiogra-
phy that categorically separates mathematical and experimental inquiries of
nature, relegating it to a lesser status that bears the adjective "applied."
Chandra Mukerji's essay underlines the importance of engineering while
setting it in a more blatant political context. What remains of interest
throughout her essay is the depiction of knowledge as constituted by a net-
work of specialized knowing and skills; a coordinated distribution that
crossed any number of disciplinary, social, and political boundaries in order
to understand nature and thereby harness it for human use. Perhaps this is
the most fitting way to investigate a period during which the loci of learn-
ing, knowledge production, and application were multiplying. Focusing on
activities and the networks required to accomplish them frees us from the
constraints entailed in studies whose inquiries are organized in terms of
either ideas or institutions.
One might respond by saying that this model is applicable only to
instances involving the creation of what Cicero long ago termed a "second
nature;" that is, a landscape engineered for human use. 20 Indeed, Mukerji
pointedly speaks of representations of nature as tools that were expressly
used to transform that which they represented. We might see the maps dis-
cussed by Alison Sandman in a similar light. Though the "network" in
which they were placed housed contention over what constituted the proper
representation of nature and how that representation ought to be used-for
the pragmatic purpose of navigation or that of "objectively" mapping the
world-and, therefore, manifested contours that could only be solidified
once the controversy was ended, the final product was one that brought
together the expertise of navigators, cosmographers, illustrators, and politi-
cians in order to assert (and challenge) Spanish imperial claims that aspired
to remake the entire globe. 21 We might, then, further see Benjamin
Schmidt's discussion of exoticism as an inverse reflection of this process
whereby the Dutch compensated for their loss of empire by informing and
controlling the market for geographic representations.

Lissa Roberts
But I think the true test for this approach-that is, a "history of doing"-
is whether it can go beyond this level of application and resolve the claimed
dichotomy of mathematics and experiment in general. This would be to
argue that the application of any and all human activity to the world,
whether mental or manual, includes an act of transformation that can there-
fore be investigated in essentially the same way. One would, in other words,
have to cast mathematical physics and astronomy in the same explanatory
mold as engineering and experimental manipulations of nature. And one
would have to present a convincing argument that the mathematical repre-
sentation of nature resulted from the same kinds of sociointellectual
processes and transformed nature through its representation in a way that
could further be impressed in networks of knowledge and power.
From the standpoint of intellectual history, this entails developing the
claim that mathematics had evolved, by the seventeenth century, from the
ancient Greek conception "as an inventory of (ideal) mathematical entities
and their absolute properties" to a formal language of relations, capable of
representing mathematical and nonmathematical relations alike and thereby
capable of "accounting" for observable changes in nature in a powerfully
general way. This need not be an idealist history, for we know that the insti-
tutional context of early modern European mathematics increasingly per-
mitted application of a method drawn from one discipline to the subject
matter of another discipline, something strictly forbidden in Aristotelian
logic as leading to category mistakesY It further involves that we take seri-
ously (that is, consider as more than handy analogies) the connections
between what came to be called the "mechanical philosophy" and the mech-
anisms-living or otherwise-that Descartes and his contemporaries expe-
rienced firsthand as well as the intimate links between Newton's alchemy
and his conception of gravity as an active principleY Finally, it involves
writing a full-blown cultural history of the mathematical sciences; that is, a
history that integratively situates them in the broader cultural context of
their time and simultaneously shows their practice to have been constitutive
of that culture. 24

THE MARKET AS MEDIATOR BETWEEN UNITY AND THE UNIQUE

If phenomena are not all linked one to another, there can be no philosophy.
-Denis Diderot, L'Interpretation de la nature

If beauty and truth came to be seen as inseparable, Anne Goldgar's essay


reminds us that, increasingly, so were beauty and profit. To quote her,
"What is rare is beautiful, and what is beautiful is profitable." While gar-

A World of Wonders, A World of One


dens, Wunderkammern and the like might therefore be regarded as reposito-
ries of intellectual capital, to borrow Mark Meadow's formulation, so too did
they relate to the nexus of cold, hard cash. Financing by private brokers,
trading companies, and empires made both "discovery" and control of the
globe increasingly possible and plausible. But a full range of technologies
was equally important and had, consequently, to be invented, "improved,"
and applied. These technologies varied from navigational techniques and
tools, on one side, to means of destruction and preservation on the other;
from technologies that named and represented to technologies that silenced
and produced transparence; from technologies that eliminated surprise to
those that encouraged awe.
The telos of this technological engagement with nature and its elements
was consumption, whether it involved ingesting herbals, imbibing facts, or
absorbing sights, sounds, and textures. But it was a consumption that was
conspicuous in the important sense that it gave rise to further productivity,
be it health, understanding, or enhancement of artistic and artisanal skills.
Hence production and consumption fed off each other in a spiral that had
more to do, this book argues, with the densely populated world of work and
wonder than with some grandly conceived process of rationalization.
Should we then be surprised if these developments involved an emphasis on
variety and uniqueness rather than unity and order, as implied by at least some
of the essays of this book? Or need we draw the conclusion that, in the end, this
was indeed the case? For if the market that supplied and benefited from the
growing urge to know and show appealed to its customers with the promise of
variety, so too did it see the rise of standardization. While this was perhaps
most obvious in the marketing of instruments and mechanical models, whose
operational success and popularity depended on the play between apparent
uniqueness and calibrated standards, the same can certainly be said for printed
matter, natural specimens, and handcrafted artificialia, all of whose value was
determined with reference to standardizing categories and notions of quality.
That we can detect the presence of a productive tension between these aes-
thetic ideals is clearly more than the result of historians' retrospection. Some of
the most stimulating philosophical debates of the early modern period rested
on this contest of emphasis. Spinoza, for example, who spent his days grinding
lenses in market-mad Amsterdam, spoke of God and the world in monist
terms. Leibniz, neither stranger to alchemy, automata, or imperial collections,
called this "the best of all possible worlds," by which he meant that God could
not but create a universe filled with as many distinct entities as did not involve
a contradiction. His views (and pretensions) led him to argue with Newtoni-
ans over the nature of God and God's relation to nature. 25 Is that relationship
marked by divine omniscience or omnipotence? Do we know God through
the unified order or the infinite wonders he creates?

Lissa Roberts
For many, the challenge was to find a way to say both. But as the eigh-
teenth century wore on, increasing numbers (though still relatively few)
questioned the need to refer to God at all when investigating nature. Where
the market led, philosophy followed. And while Adam Smith drew the
"invisible hand" to bring order to a commercial world of competing indi-
viduals, Voltaire looked around him and stated with pessimistic resignation
that this is indeed the best of all possible worlds.

Epigraph quoted in Martin Green, New York: The Armory Show and the Paterson Strike
Pageant (New York: Collier Books, 1988),249.

I. Quotations drawn from Copernicus, On the Revolution of the Heavenly Spheres (London:

David & Charles, 1976),25.


2. Robert Westman argues that Copernicus's appeal to unity was not so much an invoca-

tion of Neoplatonic ideals as it drew on the rhetoric of Horace, for which there was a high
regard at that time. "If a painter were willing to join a horse's neck to a human head and
spread on multicolored feathers, with different parts of the body brought in from anywhere
and everywhere, so that what starts out above as a beautiful woman ends up horribly as a
black fish, could you my friends, if you had been admitted to the spectacle, hold back your
laughter? Believe me, dear Pisos, that very similar to such a painting would be a literary work
in which meaningless images are fashioned, like the dreams of someone who is mentally ill,
so that neither the foot nor the head can be attributed to a single form." Horace, Ars Poetica,
lines 1-13, quoted in Robert S. Westman, "Proof, Poetics, and Patronage: Copernicus' Pref-
ace to De Revolutionibus," Reappraisals of the Scientific Revolution, ed. David C. Lindberg and
Robert S. Westman, (Cambridge, Mass.: Cambridge University Press, 1990), 183. I thank
Claudia Swan for also calling the link to Horace to my attention.
In a different vein, it is interesting to note that Thomas Laqueur goes so far as to argue
that the unity of Renaissance portrayals of human anatomy entailed depicting the female
body as a modification of the male-what he refers to as the "one-sex model." Thomas
Laqueur, Making Sex. Body and Gender from the Greeks to Freud (Cambridge, Mass.: Harvard
University Press, 1990), especially chapter 3.
3. This quotation is from Daniel Barbaro's (1567) edition of Vitruvius, cited by Jonathan
Sawday, The Body Emblazoned. Dissection and the Human Body in Renaissance Culture (Lon-
don: Routledge, 1996), 70.
4- According to Pierre Gassendi, Copernicus used the artistic skills he learned as an exten-
sion of his study of perspective to paint a self-portrait. "He [Copernicus] concerned himself
with all parts of mathematics so that, at one time, he dwelt especially upon [the study of] per-
spective, and , at another time, he took the opportunity to learn more about the art of paint-
ing, until he became skillfully practiced; and, it may be reported, as well, that he painted
himself excellently using a looking-glass. Whereupon, he began to seek advice about paint-
ing, for while contemplating travel, and above all to Italy, he resolved not only to sketch but
also, insofar as he could, to represent exactly whatever he found worthy of observation."
Quoted in Robert Westman, "Proof, Poetics and Patronage," 184. I chose to quote this excerpt
not only as a indication of the close links between artistic composition and astronomy as dis-
cussed in my essay, but also to comment on how intriguing it is to note that Copernicus appar-

A World of Wonders, A World of One


ently did not deem astronomical observation crucial to astronomy's reform (he reportedly
made no more than twenty seven observations himself), but considered observation of his
own surroundings to be crucial for artistic representation.
5. For the most eloquent, if complex, expression of this argument, see Amos Funkenstein,
Theology and the Scientific Imagination from the Middle Ages to the Seventeenth Century (Prince-
ton, N.J.: Princeton University Press, 1986).
6. While it is certainly true that Bacon did not advocate the mathematization of nature, he
did nonetheless distinguish his inductive view from "em pi ricks" who sought nothing more
than to uncover and put to use individual phenomena. See, for example, his New Organon,
aphorism 99.
7. Arnold Hauser, The Social History of Art (New York: Vintage Books, 1951), Vol. 2, IS·
8. For the categories of "representing" and "intervening," see Ian Hacking, Representing
and Intervening. Introductory Topics in the Philosophy of Natural Science (Cambridge: Cam-
bridge University Press, 1983). Hacking uses these words as a way to activate discussions of
rationalism and realism. I use them here to indicate, in an equally active way, histories of how
knowledge was constructed through its representation and how that construction process was
enabled through various means of intervening in nature, the workshop and the marketplace
(not to mention the political realm).
9. See also Lorraine Daston and Katharin e Park, Wonders and the Order of Nature,
1150-1750 (New York: Zone Books, 1998), 15 - 16.
IO. Aristotle, Physics, 2.1, 192b. 9-19.

II. J. v. Arregui, "La teologia de la belleza en Shaftesbury y Hutcheson," Themata 13


(1995): I I -35; J. Stolniz, "From Shaftesbury to Kant: The Development of the Concept of
Aesthetic Experience," Journal of the History of Ideas 48 (1987): 287-305'
12. Michel Foucault, The Order of Things. An Archeology of the Human Sciences (New York:
Vintage Books, 1973), 131.
13. There is a growing literature on the history of gardens. See, for example, John Dixon
Hunt, Gardens and the Picturesque. Studies in the History of Landscape Architecture (Cambridge:
M.LT. Press, 1994); Chandra Mukerji, Territorial Ambitions and the Gardens of Versailles (Cam-
bridge: Cambridge University Press, 1997).
14. Here I equate naturalism with Foucault's notion of the "table." See footnote IS . For
anatomy sheets, see Andrea Carlino, Books of the Body: Anatomical Ritual and Renaissance
Learning (Chicago: University of Chicago Press, 1999)' For the struggle to standardize maps,
see Alison Sandman's essay in this book.
IS, Arnold Hauser, Social History, Vol. 2, 16.
16. Svetlana Alpers, The Art of Describing. Dutch Art in the Seventeenth Century (Chicago:
University of Chicago Press, 1983), quotations on I03 and I09.
17. Bacon, quoted in ibid., 103.
18. For a view of homo faber's historical significance, see Hannah Arendt, The Human
Condition (Chicago: University of Chicago Press, 1958).
19. P. J. van Winter, Hoger beroepsonderwzjs avant -la -lettz·e. Bemoeiingen met de vorming van
landmeters en ingenieurs bzj de Nederlandse universiteiten van de 17e en 18e eeuw (Amsterdam:
Noord-Hollandsche Uitgevers Maatschappij, 1988) (Verhandelingen del' Koninklijke Neder-
landse Akademie van Wetenschappen, Afd. Letterkunde, Nieuwe Reeks deel 137).
20. The term Cicero used was alteram naturam.
21. For the relation between controversy resolution and the validation of network con-
tours in science, see Trevor Pinch, "Towards an Analysis of Scientific Observation: The
Externality and Evidential Significance of Observation Reports in Physics," Social Studies of
Science 14 (1985): 167- 187.

Lissa Roberts
22. Amos Funkenstein, Theology and the Scientific Imagination, 297-298. Funkenstein
makes the general claim that a new ideal of knowledge emerged in the seventeenth century
that equated knowing with doing, not only in the commonly accepted Baconian sense, but
more universally as a mirror of God's active intelligence.
23. For the mechanical philosophy see, for example, Otto Mayr, Authority, Liberty and
Automatic Machinery in Early Modern Europe (Baltimore: Johns Hopkins University Press,
1986). For Newton's conception of gravity as an active principle, see Betty Jo Tweeter Dobbs,
The Foundations of Newton's Alchemy, or the Hunting of the Greene Lyon (Cambridge: Cam-
bridge University Press, 1975).
24. Mario Biagioli's Galileo Courtier (Chicago: University of Chicago Press, 1993) is an
important step in this direction but falls short of offering a cultural history of the mathemat-
ical content of Galileo's natural philosophy. Instead he couples a well done history of its cul-
tural context with an internalist analysis of the mathematics itself.
On the needed symmetry of explanation-that is, the mutual constitution of science and
societylculture-see Bruno Latour, Nous n'atlons jamais iti modernes (Paris: La Decouverte,
1991).
25. See, for example, The Leibnitz-Clarke Correspondence (London, 1717). For an interest-
ing discussion of the debate, see Steven Shapin, "Of Gods and Kings: Natural Philosophy and
Politics in the Leibniz - Clarke Debates," Isis 72 (1981): 187-215.

A World of Wonders, A World of One


Questions of Representation

THOMAS DACOSTA KAUFMANN

T his book proffers a lavish array. Like some of the subjects they
treat, the essays collected here draw their materials from various
parts of Europe and touch on many parts of the globe. Including
discussions of a variety of disciplines, they offer an expanded view of early
modern history, and at the same time enrich earlier interpretations, espe-
cially of the history of science and the history of art. Yet while they intrigue
us with fascinating intellectual wares, these essays also invite us to consider
further some of the larger questions that they raise. This epilogue responds
to some of the selection's stimuli. l
Connections that may be established between individual essays and
across sections of this book indicate the general direction of this collection.
While not all the individual essays may be regarded as "interdisciplinary" in
approach, as a whole they suggest some ways in which considerations of
similar issues from multiple yet compatible points of view may mutually
enhance each other. While Mark Meadow points to an agent who he
believes played a key role in the establishment of early modern collections,
the Kunstkammer (here called, resonating with the title of the book, the
Wunderkammer), Paula Findlen discusses the trade in actual objects, the
curiosities that went into what was called in English the "curiosity cabinet."
While Chandra Mukerji points to the entrepreneurial interests that were
involved in cartography and hence in canal construction, Benjamin
Schmidt emphasizes that mapmaking itself was an important element in
what he refers to as a marketing strategy.
Most obviously, all the essays presented here thus eschew approaches that
may be called "internal" to the historiography of disciplines, which traced
transformations, or saw developments arising from within the traditional
concerns of their fields. Accordingly the trajectory of science was one that
involved methods, theories, and the accumulation of knowledge. 2 Classic
histories of science-for example, that of A. C. Crombie3-were histories
of scientific ideas; at most they may have intersected with the history of
ideas, as did an important older work by E. A. Burtt. 4 In the history of art a
comparable "internalist" tradition was the history of style, as exemplified by
Heinrich W6lfflin, Alois Riegl, or other art historians associated with
Vienna, another that of iconography, the study of subject matter in art, as in
the work of Emile Mile. 5 While it may seem that such approaches have long
been outmoded, and this collection parallels other recent efforts that move
firmly away from internal histories, it is worth recalling that the tradition
has survived, and the tendencies it expresses certainly formed a powerful
force at the time many of the contributors to this collection were being edu-
cated: so suggests the simultaneous appearance of important syntheses in the
historiography of art and the history of science thirty years ago. 6
Alternative approaches were, to be sure, long available. For example,
some of the intellectual origins of essays found here on the importance of
practical alchemy or on systems of representation may be traced back to the
works of Aby Warburg, Fritz Saxl, and other scholars associated with the
Warburg Institute on astrological and occult imagery, or to that of scholars
such as Erwin Panofsky on systems of representation, including studies of
perspective and proportion. 7 Books by scholars such as Martin Wackernagel
or Hans Floerke suggest that social and even economic histories of art were
not entirely lacking earlier, either. 8 Somewhat similarly, so witness the
works of Robert K. Merton and the influential study of Herbert Butterfield,
cultural and social historical approaches to the history of science also existed
in the earlier twentieth century.9 Nevertheless, if a self-conscious social or
economic history of art or science was practiced, it was most often to be asso-
ciated with a Marxist tradition, not just in parts of Europe dominated by the
Soviet Union, but in the West, both in the history of art, as exemplified by
Frederick Antal and Arnold Hauser, and also in the history of science, as
exemplified by J. D. Bernal. lO
At the same time, however, and almost in the same year that older sorts of
syntheses were produced-the early 197os-newer tendencies were astir in
art history." A new art history brought with it a host of new or restated con-
cerns: "gender, race, class" form a by now familiar litany. Hence, along with
arguments informed by literary theory and postmodern philosophy, arrived
other newer, or at least subtler, approaches to questions concerning the
socioeconomic aspects of art. More recently, and more to the point of this col-
lection, increasingly sophisticated studies of the commercial and economic
aspects of art of the early modern period have also appeared. The works cited
by Pamela H. Smith and Paula Findlen in their introduction are just some
among the noteworthy publications which have illuminated commercial and
economic aspects of art in the Low Countries, Germany, and Italy.'2

Questions of Representation
Similar tendencies emerged more or less simultaneously in the historiog-
raphy of science. While older surveys still marked out the course of scientific
ideas, the same years (the early I970s) that saw a new art history also wit-
nessed numerous expressions of a social history of science. 13 Similar sorts of
concerns were expressed: an increasing number of studies, including those
by some of the contributors to this volume, deal with the institutional nexus,
patronage relations, and personal networks of practitioners of science. 14 As
signaled by recent books on canonical figures such as Galileo Galilei and
Tycho Brahe, the social history of science has grown increasingly sophisti-
cated, with far-reaching implications. I) It has now been argued that even the
determination of scientific "truth" may be socially constructed.
This importance of social practices for the determination of knowledge,
in the thesis advanced by Stephen Shapin, recalls some of the arguments of
Michel Foucault-whom Shapin indeed cites.lt> In books not however cited
by Shapin, Foucault, to use his own diction, had enunciated an archaeology
of knowledge,1 7 and, more relevant to issues here, had suggested an "order
of things" which offered a view of sequentially changing epistemic systems.
For Foucault the "classic" period (roughly, of the "long" seventeenth cen-
tury, to use another term initiated in francophone historiography) was
marked by the episteme of representation. In the preface to Foucault's The
Order afThings this idea was in turn "represented" by a famous painting, Las
Meninas ofVe1<izquez. Furthermore, Foucault also spoke of "exchanging" as
well as classifying as characteristic of the classic period. ls
Foucault thus opened the path followed by many essays in this collection.
Foucault's articulation of a concern with the idea of "representation," and its
manifestations in epistemology, science, and art is reflected in the general
framing of the present collection; it resonates most clearly in essays like
Claudia Swan's, where epistemological concerns for classification are linked
to forms of artistic representation, and associated with the origins of a genre
of painting, the independent still life. Likewise, Foucault's association of
epistemological concepts of value with economic ones in the classic period is
reflected in the coupling of commerce with representation in art and science
in numerous other essays in this collection.
As the editors of this volume reiterate in their introduction, many schol-
ars in the twentieth century were certainly concerned with the way in which
art aids science, and science art, in the representation of the world. Richard
Rorty's critique of the philosophical mirror of nature demonstrates that
philosophers other than Foucault have also questioned the notion of repre-
sentation of truth as an adequate or accurate conception for epistemology.19
In addition to an increasing literature on the interactions of art and science,
there has arisen an interest in the interrogation of the concept of "represen-
tation," its many connotations and manifestations, as well as its social con-

Thomas DaCosta Kaufmann


struction and determination. The birth and early success of the periodical
Representations occurred in this atmosphere, as did the work of authors who
were initially associated with it. 20 Moreover, and independent of these ten-
dencies, other discourses on the human sciences have taken newer directions
into the investigation of the often fraught interactions of artistic and scien-
tific representation. 21
In an essay published a decade ago in a volume titled Art in History, His-
tory in Art, David Freedberg picked up many of these strands, and wove
them together in a polemic on the relations of art, science, and commerce. 22
Speaking specifically about seventeenth-century Dutch culture, Freedberg
argued that many of the interconnections that lie between these supposedly
separate realms of activity had been ignored. He challenged scholars to
address issues that seem to be located in such interstices between related but
apparently different disciplines.
This collection replies to Freedberg's challenge, bringing together a vari-
ety of essays that link art, science, commerce, and representation. Moreover,
even if it must be said that fields are not always treated in the same piece, the
ways in which connections have been made also correspond to fresher socioe-
conomic approaches. The stress on commerce, on instruments used, and on
objects studied means that practices, not theories, activities, nor attitudes, are
emphasized in the historiography of science. Networks, exchanges, and
extraneous objects are here the topics for history of science, as for one of art,
rather than scientists, systems, or for that matter masterpieces.
In addition to merchants, mechanics-meaning those who work with
their hands or have a trade, not the classic physical science called mechanics,
the laws of which were adduced by Galileo and Isaac Newton-might thus
seem to have taken center stage. In keeping with other recent syntheses and
collections of studies, this move indicates a shift in approach away from a
concern with the history of physics and astronomy, which had been a main
focus for a traditional history of science, to one in which natural history and
other subjects such as cartography or geography come to the fore as ways of
mapping nature. 23 Similarly, even in paintings with religious subjects, not
figures, but elements pertaining to what were to emerge as new, but were
regarded as lower genres -landscape, animal painting, and still life-
engage art history.
At the same time that these essays point to more paths for investigation,
they also provoke further consideration of some of the theses which they
advance. Engineers and instrument makers surface at several places in this
book, as in James Bennett's essay. This corresponds to a trend in which engi-
neers (and to a lesser degree instrument makers) have increasingly been seen
not only as ancillary to the development of the natural sciences, but to epito-
mize the connection between art and science in the Renaissance. As ins tan-

Questions of Representation
tiated by figures from Filippo Brunelleschi or Leon Battista Alberti to
Leonardo da Vinci, skills in engineering and instrument making were
closely associated with art and architecture in the Italian Renaissance. 24 The
paradigmatic "artist-scientist-genius"25 so personified by Panofsky half a
century ago has yielded to the artist-scientist-engineer engaged in designing
fortifications, making waterworks, instruments, and works of art.
Such figures were to be found over a much longer period of time and in
many more places than Renaissance Florence. Although not so famous, the
position of instrument maker-scientists such as Jobst Burgi between art and
science has in fact gained some attention. 26 But there are many other figures
such as Simon Stevin or Salomon de Caus who could claim a place equally
in histories of seventeenth-century art and architecture, as they have done in
those of mechanics or mathematics. While the previous segregation of
national and disciplinary historiographies seems to have led to their com-
parative neglect, De Caus has recently been recognized as a perspective the-
orist, and as designer of "mechanical contrivances," and Stevin's work has
been called "deeply characteristic of Dutch activities -scientific and practi-
cal-at this time (i.e., the early seventeenth century)."27 De Caus was also
directly involved not only with hydraulic theory, but with garden (and hence
garden fountain-sculpture design); Stevin also was much interested in archi-
tecture, on which he penned a treatise.28
With figures such as De Caus, Stevin, and their older Netherlandish con-
temporary Hans Vredeman de Vries, it is hard to separate the scientist or
engineer from the artist or architect. Vredeman de Vries also made water-
works, served as master of fortifications (Festungsbaumeister) in both
Wolfenbuttel and Gdansk (Danzig), but he is equally well known, if not
more famous in art and architectural history as the author of important trea-
tises on perspective and on the architectural orders. Vredeman de Vries was
also a prolific draftsman and painter. 29 Hence the story that Berkel tells here
about the Netherlandish practitioner calls up a pattern established by the
efforts and interests of other, more important figures; indeed, famous figures
such as Andreas Schliiter remained active in the intersections of art, science,
and engineering into the eighteenth century.30
The lengthy survival or recurrence of such supposedly distinctive Renais-
sance phenomena raises considerations of periodization, and thus leads to
other questions about the conceptualization of this collection. The definition
of commerce posited here is a broader one, that encompasses various kinds
of exchange, not only of trade, for example, of trade, but also of gift-giving.
Economic practices are thus described as having a longer purchase, as it
were. Consequently, Weberian notions that economic rationalism could be
linked with capitalism and with the Reformation (or Renaissance), and thus
connected with definitions of historical periods seems to be rejected. In a

Thomas DaCosta Kaufmann


world in which a prince could offer to trade a colony for a collection, or
courts could bankrupt themselves with expenditures on luxuries in a desire
to express their magnificence, it may well be that period distinctions based
on notions of economic rationality are not the best markers of modernity.31
Instead, representational practices provide the basis for the period defini-
tions assumed here. Pamela Long argues most directly for these sorts of dis-
tinctions in her essay on the relation of visual representation to the
investigation of nature. She contrasts what she finds to be the use of forms of
visual representation to legitimate knowledge claims in the period 1490- 1540S
with what she describes as the antecedent, ancient, and medieval reluctance
"to use visual images to demonstrate claims" about the natural world.
But what are the implications of this thesis? Clearly many examples exist
of images from antiquity and the Middle Ages that were used to illustrate
scientific texts, to demonstrate an argument. As exemplified by herbals and
later by tacuina sanitatis, numerous illuminated manuscripts pertaining to
medicine and natural history survive from antiquity, and in larger numbers
through the Middle Ages. 32 Moreover, other kinds of "scientific" texts,
notably astrological and alchemical handbooks, were also illustrated during
the Middle Ages, and there is evidence that some such sorts of works may
have been in antiquity as wel1. 33 These are, furthermore, some of the very
fields - natural history and alchemy - that form the focus for discussion in
this collection.
Accordingly, the period and category distinctions employed here cannot
be seen to rely simply on the presence of empirical evidence. Rather, it is the
quality of the evidence concerned that is important. And this leads back, as
indeed is suggested in part by Long's essay, to another look at the develop-
ment of representational forms and skills. Yet such an investigation also
leads further into a more extended history of artistic forms and practices,
into questions of artistic endeavor of an art historical nature-one which,
however, is not directly addressed in this book. 34 The thesis that there was a
reluctance to rely on visual evidence for truth claims is in fact based on the
evidence of theory, not practice, and the theoretical considerations seem to
derive, as Long indicates, from discussions of mechanics. This raises a fun-
damental question: To what extent did a mathematical world view offer a
different perspective on visual representation of nature?
Mechanics and celestial mechanics, or astronomy, are also not treated in
this book. But consideration of these subjects is of importance not just for
their implications for questions of periodization, but also for more general
issues concerning commerce and representation that are central to this collec-
tion. Astronomy had immense commercial importance in the period here
discussed, certainly at least in its astrological form. 35 Almanacs, ephemerides,
and horoscopes circulated widely and seem to have been avidly acquired.

Questions of Representation
New astronomical phenomena were eagerly reported and became the "hot
news" of the day.36 Astronomers were highly sought after and regarded by
courts, whom they served in various capacities, not just for casting horo-
scopes. Significantly, Tycho Brahe received many privileges from the Danish
crown; he then became the highest paid servitor of the imperial court. 37
Moreover, attention to mechanics, especially to celestial mechanics, sheds a
different light on issues of representation in art and science. Whether or not
one agrees with an older thesis that regards the "Scientific Revolution" as
connected with the mechanization of the world picture,38 developments in
mechanics were certainly related to advances in mathematics. These relied on
other forms of representation, utilizing quantification and systems for indi-
cating it. 39 For practitioners of mathematically based sciences such as astron-
omy, nature was to be read not in images, but in numbers. Ultimately the
increasing sophistication and difficulties involved with mathematics, related
to the development of professional specialization,40 in fact split the union that
had existed between art, science, and engineering. But such developments
were not of course imaginable until the end of the early modern period.
Not all natural philosophers gave images the same meaning. Differences
in opinion about the significance of visual imagery were in fact already evi-
dent in the early seventeenth century, in the "Kepler-Fludd controversy."
While Fludd believed that the cosmos could be represented in images and
diagrams, Kepler treated these images as at best symbolic, and denied that
they, or other similar images, could represent reality (even though he himself
was a master of the scientific diagram and had famously made a model of the
universe based on the inscription of the Platonic [Pythagorean] solids. Math-
ematics, not pictures, was to be trusted in the end. 41 Considerations of cos-
mology had thus already led to what can be called a crisis of confidence in the
image, at least in the realm of astronomy after Copernicus, at the very time in
which the idea of a "true" image of nature could thrive in such fields as
anatomy and natural history. While science and art could continue to coexist,
this position may be regarded as the initiation of a theme, advanced by the
Enlightenment, by which images would come to be regarded as deceptive. 42
The visual revolution of our own time reminds us, however, that there
are many views of the importance of image making. There are still many
diverse opinions about systems of representation and their relation to nature.
While new forms of biology and biotechnology have revolutionized concep-
tions of life itself, new kinds of visual imaging have proliferated, and those
also make claims to represent reality and call for a new understanding of
visuality related to newer discourses in the humanities and social sciences.43
But the issues and apparent contradictions suggested by the early modern
period are also still with us. The physical sciences still strive to produce a
unified theory, and biological scientists also claim that all science, indeed all

Thomas DaCosta Kaufmann


knowledge, may be explained by a single unified theory.44 Moreover, many
of these questions, concerning computers, biotechnology, physical fusion,
have vast commercial implications. A virtue of this collection of essays is that
it not only stimulates reconsiderations of the historiography of worldviews,
but that in so doing it encourages reconsideration of the role of representa-
tion, commerce, image-making, and science in the contemporary world.

1. The remarks made here pertain to more general issues raised by the book, and there-

fore refrain from commenting on more particular arguments in individual essays.


2. This distinction between an "internal" historiography of science and its alternatives is
made by Thomas DaCosta Kaufmann, "Empiricism and Community in Early Modern Sci-
ence and Art: Some Comments on Baths, Plants, and Courts," in Natural Particulars. Nature
and the Disciplines in Renaissance Europe, ed. Anthony Grafton and Nancy Siraisi (Cambridge,
Mass. and London: MIT Press, 1999, 401f. For the historiography of the Scientific Revolution
see most comprehensively H. Floris Cohen, The Scientific Revolution. A Historiographical
Inquiry, (Chicago and London: University of Chicago Press, 1994).
3. A. C. Crombie, From Augustine to Galileo (Harmondsworth: Penguin Books, 1959; 1st
ed., 1952).
4. Edwin A. Burtt, Metaphysical Foundations of Modern Physical Science, (New York: Har-
court Brace, 1932, rev. ed.; 1st ed. 1924).
5. The works of all these scholars are now largely accessible in translation. See, for exam-
ple, Heinrich Wolfflin, Principles of Art History. The Problem of the Development of Style in
Later Art, trans. E. Hottinger (New York: Dover, 1950; 1st English ed., 1932); Classic Art,
trans. Peter and Linda Murray (London: The Phaidon Press, 1952); Renaissance and Baroque,
trans. Kathrin Simon, intro. Peter Murray (Ithaca, N.Y.: Cornell University Press, 1964);
Alois Riegl, Problems of Style. Foundations for a History of Ornament, trans. Evelyn Kain,
annotations and intro. David Castriota, preface Henri Zerner, (Princeton, N.J.: Princeton
University Press, 1992); The Group Portraiture of Holland, trans. Evelyn M. Kain and David
Britt, intro. Wolfgang Kemp (Los Angeles: Getty Research Institute for History of Art and
the Humanities, 1999); Emile Male, The Gothic Image, trans. (New York, 1952). They have
also garnered significant historiographical attention. [See, for example, For Male.] Germain
Bazin, Histoire de l'histoire de l'art de Vasari tl nos jours (Paris: Albin Michel, 1986), 208-2IO (in
a chapter titled "Les pouvoirs de l'image: l'iconographie").
6. See, for example, Richard S. Westfall, The Construction of Modern Science. Mechanisms
and Mechanics (Cambridge: Cambridge University Press, 1971); Sydney J. Freedberg, Painting
in Italy 1500-/600 (The Pelican History of Art) (Harmondsworth: Penguin Books, 1971).
7. Warburg's essays, including many papers touching on astrology, are now accessible in
an English translation: Aby Warburg, The Renewal of Pagan Antiquity: Contributions to the
Cultural History of the European Renaissance, trans. David Britt, intro. Kurt W. Forster (Los
Angeles: Getty Research Institute for History of Art and the Humanities, 1999). See further
Fritz Saxl, Lectures (London: Warburg Institute, 1957). Panofsky's early German essays on
proportion and perspective are also available in translation: "The History of the Theory of
Human Proportions as a Reflection of the History of Style," in Meaning in the Visual Arts
(Garden City, N.Y.: Doubleday Anchor Books, 1955), 55-I07; Perspective as Symbolic Form,
trans. Christopher Wood (Cambridge, Mass.: Zone Books [MIT Press], 1991).

Questions of Representation
8. Martin Wackernagel, The World of the FlorentIne Renaissance Artist. Projects and Patrons,
Workshop and Art Market, trans. Alison Luchs (Princeton, N.J.: Princeton University Press,
1981; 1st ed. Leipzig, 1938); Hans Floerke, Studien zur niederliindischen Kunst - und Kul-
turgeschichte. Die Formen des Kunsthandels, das Atelier und die Sammler in den Niederlanden
vom 15.-18. Jahrhundert (Munich and Leipzig: Georg Muller, 1905)'
9. Robert K. Merton, Science, Technology and Society in Seventeenth-Century England (New
York: Harper & Row, 1970; 1st ed. 1938); Herbert Butterfield, The Origins of Modern Science
(New York: Macmillan, 1957; 1st ed. London, 1949). For the impact of the latter work, see
Robert S. Westman and David C. Lindberg introduction to Reappraisals of the Scientific Rev-
olution, ed. Westman and Lindberg, (Cambridge, Cambridge University Press, 1990), xvii.
ro. For example, Frederick Antal, Florentine Painting and its Social Background. The Bourgeois
Republic before Cosimo de Medici's Advent to Power, (London: Kegan Paul, 1947); Arnold Hauser,
Social History of Art, trans. Stanley Godman (New York: Random House, 1951),2 vols.; J. D.
Bernal, Science in History, (Cambridge, Mass.: MIT Press, 1971),4 vol. (1st ed. London, 1954).
11. I have dated the appearance of the "new art history," at least in its self-proclamation,
to 1972 in a response to a "Visual Culture Questionnaire," October 77,1996: 45-8. See further
"What Is New about the 'New Art History,'" The Philosophy of the Visual Arts, ed. Philip
Alperson (New York and Oxford: Oxford University Press, 1992),515-520.
12. In addition to the titles cited by Smith and Findlen, see, for example, Richard A.
Goldthwaite, The Building of Renaissance Florence. An Economic and Social History (Baltimore
and London: Johns Hopkins University Press, 1980); Neil de Marchi and Hans J. Van
Migroet, "Novelty and Fashion Circuits in the Mid-Seventeenth Century Antwerp-Paris Art
Trade," Journal for Medieval and Early Modern Studies 28 (1998) 201-246; idem, "Exploring
markets for Netherlandish Paintings in Spain and Nueva Espana," Nederlands Kunsthistorisch
Jaarboek (2000): 80- I I I; Bernd Roeck, Kunstpatronage in der Friihen Neuzeit (Gottingen: Van-
denhoeck & Ruprecht, 1999)'
13. Alan G. R. Smith, Science and Society in the Sixteenth and Seventeenth Centuries (Lon-
don and New York: Science History Publications, 1972). A new edition of Merton's Science,
Technology and Society was also published in 1970.
14- Among the volume of recent literature, an exemplary collection is Patronage and Insti-
tutions. Science, Technology, and Medicine at the European Court 1500- 1700, ed. Bruce T. Moran
(Rochester, N.Y., and Woodbridge: Boydell Press, 1991), including essays by Smith and Find-
len. See also their other publications, and studies by other scholars cited in essays above.
I s. Mario Biagoli, Galileo Courtier. The Practice of Science in the Culture ofAbsolutism (Chicago

and London: University of Chicago Press, 1993); John Robert Christianson, On Tycho's Island.
Tycho Brahe and His Assistants 1570--1601, (Cambridge: Cambridge University Press, 2000).
16. Steven Shapin, The Social History of Truth. Civility and Science in Seventeenth-Century
England (Chicago and London: University of Chicago Press, 1994),36-38, citing Foucault.
See further Shapin's more recent synthesis The Scientific Revolution (Chicago and London:
University of Chicago, 1996), emphasizing the social practices of science.
17. See Michel Foucault, L'archeologie du sa voir (Paris: Gallimard, 1969)'
18. Michel Foucault, The Order of Things. An Archaeology of the Human Sciences (New
York: Pantheon Books, 1970), English trans. of Les Mots et les choses (paris: Gallimard, 1966).
19. Richard Rorty, Philosophy and the Mirror of Nature (Princeton, N.J.: Princeton Univer-
sity Press, 1979).
20. Notably Stephen Greenblatt, Renaissance Self Fashioning From More to Shakespeare
(Chicago and London: University of Chicago Press, 1980); Svetlana Alpers, The Art ofDescribing.
Dutch Art in the Seventeenth Century (Chicago and London: University of Chicago Press, 1983)'
21. Exemplified by Barbara Maria Stafford, Body Criticism. Imaging the Unseen in Enlight-
enment Art and Medicine (Cambridge, Mass. and London: MIT Press, 1991); Artful Science.

Thomas DaCosta Kaufmann


Enlightenment Entertainment and the Eclipse of Visual Education (Cambridge, Mass. and Lon-
don: MIT Press, 1994); Good Looking. Essays on the Virtue of Images (Cambridge, Mass. and
London: MIT Press, 1996).
22. "Science, Commerce, and Art: Neglected Topics at the Junction of History and Art His-
tory," in Art in History. History in Art. Studies in Seventeenth-Century Dutch Culture, ed. David
Freedberg and Jan de Vries (Santa Monica: Getty Center for the History of Art and Humani-
ties, 1991),376- 428. Freedberg's essay has been excerpted, "Ciencia, Comercio e Arte. Topicos
negligenciados na junc;ao da hist6ri a[historia] com a hist6ria da arte (excertos)," in 0 Brasil e os
Holandeses 1530-1654, ed. Paulo Herkenhoff (Rio de Janeiro: Sextante Artes, 1999), 192-217.
23. Grafton and Siraisi, Natural Particulars. Nature and the Disciplines in Renaissance Europe.
24- Bertrand Gille, Les ingenieurs de la Renaissance (Paris: Hermann, 1964), is an older
introduction to the topic. See also for an overview Paolo Galluzzi, Renaissance Engineers from
Brunelleschi to Leonardo da Vinci (Florence: Istituto e Museo di Storia della scienza, 1996) and
Helene Venn, La gloire des ingenieurs (Paris: Albin Michel, 1993).
25. See Erwin Panofsky, "Artist, Scientist, Genius: Notes on the 'Renaissance-Dammerung,'"
in The Renaissance. Six Essays, ed. Wallace K. Ferguson (New York: Harper & Row, 1962), 121 -182
(initially given in a lecture series at the Metropolitan Museum of Art, New York, 1951-52).
26. Because of his importance for the history of mathematics, horology, and astronomy,
Biirgi, who crafted exquisite objects that became elements of princely collections, may be
picked out as exemplifying this tendency. For literature on Biirgi, see Ludolf von Mackensen,
ed., Die erste Sternwarte Europas mit ihren Instrumenten und Uhren. 400 Jahren Jost Biirgi in
Kassel (Munich: Call way, 1988; 3rd improved ed.; 1st ed. 1979), and for a more recent sum-
mary idem, "Die Kasseler Wissenchaftskammer oder die Vermessung des Himmels, der Erde
und der Zeit," in Moritz der Gelehrte. Ein Renaissancefiirst in Europa, ex. cat. ed . Heiner
Borggrefe, Eurasburg, 1997,385-390, with catalog entries 391ff.
27. Martin Kemp, The Science of Art. Optical Themes in Western Art from Brunelleschi to
Seurat (New Haven and London: Yale University Press, 1990), II2-I18; Kemp comments:
"Stevin and his fellow mathematicians were concerned with an astonishing range of applied
skills and technologies-fortifications, guns, ships, canals, navigation windmills, cranes
modes of transport, timepieces, surveying, accounting, banking, and certainly not least, the
optical instruments such as the telescope and microscope that were been exploited to revolu-
tionize the visual data of science"(1 14).
28. With characteristic insight, Frances Yates recognized De Caus's multiple talents and
importance. See The Rosicrucian Enlightenment (London and Boston: Routledge and Kegan
Paul, 1972), I1-13 passim. For De CallS, see Piet Lombaerde, "Pietro Sardi, Georg Miiller,
Salomon de Caus und die Wasserkiinste des Coudenberg-Gartens in Briissel," Gartenkunst 3,
1991: 159-171; idem, "Die Wasserkiinste des Coudenbergparks in Briissel," in Die Wasserver-
sorgung in der Renaissancezeit, ed. Albrecht Hoffmann (Mainz: Von Zuben, 2000), 277- 84. An
edition with commentary of Stevin's architectural treatise is being prepared by Charles van
den Heuvel; in the meantime see idem, "Stevins 'huisbouw' en het onvoltooide Nederlands
architectuurtractaat. De praktijk van hat bouwen als wetenschap," Bulletin van de Konijklijke
Nederlandse Oudheidkundige Bond 93 (1994): 1-18.
29. Vredeman de Vries is the subject of a forthcoming exhibition in Lemgo which will
illuminate many of his qualities. An essay by Lombarde and Van den Heuvel promises fur-
ther to relate his many-sided talents to those of De Caus and Stevin. See Petra Sophia Zimer-
mann, "Die Palastenentwiirfe des Hans Vredeman de Vries in der 'Architectura' von 1577,"
in Italienische Renaissancebaukunst an Scheide, Maas und Niederrhein : Stadtanlagen-Zivil-
bauten-Wehranlangen: Tagungshandbuch (2. Jiilicher Pasqua lin i-symposium vom 18. bis 21. Juni
1998 in Jiilich), ed. Gunter Bers and Conrad Doose, Julich, 1999; 335-337 (summary on
62 1- 622), with references to earlier studies.

Questions of Representation 421


30. For example, Andreas Schli.iter, for whom in this regard see Thomas D aCosta Kauf-
mann, "Schli.iter's Fate. Comments on Sculpture, Science, and Patronage in Central and East-
ern Europe c. 1700," in Kunstlerische Austauschl Artistic Exchange. Akten des XXVIII.
Internationalen Kongresses fur Kunstgeschichte Berlin, 15.-20. Juli 1992, Berlin, Akademie Ver-
lag, 1993, I99- 2I2.
3 I. The first story is told by Pamela H. Smith, The Business ofAlchemy. Science and Culture
in the Holy Roman Empire (Princeton, N.J.: Princeton University Press, I994), 171-192. The
overwhelming impact of expenses on luxuries and magnificence is familiarly given as a rea-
son for the bankruptcy of the French crown in the eighteenth century, and the same occurred
elsewhere, as for example in Bavaria.
32. See, for a general introduction, Wilfrid Blunt, with the assistance of William T. Stearn,
The Art of Botanical Illustration. An Illustrated History (London: Collins, 1950), 5ff. For an
introduction to tacuina sanitatis and their relation to the herbal, see Luisa Cogliati Arano, The
Medieval Health Handbook (New York and London: Barrie & Jenkins, 1976). For a recent
publication related to an ancient example, see Cesare Ruffato, ed., La medicina in Roma antica.
n Liber medicinalis di Quinto Sereno Sammonico (Turin: UTET, I996).
33. For an illustrated overview of alchemical manuscripts and books, see Gareth Roberts,
The Mirror of Alchemy. Alchemical Ideas and Images in Manuscripts and Books. From Antiquity
to the Seventeenth Century (London: British Library, I994). Astrological handbooks were dis-
cussed long ago by Warburg, Saxl, and their followers; see note 7 above. For a general
overview, see Kurt Weitzmann, Ancient Book Illumination (Cambridge, Mass.: Harvard Uni-
versity Press, 1959), not without questions, however.
34. For a very recent approach to the question of the origin of still life in Italy which, tak-
ing into account other literature, discusses historical and art historical questions, see Giacomo
Berra, "Arcimboldi, Vincenzo Campi, Figino, Fede Galizia, Caravaggio: congiunture sulla
nascita della natura morta in Lombardia," in Vincenzo Campi: scene del quotidiano, ex. cat., ed.
Franco Paliaga, Milan and Cremona, 2000, 6r-86. I deal with this in a forthcoming essay.
35. For a discussion of some aspects of the commercial relations of astrology see most recently
Anthony Grafton, Cardano's Cosmos (Cambridge, Mass.: Harvard University Press, r999).
36. This is the subject of ongoing research by Eileen Reeves.
37. In addition to the work by Christianson, On Tycho's Island, see for Brahe's later career
Victor E. Thoren, The Lord of Uraniborg. A Biography of Tycho Brahe (with contributions by
Christianson) (Cambridge: Cambridge University Press, 1990), 376ff. For the point about
Tycho's salary see Alphons Lhotsky, "Die Geschichte der Sammlungen," Festschrift des Kunst-
historischen Museums in Wien 1891-1941, pt. 2, Vol. I, P.295.
38. Cf. E. 1. Dijksterhuis, The Mechanization of the World Picture (Oxford: Clarendon
Press, I961; 1st ed. Amsterdam, I950).
39. See Alfred W. Crosby, The Measure of Reality. Quantification and Western Society
1250-1600 (Cambridge: Cambridge University Press, 1997).
40. See J. V. Field, The Invention of Infinity. Mathematics and Art in the Renaissance (Oxford,
New York, Tokyo: Oxford University Press, I997), 229ff.
41. See for these points the insightful essay by Robert Westman, "Nature, Art, and Psyche:
Jung, Pauli, and the Kepler-Fludd Polemic," in Occult and Scientifc Mentalities in the Renais-
sance, ed. Brian Vickers (Cambridge: Cambridge University Press, I984), I77-230.
42. This subject has been well elucidated by Barbara Stafford, Body Criticism, and Artful
Science.
43. See Barbara Stafford, Good Looking.
44. See Edward Osborne Wilson, Consilience. The Unity of Knowledge (New York: Knopf,
1997)·

422 Thomas DaCosta Kaufmann


Contributors

Pamela H. Smith is Edwin F. and Margaret H. Hahn Professor of History


at Pomona College and the Claremont Graduate University in Claremont,
California. She is the author of The Business ofAlchemy: Science and Culture
in the Holy Roman Empire (1994), which won the Pfizer Prize in the History
of Science in 1995, and is now completing a book on artisanal attitudes to
nature in early modern Europe.

Paula Findlen is Professor of History and Director of the Science, Technol-


ogy and Society Program at Stanford University. She is the author of Pos-
sessing Nature: Museums, Collecting, and Scientific Culture in Early Modern
Italy (1994), which won the 1995 Marraro Prize in Italian History and the
1996 Pfizer Prize in History of Science, and A Fragmentary Past: The Italian
Renaissance Origins of the Museum (forthcoming). She is currently complet-
ing a book on gender and knowledge in eighteenth-century Italy.

Antonio Barrera is assistant professor of History at Colgate University. His


areas of interest are history of science and Atlantic world history. His cur-
rent research concerns the interactions between the Atlantic world and the
emergence of modern science, with a particular emphasis on Spain and
America in the sixteenth and seventeenth centuries.

James A. Bennett is Director of the Museum of the History of Science, Uni-


versity of Oxford. His research interests focus on instruments, astronomy,
practical mathematics, and museums.

Klaas van Berkel earned his Ph.D. at Utrecht University with a dissertation,
Isaac Beeckman (1588-1637) and the mechanization of the world picture
(1983). Since 1988, he is professor of Modern History at the University of

423
Groningen. With Albert van Helden and Lodewijk Palm he edited A His-
tory of Science in the Netherlands. Survey, Themes and Reference (1999)·

Harold J. Cook is currently Director of the Well come Trust Centre for the
History of Medicine at University College London. He has authored two
books and several articles on the place of early modern medicine and natural
history in the scientific revolution, and continues his investigations into
those areas of inquiry during the Dutch Golden Age.

Anne Goldgar teaches early modern European history at King's College


London. She is the author of Impolite Learning: Conduct and Community in
the Republic of Letters, 1680-1750 (1995) and articles on seventeenth- and
eighteenth-century cultural history. She is writing a book about the social
and cultural context of tulip mania.

Deborah E. Harkness is an associate professor of History at the University of


California at Davis. Her previous work on John Dee has received prizes
from the Renaissance Society of America and the History of Science Society,
and culminated in a book, John Dee's Conversations with Angels: Cabala,
Alchemy and the End of Nature (2000). Presently she is working on Neighbor-
hoods of Science: Knowledge and Practice in Elizabethan London under the
auspices of the National Science Foundation.

Thomas DaCosta Kaufmann is a professor in the Department of Art and


Archaeology, Princeton University. Among his books are The Mastery of
Nature: Aspects of Art, Science, and Humanism in the Renaissance (1993) and
Court, Cloister and City. The Art and Culture of Central Europe1450-1800
(1995; German revised edition, 1998).

Pamela O. Long is a historian who has taught at Barnard College, St. Mary's
College of Maryland, and Johns Hopkins University. She has published
extensively on late medieval and Renaissance mechanical arts and is the
author of Openness, Secrecy, Authorship: Technical Arts and the Culture of
Knowledge from Antiquity to the Renaissance (2001).

Mark A. Meadow is Assistant Professor of the History of Art at University


of California, Santa Barbara, and Co-Director (with Bruce Robertson) of
Microcosms: Objects of Knowledge, a Special Humanities Project of the
University of California Office of the President. He has published on
Pieter Bruegel, Pieter Aertsen, Albrecht Durer; on the history of collect-
ing, the history of rhetoric, and the history of memory; on proverbs and
proverb collecting, and on Renaissance epistemology. His first book is

Contributors
Pieter Bruegel the Elder's Netherlandish Proverbs and the Practice of
Rhetoric (2001) .

Chandra Mukerji is Professor of Communication and Sociology at the Uni-


versity of California, San Diego. She has written widely on the sociology of
culture and communication, and the history and sociology of science and
technology. Her book publications include Territorial Ambitions and the Gar-
dens of Versailles (1997), From Graven Images: Patterns of Modern Materialism
(1983), A Fragile Power (1990), for which she was awarded the Robert K.
Merton award, and, co-edited with Michael Schudson, Rethinking Popular
Culture (1991).

Tara E. Nummedal has recently completed her dissertation, entitled


"Adepts and Artisans: Alchemical Practice in the Holy Roman Empire,
1550-1620," at Stanford University. Her research focuses on the careers of
working alchemists and their connections to broader scholarly, political, and
commercial pursuits in early modern Europe.

Lissa Roberts teaches the history of science and technology at the University
of Twente, the Netherlands. While more broadly interested in the cultural
history of science and technology in Europe during the eighteenth century,
she is currently writing a book that traces the cultural history of the intro-
duction of the steam engine into the Netherlands.

Alison Sandman recently completed her dissertation, entitled "Cosmogra-


phers vs. Pilots: Navigation, Cosmography, and the State in Early Modern
Spain," at the University of Wisconsin, Madison. Her research uses disputes
about the practice of navigation to probe the construction of the idea of the
utility of science and examines reasons for the growth of state patronage of
science in the early phases of the Scientific Revolution.

Benjamin Schmidt is Assistant Professor of History at the University of


Washington. His publications include Innocence Abroad: The Dutch Imagina-
tion and the New World (2001) and numerous articles on early modern cul-
tural history. He has held recent fellowships from the NEH and Ahmanson
and Getty foundations. His current research focuses on travel, expansion,
and exoticism in Baroque Europe.

Larry Silver is Farquhar Professor of Art History at the University of Penn-


sylvania. He taught previously at Berkeley and Northwestern and is a special-
ist in early modern prints and paintings of Germany and the Low Countries.
He has served as President of both the Historians of Netherlandish Art and

Contributors
the College Art Association and is also the author of Art in History (I993), a
general survey.

Claudia Swan is Assistant Professor of Art History at Northwestern Uni-


versity. She is completing a book titled Art, Science, Witchcraft; Jacques de
Gheyn II and the Representation of the Natural World in the Netherlands ca.
1600; her Clutius Botanical Watercolors (a compendium of sixteenth-century
scientific watercolors employed in the medical curriculum at Leiden Uni-
versity) appeared in I998.

Contributors
Index

Accademia Fisicomatematica Romana, 278, tinctures, 207


285-290,405 uses of, 205-212
Achromatic lenses, 381 values of, 216-217
Ackerman, James, 13 Alchimia, 202
Acontius, James, 150 Aldemare, Caesar, 137
Ad vivum illustrations, 3 Aldrovandi, Ulisse, 8, 14, 109-111,300,303-
Adams, George, 379, 382, 384 314
Alaertes, Guillaume, 144 Alpers, Svetlana, 13, 19,404-405
Alaertes, Lieven, 144 Altdorfer, Albrecht, 36-37, 41, 44, 46, 52
Alberti, Leon Battista, 8,65,416 The Amboinese Curiosity Cabinet
Albinus, Bernard Sigfried, 243 (Rumphius),347
Albrecht V (Duke of Bavaria), 182-183, 185, The Amboinese Herbal (Rumphius), 347
187, 189, 191-194 Antal, Frederick, 413
Alchemy Apologia (Meijer), 290-292
commerce and, 201-202, 211-212 Apothecaries, 17, 144,305-307
as commodity, 202-205 in Elizabethan London, 144-145
critiques of, 212-216 Appadurai, Arjun, 5
in Elizabethan London, 151-155 Architectural styles, 74-75
Frobisher's gold, 152 Aretino, Pietro, 74
generation by putrefaction, 51 Aristotle, 47, 63, 402
in Holy Roman Empire, 201-217 Arnold, John, 382
"how-to-books" for, 202 Arnold, Ken, 184
laboratories for, 207-208 Art; see also Diirer, Albrecht
manual work and, 41 consumption and, 18-19
market for knowledge of, 203-204 in Hypnerotomachia Poliphili, 65-67, 74
medicine and, 206-207 "internalist" tradition, 413
mining and, 206 market for, 15-18
as natural philosophy, 201 natural knowledge and, 404
patrons of, 209-211 nature and, 29, 65, 72, 326, 402
practical skills of, 205-209 science and, 13-14, 17-19
profit motives of, 204 Artificialia, 124, 182, 325, 330, 333, 402,
smelting and, 210 408
as spiritual art, 214-215 "Artist-scientist-genius," 416
sulfur/mercury dual principles, 42-44 Artistic representation, 112-113

427
Artists-artisans, 13-14 Brahe, Tycho,414,418
Ascham, Roger, 190 Brouncker, William, 371
Astronomy Brueghel, Jan, 31
instruments for, 370-371, 377 Brunelleschi, Filippo, 416
sea charts and, 85, 95-96, 101 Brunner, Otto, 4
Atkinson, Christopher, 144 Buffon, Comte de, 110
Atlases, 352 Bugge, Thomas, 381-382, 391-392
Ayahibex, Catalina de, 166-167, 169 Burgi, Jobst, 416
Burtt, E. A., 413
Bacon, Francis, 16, 150- 151 Butterfield, Herbert, 413
Bacon, Roger, 228 Butterfield, Michael, 377
Baconians
commercial interests, 404 Cabot, Sebastian, 85-87, 89-92, 95, 97, 100
Dutch painting and, 405 Cal vino, Italo, 185
Baker, George, 144, 146 Calzolari, Francesco, 9, 305-306, 315-316
Balsam; see also Santa Domingo balsam Campin, Robert, 8
in classical medicine, 165 Canal du Midi
Matarea balsam, 166 as a commercial/entrepreneurial enter-
Barlow, Edward, 144 prise, 249- 255, 269
Barnes, John, 150 financing arrangements, 251
Barometers, 374 geographical measurement and, 257
Barreda, Licenciado, 170, 174 labor force/social relations for, 253-255
Barrera, Antonio, 163 land survey methods, 255-259
Basilisks, 307- 318 political maneuvering for, 252
Baxandall, Michael, 15, 19 port/terminus, 263-269
Baxter, William, 144 risk assessment, 259-269
Baylie, Walter, 144 water system of, 260-263
Becher, Johann Joachim, 211-212 Capitalism, science and, 15-16
Beeckman, Andries, 352 Carrara Herbal, 11-12
Belon, Pierre, 310 Cartography, 256-257, 358, 406, 412, 415; see
Benavides, Pedrarias de, 172 also Canal du Midi; Sea charts
Bennett, James A., 370, 415 during reign of Louis XIV, 248-249
Berge, Matthew, 387 in Dutch Republic, 352
Berger, William de, 150 marketing strategy and, 412
Bernal,]. D., 413 military surveys, 257
Bernini, 280, 282 narratives of places, 258-259
Bernoulli, Jean, 370, 379-381 as political tools, 258
Bird, John, 377-378, 380 in Spain (17th c), 83
Blagrave, John, 149 Casa de la Contrataci6n, 84-85,96, 164-165,
Blowfish, Dutch classification system and , 168-170,172,174-175
111- 122 Cassini, J. D., 384-386
Board of Longitude, 380 Castelletti, Bernardo, 303-304
Boerhaave, Herman, 229 Cecil, William, 142, 150-151
Book of Isaac, 203 Cellini, Benvenuto, 50
Botanical portraits, 129 Cennini, Cennino, 40
Botany, 121; see also Gardens; Tulips Ceramic art, 50
classification system for, 127-128 white glaze, 50
Bourne, William, 147, 149 Ceruti, Benedetto, 313
Boyle, Robert, 372 Charles I (King of England), 86
Bragadino, Marco, 207 Charles II (King of England), 372-373

Index
Charles V (Holy Roman Emperor), 86, 166, for European market, 169
190 in New World, 167-168
Chaves, Alonso de, 85-86, 89, 94-95, 101- Cook, Harold J., 223
102 Copernicus, 284, 399-400, 406, 418
Chiocco, Andrea, 313, 315 Cornet, Charles, 141
Chirurgia parva (Lan frank), 144 Cortes, Hernan, 358
Chronicle of Prodigies and Portents (Lycos- Cosmography, 84-85, 350
thenes),308 politicization of, 103
Chronometers, 377 sea charts and, 86-87, 98
Ciampini, Giovanni Giustino, 286-288 territorial disputes and, 100
Clajus, Johannes, 212-213 Council of the Indies (1524), 97-98,101-102,
Classification systems, 10, 109-131 164-165, 175
blowfish, in Dutch natural history, Craftsman, as artist, 278
111-122 Cranach, Lucas, 34-36
dwarf tulip, 120-121 Croll, Oswald, 231
tabulation, 122-131 Crombie, A. C., 413
Clement X (Pope), 280-281 Culpeper, Edmund, 374, 390
Clowes, William, 146 Cummings, Alexander, 382
Clusius, Leidenaar Carolus, 112-122, 128, Curiosity cabinets, 182,297-301,403
175 collections as communal activity, 184
Cluverius, Philip, 350 dried objects for, 226
Cockerman, Philip, 150 Dutch objects, 353-354
Colbert, Jean-Baptiste, 249-259, 264-267 hydralbasilisk fabrications, 307-320
Cole, Humphrey, 148-149 as status symbols, 300
Collections; see Curiosity cabinets; trade objects in, 182,226,297-301
Wunderkammer trafficking in nature, 301-307, 412
Colon, Hernando, 85-86,95, 100 travel and, 184
Colonna, Francesco, 66
Columbus, Christopher, 85 Da Vinci, Leonardo, 14,44,65, 67-71, 250,
Commerce 307-308,416
alchemy and, 211-216 Dade, John, 138
art and, 15, 71-74 D' Albuquerque, Alfonso, 1
Baconians and, 404 Dannenfeldt, Karl, 230
Calvinism and, 15 Dapper, Olfert, 358-360
critiques of new alchemical commercial- Dapperus exoticus curiosus, 358
ism, 212-217 Daston, Lorraine, 18, 184, 325
curiosity cabinets and, 184-185, 193-195 Daumas, Maurice, 377
inventing nature and, 302-303, 311, 319 Davila's expedition to Panama, 170
inventory and, 224-225 De Bils, Louis, 229-230, 232-234, 238,
ivory trade, 183 240-241
market for art and, 15 De Bruyn, Cornel is, 359
natural philosophy and, 17 De CallS, Salomon, 416
nature and, 384-391 De Constantia (Lipsius), 130
patronage and, 3-7 De Gheyn, Jacques, 112- 113, 115, 121-122,
science and, 15-18 128-129
sea charts and, 83 De Historia Stirpium, 8
timellong-term arrangements and, 224 De Hortis Germaniae, 326
Commodities, 4-7; see also Fugger family; De Humani Corporis Fabrica (Vesalius), 8, 65,
Santa Domingo balsam 400
alchemy as, 202-205 De Leon, Ponce, 228

Index
De Mellayne, John, 149 husband-and-wife medical partnerships,
De Revolutionibus (Copernicus), 399-400 144
Dee, John, 137-138, 152 instrument makers in, 148-149
Descartes, Rene, 18,224,228,407 large-scale engineering/mechanical feats,
Diderot, Denis, 110,407 150
Digges, Leonard, 149 mechanical marvelslroyal patents, 147-
Diosc6rides, 171 151
Dodonaeus, Rembertus, 121 medical practitioners, 143-147
Dollond, Peter, 380-381 natural science practitioners/exchange
Don Manuel I, I in, 137-155
Durer, Albrecht, 1-14, 19,39-40,65,71-74, Paracelsian medicine in, 146--147
308 royal patents, 142
alchemy themes in, 41 surgical practices, 145-146
Madonna with the Mayfly, 29-30, 32 technical progress in, 147
Madonna with the Monkey, 31-34 Embalming methods, 230-236; see also
Madonna outdoors, 29-31, 34, 36, 46, 51 Preservation arts
Dutch East India Company (VOC), 225, Entrepreneurs; see also Canal du Midi
348-349,356 in 17th c. France, 248-249
Dutch geography, 347-364 state-based entrepreneurial ism, 249
exoticismlimage of world, 356--360 Ercker, Lazar, 204, 212
marketing of "World," 349, 356--360 Eschinardi, Francesco, 287-288
neutrality of, 360-362 Evans, R. J. W., 209
prank (sumptuousness of), 359- 360 Evelyn, John, 299
Dutch Republic Exoticism, 347, 352, 364, 406
Baconians and painting, 405 colonialism, 349
exoticism/image of world, 356--357 Dutch image of world, 356-360
geography project; see Dutch geography early observers of, 347
Golden Age of, 356
mathematics in, 406 Fabrica (Vesalius), 74-78, 80
natural histories of, 350 Falero, Francisco, 86-89, 100-101
painters of, 353 Fama of the ruler, 4
representation of the world, 362 Farnese, Alessandro, 190
traditional geographies, 350 Ferdinand I (King of Bohemia), 190
travel narratives, 350-352 Fernandez de Oviedo, Gonzalo, 7
wonders and "things-of-the-world," 352, Feudal gift economy, 4
364 Fevre, Nicaise Ie, 372
Findlen, Paula, I, 111,297,401,412-413
Eamon, William, 202 Flight into Egypt (Durer), 31
Eckhout, Albert, 353, 355 Flight into Egypt (Schongauer), 5 I
Economic rationalism, 416 Floerke, Hans, 413
Eichberger, Dagmar, 73 Fontana, Carlo, 280, 282, 284, 290-291
Electricity, 383 Forestry surveyor, 267
Elias, Norbert, 4, 224 Forster, Richard, 138, 144
Elizabethan London Foucault, Michel, 119,403,414
commercial practices/advertising in, 140- Fountain of Life, 44, 46, 52
142 Fountain of youth, 228
English alchemical tradition vs. foreigners, Four Elements (Hoefnagel), 36
151-154 Francis I (King of France), I
foreigners and "strange" ideas, 139, 141 Franeau, Jean, 327
Freedberg, David, 415

Index
Friedrich I (Duke of Wuttemberg), 211 Graham, George, 378, 382, 391
Frobisher, Martin, 152 Gravesande, Willem 's, 406
Froidour, Louis de, 267-268 The Great Piece ofTulf(Durer), 72, 74
Frow, John, 6 Greatorex, Ralph, 372-373
Fuchs, Leonhart, 8,14,110 Greenwich Observatory, 378
Fugger family, 183-184, 186--189 Gresham, Thomas, 137, 152
communication system of, 188 Guilds, 389
"factories" of, 188 Gurlach, Martin, 203
history of, 186-187 Gutierrez, Diego, 86-87, 90-97, 101
international trade and, 188
as patron of arts, 191-192 Hadley, John, 375
Fugger, Hans Jacob, 6,183-186,189-193 Hainhofer, Philipp, 9, 183,300
biography of, 189-193 Hall, John, 144-145
career as librarian, 191 Harkness, Deborah E., 4, 137
education and fellow students of, 189- Harrison, John, 377
190 Harvey, Gabriel, 149
Fugger, Max, 183 Harvey, William, 223
Hauksbee, Francis, 375-376, 379, 389, 417
Gabbey, Alan, 63 Hauser, Arnold, 401, 404, 413
Gale, Thomas, 145 Hearne, Thomas, 148
Galileo, 9-10, 70, 224, 401, 406, 414-415 Herbals/herbalists, 17, 417; see also Santa
Gardens Domingo balsam
as collection sites, 403 Hernandez, Francisco, 317
profit motive and, 408 Herwart, Johann Heinrich, 326
as a "third nature," 403 Hexameron of Bernardus, 203
Garzoni, Tommaso, 299 Hirschenberg, Cristoff, 205
Gelenius, Sigmund, 190 History ofAnimals (Gessner), 305, 310
Gemini, Thomas, 140, 149 History of Selpents and Dragons (Aldro-
Geographia generalis, 350 vandi), 309, 311
Geography, 415; see also Mapmaking; Sea Hlavsa, Petr, 205
charts Hodgson, James, 376
Dutch exoticismlimage of world, 356- Hoefnagel, Joris/Georg, 11,36
360 Holy Family by a Fountain (Altdorfer), 38
Dutch geography, 347-364 Holy Family with Three Hares (Durer), 30
in Dutch Golden Age, 355-356 Holy Roman Empire, alchemy in, 201-217
geographic movement, 355-356 Honauer, Georg, 211
measurement, 257 Hondius, Petrus, 327
of non-European world, 347-350 Honricke, Garrard, 150
representation of the world, 362 Hoogewerff, G. J., 278, 292
George III (King of England), 379 Hooghe, Romeyn de, 355
Gessner, Conrad, 14, 127, 172, 190, 192,305, Hooke, Robert, 19,372,377,389
308- 310,326 Hornius, Georg, 350
Gift economy, 1,4-5,7 Hornsby, Thomas, 378
Gilbert, Samuel, 327, 329 Hortus conclusus, 30-31
Gilpin, George, 150 Hosius, Stanislaus, 190
Glass cutting, 374 Human body, 74, 400- 401
"Glass teardrops," 372 anatomical investigation, 240-241
Global trade, 3, 6 preservation techniques, 240
Goldgar, Anne, 324, 407 sensory organs and, 18
Goldthwaite, Richard, 7, 15 Vesalius' illustration, 77

Index 43 1
visual representation, 77 Leber, Hermann, 72
Hund, Wiguleius, 190 Leclerc, Georges-Louis, 110
Huygens, Christian, 373 Leeuwenhoek, Antoni van, 277-278
Hydras, 307-318 Leiden University Garden, 124
hydra of Hamburg, 318-320 Le Monnier, Pierre Charles, 378, 381
Hypnerotomachia Pollphih 65-67, 74 Leo X (Pope), 1
Life of the Virgin (Diirer), 31
Imperial power, 17 Lindeboom, G. A., 232
Innocent XI (Pope), 281 Linnaeus, Carolus, 110
Indroductionis in universam geographiam Lipsius, Justus, 130
(Cluverius),350 Lister, Martin, 377
Inscriptiones vel titull Theatri amplissimi L'Obel, Matthew, 145
(Quiccheberg), 190, 192 London shops; see also Elizabethan London
Instrument makers, in Elizabethan Lon- Bernoulli, Bugge, and Van Marum, 370,
don, 148-149; see also Scientific 379-384
instruments Cassini's analysis, 384-391
International trade; see Commerce as commercial enterprises, 389
Inventory investment, 224-225 Monconys, Von Uffenbach, and Lalande
Invisible Cities (Calvino), 185 in, 371-379
Italian Renaissance, 15,416 Long, Pamela 0., 63, 417
I vory trade, 183 Lycosthenes, Conrad, 308

Jamnitzer, Wenzel, 47-49 McConnell, Anita, 388


Jardine, Lisa, 7 Machines, in Madrid Codex I, 65, 67-71
Jodar, Andres de, 173 "Madonna of Humility" (Diirer), 29
Jonstonius, Joannes, 313-314 Madonna with the Mayfly (Diirer), 29-31
Josephus, Flavius, 190 Madonna with the Monkey (Diirer), 31- 34
Madonna with a Multitude of Animals
Kaempfer, Engelbert, 347-349, 364 (Diirer), 31
Kaufmann, Thomas DaCosta, 13,412 Madrid Codex I (da Vinci), 65, 67-71, 80
Kemp, Martin, 13, 19 Madruzzo, Christoforo, 190
"Kepler-Fludd controversy," 418 Magellan, Ferdinand, 98, 102
Khunrath, Heinrich, 215-216 Magic lantern, 371-372
Kircher, Athanasius, 318 Maier, Michael, 215-217
Koyre, Alexandre, 16 Male, Emile, 413
Kiiffeler, Johann, 372 Maler, Hans, 192
Kuhn, Thomas, 404 Malocchi, Francesco, 304
Kunstkammern,4-6, 182- 183,412; see also Mapmaking; see Cartography; Sea charts
Curiosity cabinets; Wunderkam- Marketplace of Natural Marvels (Serpetro),
mern 297,299
Kynvin, James, 148-149 Marshall, John, 374- 375
Martin, Benjamin, 379
Ladurie, Emanuel LeRoy, 254 Martyr, Peter, 166
Laguna, Andres, 172 Master of the Enclosed Garden, 29
Land survey methods, Canal du Midi, 255- Materia medica (Dioscorides), 113, 171
259 The Mathematical Jewel (Blagrave), 149
L'arte di restituire (Meijer), 289- 291 Mathematical vs. optical instruments, 390
Las Meninas (Velazquez), 414 Mauss, Marcel, 4
Lauterwald, Alexander, 205-206, 210, 214- Maximilian II (Holy Roman Emperor), 190,
216 193

43 2 Index
Meadow, Mark, 6, 182,408,412 Mylius, Johann Daniel, 45
Mechanical arts, 63, 407
in Elizabethan London, 147-151 Nairne, Edward, 379
in Madrid Codex I, 65, 67-71 Narratives of place, 258-259
Medical practitioners, 143-147 Natural history, 17,302
Medicina diastatica (Tentzel), 231 basilisks as iconogra ph y of, 313
Medina, Pedro de, 89, 91-92, 94, 96 inventing nature, 310-311, 319
Meijer, Cornelius, 278-285, 405-406 science vs. commerce, 302-303
Accademica Fisicomatematica and, 288- trafficking in, 301-307
289 Natural History (Pliny), 301, 304
Apologia of, 290-292 Natural History of Serpents (Jonstonius), 313
biography and early work of, 279-280 Natural History of Strange Fish (Belon), 310
conflict with Fontana, 282-283,290-291 Natural philosophers, 13-14, 17
defense and defeat of, 290-292 Natural sciences, 9-10
engineering skills of, 405 practitioners in Elizabethan London,
L'arte di restituire (Meijer), 289-291 137-139
passonata in the Tiber, 280-282, 290-291 Naturalia, 5, 124,402
plan for St. Peter's Square, 284-285 vs. artificialia, 325
Merian, Maria Sibylla, 14 in curiosity cabinets, 182- 183
Merkel Centerpiece (Jamnitzer), 49 dried objects, 226-227
Mersenne, Marin, 19 medical instruction and, 113
Merton, Robert K., 15-19,413 preparation/preservation of, 223-241
"Merton thesis," 15-16 trafficking in, 301-307
Mexia, Pedro, 87, 89, 101 tropical naturalia, 347
"Microcosms: Objects of Knowledge" (Univ. Naturalism, 8-9, 11, 13,47
of California), 186 Naturalistic representation, 8-9, 109
Micrographia (Hooke), 372 blowfish, 112-113
Microscopes, 372 Nature
Military cartographers, 257 art and, 65, 72,324-328,402
Mining enterprises, 210-212 consumption and, 18
Mistrell, Eloy, 149 in Hypnerotomachia Poliphilt: 65--67, 74
Modern science; see also Scientific Revolu- representations as standardization of,
tion 404
capitalism and, 401 visual representations, 417
Monardes, Nicolas, 169, 172 Navigational techniques/instruments, 83,
Monconys, Balthazar de, 371-374 165,370
Montanus, Arnoldus, 347-349, 359-360 Neoplatonism, 44, 401
Montias, John Michael, 15 New World; see also Santa Domingo balsam
Morales, Garciperez, 169 commerce and knowledge in, 163-165
Moran, Bruce, 209 empirical information/validating knowl-
Morland, Samuel, 374 edge of, 164-165, 175
Morland's calculating machine, 374-375 health issues in, 163
Mornay, Robert, 371 history of science and, 164
Moscardo, Lodovico, 315 medicine of, 164
Motion, Leonardo's studies of, 68-70 natural resources of, 164
Mountebanks, 305 trading and commerce, 163
Moxon, Joseph, 374 Newton, Isaac, 19,375,407,415
Mukerji, Chandra, 18,228,248,406,412 Newton's reflector, 375
Mummies, 230-232 Nieuhof, Johan, 357, 360
Muzafar II (Sultan), 1 Nonez, Hector, 144

Index 433
Nummedal, Tara E., 201 Portugal, territorial disputes with Spain, 83-
94
Observatoire de Paris, 384-386 Post, Frans, 352-353, 355, 360
Octants,371 Pound, John, 375
Ogilby, John, 355, 360, 363 Praun, James, 374-375
Oldenburg, rIenry, 371, 373 Preservation arts
On Fish (Aldrovandi), 311 animal parts/structures, 228, 240-241
Optical instruments, 390 botanical materials, 228
The Order of Things (Foucault), 414 chemical processes for, 227-228
Oviedo, Gonzalo Fernandez de, 18, 166 De Bils' process, 234-235
dissection and, 229
Pacht, Otto, II, 19 dried objects, 226-227
Padron real (royal pattern chart), 85-96 embalming methods, 230-236
territorial disputes and, 97-101 lifelike bodies, 229-235
Pagden, Anthony, 17,360 mummies, 230-232
Palissy, Bernard, 50-51 turpentine oil, 238
Paludanus, Bernardus, 124,226 "Prince Rupert's drops," 372
Panofsky, Erwin, II, 13,413,416 Pringle, John, 379
Panvinius, 190 Puritanism, 15
Paper museums, 8 Pyefinch, rIenry, 379
Paracelsus, 46-48, 145, 147, 151,202,206,
209,214,231 Quadrants, 378
Paris shops, 382-383 Quiccheberg, Samuel, 190-195
Bugge's visit to, 391-392
Cassini's "workshop" proposal, 385 Ramsden, Jesse, 380, 386, 388
as commercial enterprises, 389 Ramus, Petrus, 41
Van Marum's visit, 382-383 Rariorum Plantarum rIistoria (Clusius), 120
Park, Katharine, 325 Rationalism, 17,401,404,416
Parke, Edward, 140 Ray, John, 305-306
Patrick, John, 417 Rea, John, 328
Patronage, commerce and, 3-7 Recchi, Nardo Antonio, 175
Pauw, Pieter, 124, 128 Reeve, Richard, 371-374, 379
Penny, Thomas, 144 Relics, trade in, 302
Perseus Beheading Medusa (Cellini), 50 Representations (periodical), 415
Perspectograph, 373 Rest on the Flight into Egypt (Altdorfer),
Pharmacy, 305-306 34-38,44,52
Philip II (King of Spain), 183 Rhinoceros (Rhinoceros unicorni.;), 1-4, 13
Philipps, Edward, 144 Riegl, Alois, 413
Philipps, Emma, 144 Riquet, Pierre-Paul, 249-256, 259-260,
Philosophia naturalis, 63 264-265,270
Physica,63 Roberts, Lissa, 399
Physico-Mechanical Experiments (rIauksbee), Rome in 17th c., 277-292; see also Meijer,
376 Cornel is
Physiologia, 63 Accademia Fisicomatematica, 278, 285-
Pickering, William, 137 290,405
Pigott, Nathaniel, 382 Rorty, Richard, 414
Pizzaro, Francisco, 188 Rowley, John, 375, 390, 417
Pliny, 18,64,301-302,304,311 Royal Academy of Mathematics (Madrid),
Polhaimer, Michael, 205 164
Pomian, Krysztof, 4 Royal College of Physicians, 141, 143-144

434 Index
Royal patents, 147-151 Paris shops, 377
Royal Society, 371-373, 376-377, 390-391 Royal Society and, 390- 391
Rudolf II (Holy Roman Emperor), 36, 182, trade affiliations for, 389-390
191, 193,209 Scientific Revolution, 3, 11, 16, 18-19,399,
Rumphius, Georg, 347-349 405
Russwurin, Valentine, 145-147 early practitioners, 405-406
Ruysch, Frederik, 241-242 mechanics/mathematics and, 418
nature and, 16
Sadeler, Aegidius, 31 social divide and scientific knowledge,
St. Johannis Kirche (Durer), 71 277
Saint John the Evangelist and John the Baptist Sea charts
(Altdorfer),36-37 astronomical navigation and, 95-96
Saltpeter, 150 commerce and, 83
Sandman, Alison, 83, 406 compass navigation, 96-97
Santa Cruz, Alonso de, 86-87, 89-90 controlling navigation, 91-97
Santa Domingo balsam, 164-165 cosmographers, 86-87
Barreda's report, 170, 174 in Dutch Republic, 352
commercialization of, 169-170 Falero's proposals, 89
crown's and knowledge production, information reliability in, 88-91
172-173 magnetic declination problem, 92-94
testing experience of, 170-175 navigation and, 84
Villasante's report on, 166-170, 174 padron real, 85-87
Saxl, Fritz, 413 sea pilots and, 88, 90
Schafer, Ernst, 166 Spanish-Portguese disputes, 83-84
Scherdinger, Abel, 203, 205 territorial claims/disputes, 83, 97-101
Schl uter, Andreas, 416 Sea pilots
Schmidt, Benjamin, 347, 406, 412 astronomy and, 95- 96
Schongauer, Martin, 51 compass navigation, 96-97
Science sea charts and, 88, 90
art and, 13-14, 17-19 Seba, Albert, 318-319
capitalism and, 15-16 Securie, John, 137
classical histories of, 413 Seghers, Daniel, 325
commerce and, 15- 18 Serlio, Sebastiano, 65, 74-79
construction of, 17 Serpetro, Niccolo, 297, 299, 301
consumption and, 18-19 Settala, Manfredo, 300
foundations of, 3 Shapin, Steven, 414
historiography of, 17,414-415 Shoring, John, 144
as matrix of marvels, 402-404 Short, James, 375, 377
Puritanism and, 15 Silver, 210
The Science ofArt (Kemp), 13 Silver, Larry, 29
Science, Technology, and Society in Seven- Sisson, Jeremiah, 378, 380, 382
teenth-Century England (Merton), Smith, Adam, 409
15 Smith, Adison, 381
Scientia, 47 Smith, Pamela H., 1,29,401,413
Scientific instruments, 372, 402 Smythe, John, 140
as commercial enterprises, 389 Sommering, Philipp, 203, 205, 213, 217
in Elizabethan London, 147 Sorbiere, Samuel, 229, 372
foreign commissions for, 380 Spain
in France, 381 in 16th century, 83-102
mathematical vs. optical instruments, 390 controlling navigation, 91-97

Index 435
navigation!cartography in, 83 exchange/social relationships and, 335-
padr6n real, 91-92, 95 336
sea charts as territorial claims, 97-101 first description of, 326
territorial disputes with Portugal, 83-84 luxury and, 330-331
Spanish America; see New World as works of art, 324-328
Sphere (Sacrobosco), 95 Turba philosophornm, 43
Spinoza, 408 Turner, Peter, 137
Splendor Solis, 42 Turpentine oil, 238, 240
Standardization, 408
Stevens, Richard, 148 Universal Piazza (Garzoni), 299
Stevin, Simon, 416
Still-life paintings, 109, 129 Valkenburg, Dirk, 352
Stowghberghen, Peter, 150 Vallisnieri, Antonio, 318
Strada, Jacopo, 190-194 Valnay, Nicolas de, 324-326
Strieder, Peter, 71 Van Berkel, Klaas, 277, 405
Suarez de Carbajal, Juan, 87 Van Campen, Jacob, 353
Subterranean World (Kircher), 318 Van der Aa, Pieter, 352
Sumario de la Historia Natural (Oviedo), 166 Van Duran, Peter, 141
Surveying instruments, 6 Van Edema, Gerard, 352
Swammerdam, Jan, 229, 236-237, 240-241, Van Eyck, Jan, 8
300 Van Horne, Joannes, 230, 232-233, 236, 240
Swan, Claudia, 10, 109,403,414 Van Mander, Karel, 44
Swan en burgh, William, 113 Van Marum, Martinus, 382-383
Van Meurs, Jacob, 348-349
Tabular diagram, 122-130 Van Winle, Casper, 278-279
Tacuina sanitatis, 417 Varenius, Bernard, 350
Taxonomy; see Classification system Vellam, Hugh and Ann, 144
Taylor, Paul, 325 Vesalius, Andreas, 8,65,74-78, 80, 400
Telescopes, 371 , 374-375 Villasante, Antonio de, 166-168, 173-175
Tentzel, Andreas, 231 Villasante's report on balsam, 166-170, 175
Thompson, Anthony, 373 Visual images, 7-11,109,417
Time; see also Preservation arts Durer's work, 71-74
commerce and, 224, 241 in Hypnerotomachia Poliphili, 66-67
mechanical timekeepers, 224 in Madrid Codex I, 65, 67-71,80
physical vs. social time, 224 natural world and, 63-66
Titian, 74, 192 Serlio and Vesalius, 74-80
Trade affiliations, 389-390 Vitruvius, 75-79
Trading places, 185 Vives, Juan Luis, 41
Travel narratives, 350- 352 Voltaire, 409
Treasure of Medical Things of New Spain Von Uffenbach, Zacharias Conrad, 374-376
(Hernandez), 317 Vredeman de Vries, Hans, 416
Treatise of Commerce (Wheeler), 142 Vulpe, Johann, 144
Treaty of Tordesillas, 84, 98
Triumph, with Treasures of the Indies (van Wackernagel, Martin, 413
Campen), 353 Waldberg, Ono Truchess von, 190
Tulipmania (1634- 37), 324 Warburg, Aby,413
Tulips Warburg Institute, 413
aesthetic comments on, 333 Watchmakers, 382
cultivation as art, 328-329 Watkins, Francis, 377-378
dwarf tulip, classification of, 120-121 Weber, Max, 4,15

Index
Weiherhaus,31 commerce in, 193-195
Wheeler, John, 4,142 Dutch objects, 353-354
White, John, 8,148 Ferdinand I, as founding, 190
Wire Drawing Mill (Durer), 71 as intellectual capital, 193-194
Wolf, Caspar, 127 profit motive and, 408
Wolf, Hieronymus, 190-191 purposes/structure of collection, 195
Wolffin, Heinrich, 413
Wollaston, Francis, 388 Zieglerin, Anna, 206-207, 209
Wren, Christopher, 373
Wunderkammer, 124, 182-186,403,412; see
also Fugger, Jacob

Index 437

You might also like