Forms of Truth

and the

Unit y of Knowledge

Edited by

VITTORIO HÖSLE

-
 Forms of T ruth
and t he
Unit y of Know ledge
 Forms of Truth

and the

Uni t y of Kn owledge

Edited by

VITTORIO HÖSLE

-
University of Notre Dame Press

Notre Dame, Indiana

 Copyright © 2014 by University of Notre Dame
Notre Dame, Indiana 46556
undpress.nd.edu
All Rights Reserved

Manufactured in the United States of America

The Press gratefully acknowledges the support of the Institute for Scholarship
in the Liberal Arts, University of Notre Dame, in the publication of this book.

Library of Congress Cataloging-in-Publication Data

Forms of truth and the unity of knowledge / edited by Vittorio Hösle.

pages cm
Includes bibliographical references and index.
ISBN 978-0-268-03111-4 (pbk. : alk. paper) —
ISBN 0-268-03111-8 (pbk. : alk. paper) —
ISBN 978-0-268-08177-5 (e-book)
1. Knowledge, Theory of. 2. Truth. 3. Science —Philosophy.
4. Humanities—Philosophy. 5. Art and religion.
I. Hösle, Vittorio, 1960– editor.
BD161.F587 2014
121—dc23

2014022434

∞ The paper in this book meets the guidelines for permanence

and durability of the Committee on Production Guidelines
for Book Longevity of the Council on Library Resources.
 C o n t e n t s

Introduction 1
Vi t to ri o Hösle

P a r t I
T H E H I ST O RI CAL DEVELOPMENT
O F T H E T RE E O F KNOWLEDGE

C h a p t e r O n e
How Did the Western Culture Subdivide Its Various
Forms of Knowledge and Justify Them? Historical Reflections
on the Metamorphoses of the Tree of Knowledge 29
Vi t to ri o Hösle

P a r t I I
EPI ST E M O L O GY, L O GI C, AND MATHEMATICS

C h a p t e r T w o
Intuition and Coherence in the Keystone Loop 73
Ke i t h L ehre r

C h a p t e r T h r e e
What Is the Nature of Inference? 89
Ro b e r t Hanna

C h a p t e r F o u r
Speculation and Narration in Mathematics 102
L aure nt L afforgue
 vi Contents

P a r t I I I
EXPL AN AT I O N I N T H E N ATURAL SCIENCES

C h a p t e r F i v e
A Molecular Glimpse of How Mother Nature
Can Regulate Our Being 115
Th o m as N owak

C h a p t e r S i x
What Light Does Biology Shed on the Social Sciences
and the Humanities? 140
Franci s co J. Ayala

C h a p t e r S e v e n
What Is the Nature of Perception? 159
Z y gm unt Piz l o

P a r t I V
INT RO SPE CT I O N AN D UNDERSTANDING
I N T H E H UM ANITIES

C h a p t e r E i g h t
Is Introspection (a First-Person Perspective)
Indispensable in Psychology? 181
O s b o rne Wi ggins

C h a p t e r N i n e
Could Normative Insights Be Sources
of Normative Knowledge? 195
Al l an Gi b bard

C h a p t e r t e n
Truth and Knowledge in Literary Interpretation 216
Cars t e n Dut t

C h a p t e r E l e v e n
Historical Truth 232
Av i e z e r Tucke r
 Contents vii

C h a p t e r T w e l v e
Truth and Unity in Chinese Traditional Historiography 260
N i co l a Di Cosmo

P a r t V
ART AN D RELIGION

C h a p t e r T h i r t e e n
The Architecture School within the University 289
M i ch ae l Ly koudis

C h a p t e r F o u r t e e n
How Is Theology Inspired by the Sciences? 300
Ce l i a De ane - Drummond

About the Contributors 324

Index 332
 Introduction

-
Vittori o H ö s l e

The Notre Dame Institute for Advanced Study was founded in 2008 in
the belief that every research university needs a place where scholars
come together so as to be inspired by colleagues in disciplines differ-
ent from their own. There is no doubt that the subdivision of universi-
ties into departments is reasonable: given the increase of knowledge in
every discipline, controlling the quality of work in a given academic
field presupposes expertise in that field. At the same time, the dangers
connected with this departmentalization of knowledge are various. For
example, people may simply ignore what occurs outside of their own
fields. Sometimes this may not affect the quality of their research; some-
times, however, this leads them to miss opportunities to connect their
own studies with those of others and thus achieve more general insights.
This danger is particularly menacing whenever basic insights are ignored
that have repercussions for more applied research. As physics cannot
progress without mathematics, so also the humanities, to give only one
example, inevitably operate with basic philosophical categories, such as
meaning or value; the lack of an explicit reflection on these concepts
does not mean that they are not presupposed. On the contrary, the less
reflection occurs, the more likely the concepts used are imprecise and

1
 2 Vittorio Hö sle

perhaps even inconsistent. But the disconnect of philosophy from the

other disciplines is due not only to the specialization of those disci-
plines: philosophy itself has, to a considerable degree, given up its am-
bition to address a larger audience and withdrawn into very technical
problems that are of interest only to “specialists of the universal,” as one
could call such philosophers. The decline of public intellectuals, the
topic of the Institute’s 2013 annual conference, which was organized by
Michael Desch, who will edit the corresponding volume, is a necessary
consequence of this development—one with dangerous consequences
for politics: without the advocacy of public intellectuals, the political
system is far more likely to be manipulated by populists and those ter-
ribles simplificateurs in the mass media who care more for advertising
revenue than the truth.
Its three inaugural conferences in 2010, 2011, and 2012 were, to-
gether with the residential life of scholars working on research projects
that deal with questions both interdisciplinary and normative, the two
foci of the Notre Dame Institute for Advanced Study. These three con-
ferences addressed the three concepts that have often been considered
by the philosophical tradition the most fundamental—namely, beauty,
goodness, and truth—by bringing together scholars from as many dis-
ciplines as could fruitfully interact in three days. The first two volumes,
The Many Faces of Beauty and Dimensions of Goodness, appeared in
2013 with the University of Notre Dame Press and Cambridge Scholar
Publishers respectively; this volume on conceptions of truth and the
unity of knowledge completes the trilogy. The participants of the three
conferences did not overlap—with the exception of myself: as founding
director of the Institute from 2008 to 2013, I contributed to these three
volumes three essays that complement one another. Those invited to the
conference on truth were chosen to represent philosophy—including
epistemology, logic, and ethics—theology, mathematics, chemistry, bi-
ology, psychology, literary criticism, historiography, and architecture.
They hailed not only from the United States but also from Israel and
various European countries such as France, Germany, Italy, Poland,
Spain, and the United Kingdom, and one essay deals with the Chinese
contribution to the concept of history. The essays in this volume are
collected in five parts, the first dealing with the historical development
of the tree of knowledge; the second with the foundational disciplines of
 Introduction 3

epistemology, logic, and mathematics; the third with explanation in the

natural sciences; the fourth with introspection and understanding in
those disciplines dealing with humans; and the fifth with the contribu-
tion of art and religion.
Some of the common, overarching questions that the contributors
to this volume address are these: By which different methods do the
various disciplines achieve knowledge of truth? What is common to
their methods, and what distinguishes them? Are some disciplines more
foundational than others, that is, can they be understood on their own
while the others presuppose them? Which forms of knowledge influ-
ence each other, and which disciplines have very little overlap? Are there
different ontological realms connected with the various epistemologi-
cal activities? And since it is impossible to give up the belief that the
various disciplines contribute to an ultimately coherent vision of reality,
how should we imagine this tree of knowledge?

Vittorio Hösle’s essay, “How Did the Western Culture Subdivide Its
Various Forms of Knowledge and Justify Them? Historical Reflections
on the Metamorphoses of the Tree of Knowledge,” gives an overview
of how Western culture has grouped its various disciplines from antiq-
uity to the present, often under the metaphor of the tree of knowledge,
which points to the common roots and thus to the underlying unity of
all knowledge. It starts with reflections on the budding of new and the
withering of old branches of that tree, partly due to the uncovering of
unsuspected strata of reality and partly thanks to the development of
new theoretical tools— tools that occasionally allow the unification
of disciplines that earlier were regarded as independent. The develop-
ment of the tree is not always caused by empirical discoveries: shifts in
philosophical categorization based on purely theoretical arguments also
help to explain the various shapes that the tree of knowledge has as-
sumed in its history. Within ancient philosophy, the first insight into
different types of knowledge comes with the Eleatic school’s distinction
between the way of opinion and the way of truth. Plato builds upon
this distinction and connects it to the subdivision of the mathematical
disciplines—even while pointing to a fundamental difference between
philosophy and mathematics. A thinker as early as Xenocrates proposes
 4 Vittorio Hö sle

the subdivision of philosophy into logic, physics, and ethics—a subdi-

vision that would become canonical within Stoicism—but it is Aristotle
to whom we owe the most elaborate system of knowledge of antiquity:
theoretical knowledge encompasses physics, mathematics, and the-
ology; practical knowledge, ethics and politics; and poietical knowledge,
poetics and several other less important subdisciplines. Logic, how-
ever, is difficult to fit into this tripartition: in Hugh of Saint Victor’s
Didascalion, it will form a fourth part. The two main works from the
Middle Ages analyzed in the essay are Bonaventure’s De reductione
artium ad theologiam and Ramon Llull’s Arbor scientiae. Important in
Bonaventure is the attempt to include the trivium of grammar, logic,
and rhetoric into this system as well as the insertion of economics—
that is, the discipline of the management of a family— between ethics
and politics. The climax of all knowledge is a theology based on the
fourfold interpretation of scripture. In early modernity, Francis Bacon
proposed the most articulate and influential system of disciplines based
on the three faculties of memory, imagination, and reason. However,
his system is soon challenged by Descartes’s radical separation of the
knowledge of extended substances from that of thinking substances,
which splits any science of humans in two and raises the difficult
question of how we may have knowledge of other selves, since intro-
spection is limited to oneself. In the eighteenth century, Giambattista
Vico proposed the idea of a new science that addresses human culture,
understood as a realm beyond those of nature and the mind. The essay
then explores the subtle alterations of the Baconian system in Jean Le
Rond d’Alembert’s introductory “Discours” to the Encyclopédie and
delves into Kant’s epistemological challenge and his idea of a new sys-
tematization of knowledge, which was elaborated by the greatest ency-
clopedist among the German Idealists, namely, Hegel, who recognizes,
against Bacon and d’Alembert and anticipating Frege and Husserl, the
irreducibility of logic to any other discipline. Hegel’s subdivision is fi-
nally contrasted with the almost simultaneous one by Auguste Comte,
who had an enormous impact on the structure of nineteenth-century
universities by conceptualizing the peculiar status of engineering and
having his doctrine of the sciences culminate in the new discipline of
sociology.
 Introduction 5

It is one of the paradoxes of our search for truth that, while knowl-
edge has grown exponentially in the last few centuries, epistemology
has not been able to answer in a satisfying way its most basic problem,
which a thinker as early as Plato addressed at the end of his Theaete-
tus. Some of our knowledge is inferred from other knowledge, but this
inference process cannot go on forever. But how do we grasp the ulti-
mate premises of our inferences? Is there an immediate knowledge—
is there, that is, what some have called “intuition”? But how, then, do
we react to those who do not share our intuitions, which may be ei-
ther conceptual or perceptual? If we try to justify our intuitions, we
seem to deprive them of their immediacy and thus of their evidence,
which must not depend on anything else. If, however, we refuse to do
so, we seem to give up a basic demand of rationality. Therefore, coher-
entism has been developed as an alternative to foundationalism: there
are, it holds, no basic premises from which to start, and so it is only the
coherence of our whole set of beliefs at which we should aim. But does
this not imply either a circular “justification” or an infinite regress?
The quarrel between foundationalists and coherentists is old: Hegel
may be regarded as paradigmatic coherentist, Husserl as quintessential
foundationalist.
In his essay, “Intuition and Coherence in the Keystone Loop,”
Keith Lehrer, one of America’s leading epistemologists, tries to pave a
middle ground between foundationalism, which for him is sympto-
matically exemplified by the eighteenth-century Scottish philosopher
Thomas Reid, and coherentism, as whose champion he regards the
twentieth-century American thinker Wilfrid Sellars: “It may be that
intuition and coherence must be joined to yield the kind of evidence
required for knowledge.” In fact, Lehrer tries to prove that Reid is less
of a foundationalist than he seems at first glance, for he teaches the
dependence of the principles of intuition on each other and on what
Lehrer calls “the First First Principle,” to wit, “that our faculties by
which we distinguish truth from error are not fallacious.” This prin-
ciple, furthermore, vouches not only for the other first principles but
also for itself — like light, which reveals itself as well as other objects.
(Lehrer calls such principles “keystone principles.”) But this means
that other first principles are not by themselves maximally evident,
 6 Vittorio Hö sle

since they are justified by the First First Principle. Sellars’s rejection of
the myth of the given is based on the insight that a person who has sen-
sory states does not necessarily have a conception of them. But an intu-
itionist could counter that, while this is true, it does not yet show that
the evidence connected with the conception has to be inferential, for
“not every transition from one state to another is an inference.” Still,
for Sellars evidence of truth is explained by a system, while for Reid
immediate convictions are “born justified.” Lehrer then discusses the
challenge that skepticism represents to both the foundationalist and the
coherentist. He recognizes that the skeptic cannot be confuted without
begging the question— but also that this does not endanger any claim
to knowledge: “we can know what we suppose we know, even though
we cannot prove this to the skeptic.” Lehrer insists, furthermore, that
the explanation of how we can know something is not a premise in
the justification of that belief; therefore, we may have “an answer to the
question of how the justification of all beliefs can be explained though
some beliefs are noninferentially justified.” Lehrer avers with Sellars
that all our beliefs are subject to revision. Reflections on the difference
between primitive and discursive knowledge, of which only the second
may reply to objections—as well as on the nature of exemplarized ex-
emplars, which are instances of themselves—conclude the essay.
In “What Is the Nature of Inference?” Robert Hanna, author of
the standard work Rationality and Logic, discusses in detail the nature
of the process that was the basis of the earlier epistemological inves-
tigation. He divides the issue into four separate questions regarding
inference—which can be deductive, inductive, or abductive: he inves-
tigates its metaphysics, its purpose, its justification, and its mechanism
before proposing as a solution what he calls “Contemporary Kantian
Moralism about Inference.” Concerning the first, he defends transcen-
dental mentalism in opposition to both psychologism—which does not
capture the objectivity, necessity, and apriority of inferential facts—
and Platonism—which is supposed to be incompatible with the causal
triggering of human knowledge. With regard to the second, he subsumes
the theory of inference under the metaphysics of morals, for it deals
with the summum bonum of reasoning. (He later speaks of a “special
logico-practical ought.”) Emotivism and instrumentalism (or pragma-
 Introduction 7

tism), on the other hand, imply that “anything goes,” provided that
everyone shares the same feelings and good results are produced from
the standpoint of human interests respectively. Third, he deals with
the old problem that the inferential principles to be justified are them-
selves presupposed in the process of justification. Hanna rejects non-
cognitivist, holist, and inferentialist strategies as unable to warrant the
objectivity, necessity, and apriority of our inferential principles and
defends categorically normative logical laws. (He does not investigate,
however, the problem of the plurality of logics.) The mechanism of
inference finally rests on the causal power of my intentional act of in-
ference. Inference requires both consciousness and consciously free
willing —an “inferential ‘zombiehood’” is doomed to fail. In his con-
clusion, Hanna characterizes his idealism as liberal and natural, not
scary. It simply states that fundamental mental properties are co-basic
in nature with fundamental physical properties.
The discipline in which inference plays the most exclusive role is
doubtless mathematics—excepting, of course, logic itself. Laurent Laf-
forgue, Fields Medalist and one of the world’s leading mathematicians,
honored the conference with his presence and delivered an essay, fas-
cinating in both its content and its literary quality, titled “Speculation
and Narration in Mathematics.” Lafforgue starts from the fact that
truth cannot simply be seen, but must be lived, for otherwise the con-
cept of truth would be pornographic rather than nuptial. And so he
offers a confession of what it means to live a mathematician’s life, even
if mathematics seems the most impersonal of all disciplines. But he be-
gins his exploration with an approach, as it were, from outside: a non-
mathematician who observes a department of mathematics will first
notice the far higher proportion of people to equipment than in most
other settings and then find out, “not perhaps without feeling a degree
of terror,” that they use incomprehensible words. If he then ventures
to open one of their texts, he will discover that it is “as if it told a story
in chronological order and . . . as if the specific story . . . were part of a
general step in mathematical science and in the history of that onward
march.” At the same time, the mathematical text uses only the present
tense and in fact refers to timeless structures of logical implication.
The tension between the narrative and the speculative element is what
 8 Vittorio Hö sle

constitutes mathematics; this tension may generate the erroneous belief

that implication is a form of causality.
However, a mathematical text is not only a temporal manifestation
of something eternal: if we consider the mental life of the individual
mathematician, it occupies a place within the history of mathematics as
well—as becomes obvious when one considers those theorems named
for the mathematicians who discovered them. Lafforgue distinguishes
two ways of employing the combination of speculation and narration
peculiar to mathematics: one can either use old methods in new analy-
ses or explore mathematical terrain that is entirely new. Paradoxically,
Lafforgue notes, mathematicians speak of events in their mathemati-
cal texts: something “happens” in mathematics when a new vista opens
through the development of a new concept or the solution of an old
problem. Regarding the inner motivation of mathematicians, Lafforgue
recognizes the striving for recognition and honors as a common mo-
tive. But far more important is the desire to share precious things, such
as mathematical insights— even if this leads to an overproduction of
texts that are ignored by almost all human beings; in fact, even the most
famous mathematicians “are not sheltered from self-doubt and de-
spair.” Therefore, another motive is the desire to write to oneself: “a
mathematical narrative is like a travel story in which, in fact, the jour-
ney consists of the narrative itself.” The enormous effort in concentra-
tion, so Lafforgue ends his analysis, seems to go hand in hand with the
experience of inspiration, and even if few mathematicians reflect on the
author of these inspiring words, “they find their only joy in becoming
servants to these words.”
Truth can be found not only in logic and mathematics but also in
the natural sciences—those that deal with the physical world, not with
ideal entities. Pure thought does not suffice to get at truth in this realm;
experience is indispensable. Within nature, one class of objects is of par-
ticular importance: humans. Of all the animals known to science, it is
humans alone who seem to be able to grasp truth, and this is due—at
least in part — to some of their biochemical and biological properties.
In “A Molecular Glimpse of How Mother Nature Can Regulate Our
Being,” the biochemist Thomas Nowak addresses both the criteria that
a scientific theory has to satisfy in order to be taken seriously in its
 Introduction 9

claim to approach truth and some specific regulatory mechanisms that

keep organisms alive. While a valid inference guarantees the transfer of
truth—if the premises are true, then also the conclusion must be true—
logic cannot answer the question of which premises should be affirmed
as true. For science, it is clear what is needed in order to determine the
truth of premises: observation and experimental data. Like logical posi-
tivism, Nowak sharply distinguishes between evident data (which he
calls “objective truth”) and its interpretation, which, though it aims at
and sometimes approximates truth, remains inevitably subjective. Inter-
pretations of data are called hypotheses or theories, and they can achieve
a very high degree of plausibility: for example, “virtually all credible
scientists believe in the phenomenon of evolution.” Still, Nowak in-
sists that the evolution of one species into others is only a hypothesis,
not itself an immediate datum. The problem in moving from data to
theories is that there are almost always many reasonable explanations
of the same data. True enough, further experiments may help eliminate
some of the competing explanations, for “hypotheses that are incon-
sistent with the data are clearly incorrect.” But since the choice of ex-
periments depends, among other things, on the technology available
and the technology of one age will always be surpassed in accuracy by
that of the next, absolute certainty cannot be gained. No less impor-
tant than technology in devising new experiments is the imagination
of the scientist using that technology. Laws of nature, according to
Nowak, are nothing other than hypotheses that have withstood exper-
imental challenge for a long time. Nowak then gives several examples
where doctrines once cherished, such as Stahl’s phlogiston theory,
were dropped during the evolution of science: echoing Thomas S.
Kuhn, Nowak speaks of paradigm shifts, such as those that led to rela-
tivity and quantum theory. He then focuses on his own research, the
study of a key step in the regulation of glycolysis: “this study reflects
the asymptotic process of better understanding the details of how a
specific regulatory process occurs.” He chooses this specific meta-
bolic pathway because it is critical to life and virtually ubiquitous in
living things. Among the regulatory mechanisms devised by nature to
respond to physiological challenges, two alternative ones are paramount:
an on/off process in which the catalytic reaction is active or inactive or
 10 Vittorio Hö sle

a determination of the speed of the reaction rate by an enzyme. Light

switches and rheostats are, respectively, the technological equivalents
of these biological mechanisms. Modifications of the enzymes can be
reversed; the secretion of metabolites and hormones may trigger a cas-
cade of regulatory steps. Nowak then analyzes in detail the mechanism
of regulation of the enzyme pyruvate kinase, discussing, among other
things, the feed-forward activation, in which a metabolite formed early
in the pathway activates an enzyme farther down the pathway. The rele-
vant hypothesis was exposed to both kinetic and thermodynamic stud-
ies, and since it proved consistent with the information provided by the
experiments, it continues to be held—and will be until “the results of
further studies become inconsistent with the current theory.”
One might voice against Nowak’s reflections on the nature of sci-
ence three possible objections: first, the demarcation between data and
hypotheses is not easy, since, at least in modern physics, many data
are gained through experiments that already presuppose theories. Sec-
ond, one could try to defend certain basic principles of the natural sci-
ences, such as conservation laws, as a priori valid because, without as-
suming them, experimentation itself would not be possible. Third, in
the case of living things, there are certain constraints on the chemical
processes that occur within a given organism that follow from the na-
ture of that organism—specifically, from its need to survive and repro-
duce. Do such constraints, one might ask, allow us to exclude a priori
certain possibilities that in themselves would be logically possible?
Francisco J. Ayala is one of the most famous biologists of our time,
former president of the American Association for the Advancement of
Science, member of many national and foreign academies, and recipient
of both the National Medal of Science, in 2001, and of the Templeton
Prize, in 2010. His essay, “What Light Does Biology Shed on the Social
Sciences and the Humanities?,” probes the question of how the disci-
plines that deal with that very unique species of animals, namely, hu-
mans, can benefit from the most complex natural science, biology.
After discussing both the origins and the distinctive anatomical traits
of humans— such as our bipedal gait, opposing thumbs, cryptic ovu-
lation, large brains, and vocal-tract modification—he looks at our be-
havioral traits— such as abstract thinking, symbolic language, and
 Introduction 11

toolmaking—which, together, led to a new form of evolution, cultural

evolution. Within human culture, Ayala focuses on ethical behavior
and investigates the issue of “whether ethical behavior was directly pro-
moted by natural selection or has rather come about as an epigenetic
manifestation of some other trait that was the target of natural selec-
tion.” Ayala points to Darwin’s account in the third chapter of The De-
scent of Man, whose two basic points are these: moral behavior is the
result of human intelligence and thus of biology, even if it is only in-
directly promoted by natural selection, but concrete moral norms are
culturally determined. Ayala then touches upon post-Darwinian evo-
lutionary ethics, such as the theories of Herbert Spencer and Julian Hux-
ley, and rejects the idea that the evolutionary process is itself pointing
toward progress: from a biological point of view, the success of bacte-
ria is no less impressive than that of vertebrates, including humans. To
return to the evolution of moral behavior (the justification of which
belongs to a discipline other than biology), Ayala compares the issue at
stake to language: clearly, humans have an innate capacity for symbolic
language, but the particular languages are a result of culture. Analo-
gously, the moral sense is a result of the biological capacities of anticipat-
ing the consequences of one’s own actions, of passing value judgments
(which presupposes the capacity to abstract), and of choosing between
alternatives. (Later, empathy is added.) These conditions come about
after crossing an evolutionary threshold; thus, there is a radical breach
between humans and other animals, even if we can hardly know ex-
actly when in hominid evolution morality emerged. Ayala contrasts his
theory with that of Richerson and Boyd, gene-culture-coevolution,
which “would rather lead to a more nearly universal system of mo-
rality.” According to Ayala, the obvious differences between the moral
codes of various nations are due to cultural evolution. Still, he recog-
nizes that some concrete moral norms must have a biological basis,
because they are ubiquitous and are clearly conducive to biological
fitness—think of parental care—while other norms are even contrary to
fitness. Finally, Ayala delves into the peculiarities of cultural evolution—
peculiarities more Lamarckian than Darwinian— such as its horizon-
tal transmission and its capacity to incorporate elements from differ-
ent cultures. The norms of morality are clearly the product, then, of
 12 Vittorio Hö sle

cultural evolution, even if civil authorities and religious institutions

may reinforce them. (Of course, these institutions are themselves the
products of cultural evolution.)
Zygmunt Pizlo works both in the Department of Psychological
Sciences and in the School of Electrical and Computer Engineering at
Purdue University and is thus doubly qualified to address the other
great source of knowledge beside inference, namely, perception—even
if he shows how much inference is involved in perception itself. In his
essay, “What Is the Nature of Perception?” he shows, among other
things, a great familiarity with the historical reflections on the prob-
lem, especially those of the nineteenth century. Pizlo defends a form
of innatism—one based on strong empirical evidence. He starts by dis-
tinguishing, with Gustav Fechner, three different types of phenomena,
which are causally connected: distal stimuli (physical objects or events
described in the language of physics), proximal stimuli (a distribution
of light, mechanical, or chemical energy on the surface of receptors),
and finally percepts (i.e., mental phenomena). The problem is that sen-
sory data are never sufficient to describe the external object, as be-
comes obvious when we reflect on how people construct out of a two-
dimensional retinal image the corresponding three-dimensional object.
While the mapping from object to image (the so-called direct problem)
is many-to-one, the inverse problem, mapping from the data to the ob-
ject, is one-to-many. But how, given this problem, do we succeed in hav-
ing veridical perceptions—the existence of which Pizlo asserts, rightly
pointing to the fact that even radical skeptics use and presuppose the
functioning of their senses in experiments purported to show the un-
reliability of perception? But Pizlo does not follow Reid’s common-
sense assumption of the trustworthiness of perception: he is closer to
Kant in asserting the existence of hardwired intuitions of space, time,
and causality—even if Kant’s concept of the synthetic a priori does not
imply innate knowledge.
In order to know how other people perceive, science must start
from public events—that is, phenomena that all can observe—and infer
from them private perceptions. It is crucial to separate response bi-
ases from perceptions, which can be achieved by the signal detection
methodology. Pizlo shows by means of the Müller-Lyer illusion that
 Introduction 13

perception proper is far more independent of individual experiences,

such as learning, than was earlier assumed. Our visual system does not
use look-up tables influenced by experience but rather innate computa-
tional algorithms. The inverse problem can only be solved, moreover,
if there are a priori constraints: otherwise, the visual system would have
to try all partitions of the set of six million receptors, and there is no way
this could be achieved in a time even as large as that elapsed since the
Big Bang. “Were the tabula rasa anywhere near blank or clean, the task
of training feature detectors in our visual system would be hopeless.”
However, the very computational intractability of the inverse prob-
lem can be a blessing because it points toward the constraints that we
have to assume exist. Pizlo follows a tradition from Plato to the Gestalt
psychologists in claiming that we perceive perfectly regular shapes not
because we have encountered them in reality but because universals
rather than particulars structure perception: “concrete objects are in-
stantiations of the abstract concept of symmetry rather than symmetry
being abstracted . . . from particular objects.” Finally, Pizlo describes
the way in which a robot constructed by his team according to such
principles can deal with the inverse problem achieving results not very
different from those of humans.
Osborne Wiggins is one of the best-known American philosophers
working in the phenomenological tradition. That philosophical school,
founded by Edmund Husserl, insists on the irreducibility of the first-
person perspective to the third-person perspective, and so Wiggins’s
essay is in a way complementary to that of Pizlo, who, as a scientist, an-
alyzes perception from outside—that is, from a third-person perspec-
tive. Wiggins starts with reflections on the proper object of psychology:
is it mind, overt behavior, or the brain? Even if he is sympathetic to the
answer that all three are legitimate, his main concern is to find out what
“it is like to be a human being for the human being himself or herself.”
He dislikes, however, the traditional term introspection—which is used,
for example, by William James—since it is loaded with dualistic meta-
physical assumptions that Wiggins rejects; therefore, he prefers terms
like self-awareness. Mental processes are experiences of something, of
my body and the world beyond it, and the mind given through self-
awareness is an embodied mind. At the same time, body and world are
 14 Vittorio Hö sle

given as experienced. The need for a study of the latter as experienced

becomes particularly obvious in psychopathology, for the spatiotem-
poral world the schizophrenic experiences must not be identified with
space and time experienced by normal people. When I am aware of men-
tal processes, I am thus aware of processes that are themselves aware
of something. Wiggins then distinguishes two types of self-awareness.
On the one hand, there is a prescientific one, which involves a direct, al-
though indistinct, givenness of ourselves to ourselves. Connected to it is
an indirect awareness of the experiences of other people, “and these in-
tertwined and comingled experiences of the subjective teach us . . .
the workings of human subjectivity as such.” This first sort of self-
awareness is implicit and tacit; it forms part of the field within which I
focus on an object. On the other hand, the explicit self-reflection, which
leads to psychology as science, draws some of its meaning from our pre-
scientific self-awareness. The self disclosed in such reflection is embod-
ied as well, and it is directly given to one’s attention. This does not imply
that its data are uninterpreted—but they are still data. The direct given-
ness is an epistemological advantage, since there is no need to bring in
inference. Even if this direct givenness is the ultimate court of appeal, we
still want claims about it to be recognized intersubjectively. And there is
only one way to do this—specifically, “by inviting one’s critic to try to
directly see for himself.” Of course, the decisive problem, which “has al-
ways constituted the Achilles’ heel of a psychological science utilizing
‘introspection,’” is this: how can I hold the justified belief that my own
experiences exemplify general features of humans? Still, Wiggins con-
vincingly argues that there is no way of avoiding introspection. Even
in the third-person approach there is an implied appeal to the reader to
“imagine” the experiences the object of the analysis is supposed to have.
Only thus can we examine reflectively the properties of those experi-
ences. Imagination has the peculiar power to create the experiences that
it imagines having, and all of the social sciences presuppose this capacity.
This leads to final reflections on the importance of expertise—for even
direct self-reflection is a hard achievement that requires schooling.
One of the most difficult types of knowledge to understand is ethi-
cal knowledge. For a society, it seems crucially important that people
know, or at least believe to know, that extorting money is wrong, and
 Introduction 15

yet philosophers do not agree on the peculiar nature of this knowl-

edge. Some philosophers, most famously Kant, have tended to assimi-
late it to logical or mathematical knowledge; not so the British tradition.
It is in this tradition that Allan Gibbard, doubtless one of the most in-
novative and famous ethicists of our time, is rooted. His essay, “Could
Normative Insights Be Sources of Normative Knowledge?,” begins with
a depiction of ethical intuitionism as it was upheld by Henry Sidgwick
and George Edward Moore. Neither was an empiricist: they did not be-
lieve that moral questions can be answered by experiment and observa-
tion, even introspective observation, alone. Rather, any answer to moral
questions ultimately presupposes specifically moral intuitions. Good,
according to Moore, is a simple, non-natural property, and this holds
also for the alternative basic concept, being warranted. Warrant is the
subject of normative inquiry, of which ethics is only a part: normative
epistemology is another, and thus the concept of probability is norma-
tive, too. Even denying the existence of such non-natural properties
seems self-defeating, for, if we claim that we should be skeptical regard-
ing them, we must ask, “is this ‘should’ itself something non-natural,
and so something we should banish from our thinking?” The use of
“should” seems ineludible. The grasp of fundamental normative facts,
so Sidgwick concluded, occurs in intuitions. This philosophical con-
cept must be distinguished from the psychological one, according to
which intuition is a rapid and effortless type of thinking, for the latter
does not imply that intuitions convey truth. Gibbard does not deny
that ethical disagreements are often due to the difficulties of knowing
non-normative facts. But this does not alter the outcome that all nor-
mative knowledge must rest on fundamental normative knowledge—
for example, that misery is bad. This is not controversial, but other
equally fundamental normative claims are: think of the question whether
happiness and misery are all things that matter in themselves. Since Gib-
bard rejects deep analogies between ethical and mathematical knowl-
edge but is unwilling to pay the very high price of skepticism regard-
ing normative issues, he turns to naturalistic theories. Some of the most
sophisticated are ideal response theories—the main problem with which
is that Moore’s objections against naturalistic definitions of good can
be directed also against naturalistic definitions of ideal.
 16 Vittorio Hö sle

Gibbard, however, fundamentally agrees with Richard Brandt’s

“qualified attitude method”: our affective capacities must be impartial
and informed, must not stem from an abnormal state of mind, and must
accord with general principles. Since this is not intended as a definition
of what “morally required” means, it does not violate the prohibition
against naturalistic definitions. Gibbard then deals with expressivism
and criticizes Alfred Jules Ayer’s version, since I can feel disapproval
of something without thinking that it is really wrong. Gibbard’s own
version of expressivism is more complex: calling a state of mind such as
approval “warranted” expresses a state of mind akin to planning how
to feel. In order that the flashes of moral affect that constitute moral in-
sights become knowledge, they have to deliver truths in a nondefective
way. “For a belief to constitute knowledge . . . it must be true, it must
be warranted, and the processes that eventuate in true belief must be re-
liable.” For ethics, moral feelings are warranted if and only if they meet
the standards of the qualified attitude method. Finally, Gibbard discusses
one of the main objections against his theory of moral knowledge—the
fact that one’s moral insights may contradict those of another. This is
particularly true for different cultures, even if the attitudes of both meet
the standards of the qualified attitude method. But how can we speak of
knowledge if my moral attitudes are determined by the happenstance of
belonging to one group or another? In his answer, Gibbard points to
what he calls “tempered moral realism”: we are warranted in our con-
viction that cultures with other moral sensibilities are wrong, even if
they do not manifest any epistemic defect apart from disagreeing with
us. That is how ethics differs from science. Therefore, we should speak
only of moral “quasi-facts” and normative “quasi-knowledge.”
The humanities are probably best understood as interpretive
disciplines—that is, as disciplines that try to reconstruct the inten-
tional acts of other subjects based on the expression of these intentions
in physical events or objects. They aim at mental acts, even if they are
not based on a first-person approach, since they deal with other per-
sons’ expressions. But how can we access other persons’ mental acts?
Against the widespread skepticism that has paralyzed much of what
occurs today in the humanities, Carsten Dutt, a German philosopher
and literary critic who was Hans-Georg Gadamer’s main intellectual
 Introduction 17

interlocutor in the last decade of his life, defends in his essay, “Truth
and Knowledge in Literary Interpretation,” the claim that the paradig-
matic hermeneutical discipline on which he focuses, namely, literary
criticism, can acquire knowledge, which he explicates, with a definition
going back to Plato, as justified true belief. However, literary interpre-
tation is not only a cognitive process; it is also a creative one, for the
text has “zones of indeterminacy . . . embedded in well-determined
structures.” This feature makes the interpretation of a text analogous to
the performance of, say, a musical piece, since not every aspect of the
performance is determined by the score. However, while various per-
formances are mutually exclusive, in the case of two incompatible first-
order interpretations, a second-order interpretation may be offered—
averring, perhaps, that either the literary work is ambivalent or the
author is undecided. According to Dutt, interpretation presupposes a
non- or preinterpretive perception and then consists in a classification,
such as “this text is a sonnet.” It involves effort because, as Dutt un-
derstands it, interpretation deals with nonobvious states of affairs and
has to be justified by giving reasons. “Felicitous interpretation adds to
the knowledge we must have and make use of in order to get our inter-
pretive business going” (italics omitted).
Dutt exemplifies the conflict of interpretation to be settled by ar-
guments with two possible interpretations of a painting by Francesco
Maffei; Erwin Panofsky finally determined its meaning by relying on
the history of types. In the case of literary criticism, the interpretation
may focus on single words or phrases, on larger parts of the text, on
the text as a whole, or on its relation to its social and cultural context.
“The web of interpretive issues in dealing with literature is chronically
vast and full of repercussions: findings on the level of microinterpreta-
tion often affect our results on the level of macrointerpretation—and
vice versa. A single ironic word in a seemingly full-blown happy ending
novel can alter the work as a whole.” A literary critic must master a set
of concepts; their application will at the same time lead to a refinement
of the conceptual framework. Conceptual work and concrete interpre-
tation enhance each other. Dutt grants to philosophers of interpretation
such as Robert Matthews that there is a creative aspect in interpretation
that follows from the (syntactical or narrative) indeterminacy of the
 18 Vittorio Hö sle

work; nonetheless, the work has features that cannot be interpreted

away— for example, linguistic content on the level of individual pas-
sages and literary content on the level of the work as a whole. Against
radical anti-intentionalism, Dutt ultimately links the task of the inter-
preter of a text to “the intentional activity of its historically situated au-
thor,” even if he grants that this does not exclude that a work of art may
have more levels of complexity than its author was aware of.
While hermeneutics deals with the interpretation of products of
the human mind, whether present or past, historiography studies truths
about the past. According to the common understanding of the term,
it aims at past events in which humans played a role, but this is not a
necessary component of the concept of historical research, since this
also forms an aspect in the natural sciences: think of paleontology.
Aviezer Tucker, a well-known scholar of the methodology of histori-
ography, counters in his essay, “Historical Truth,” the objection that
there are no universally recognized classics of historiography that enjoy
the paradigmatic status that, say, Newton and Einstein do in physics
and that could be the starting point of normative conclusions about
what constitutes good historiography: Tucker points to figures like
Jacob Burckhardt and his theory of the Renaissance. He recognizes,
however, that most of the methodology of historians is implicit in their
practice and that there is today even a certain animus against any ex-
plicit theory of historiography: “Collective tacit knowledge is acquired
through social embedding. Explicit knowledge, by contrast, may be
transmitted via intermediaries.” All disciplines that generate historical
truth “use different information theories about the transmission of in-
formation in time via different media.” The nature of the information
transmitted varies— just think of the differences among documents,
languages, and DNA— but “the stages of inference are identical: they
first attempt to prove that the evidence is more likely given some com-
mon source of its shared preserved information than given separate
sources by finding dysfunctional homologies that are unlikely given
separate sources.” These dysfunctional homologies may be biological
rudiments or infelicities in texts. Tucker confutes the naive realistic as-
sumption that we can “see” the past. In fact, we can only infer it, and
thus “epistemology is built into the historiographic enterprise, un-
 Introduction 19

avoidable and inevitable.” Where we refuse epistemology, skeptical

consequences force themselves upon us. Tucker wants to defend both
the specific evidential basis for historiographic claims and the proba-
bilistic inference from the evidence to the fact.
The difference between most natural sciences and historiography
is that the former are interested in replicable types of events, the latter
in individual tokens such as the French Revolution. When a natural
science aims at an individual event such as the Big Bang, according to
Tucker, it becomes itself historiographical and is no longer a theoreti-
cal science. Needless to say, historical sciences, too, “are theory laden
in the sense that they are founded on information transmission theo-
ries.” Tucker points to the difference between repeated visits of the
same archival sources and replicated experiments—only in the second
case are different tokens studied— and fights the theory according to
which historiography is always built on communal narratives with a
“unity of story, storyteller, audience, and protagonist,” since this is
only one of the origins of historiography. He rejects also the doctrine
according to which historiography simply embellishes aesthetically
the facts: all such facts are the result of inferences, which therefore
have to be justified. Tucker furthermore argues against Michael Dum-
mett’s subjunctive justificationism and insists that certain facts may be
inferred today even if they could not have been observed at the rele-
vant time. “So, while the tendency in cosmic time is for information to
decay, the tendency in the much shorter modern era of historical sci-
ence has been to discover more of the past by developing new meth-
ods for extracting existing nested information.” Against all skepticism
with regard to the truth claims of historiography, Tucker points to un-
coerced consensus regarding many historical facts and the stability of
basic historical theories, a consensus that cannot be explained merely
sociologically since historians as a group are more heterogeneous than
natural scientists. Next, he distinguishes cases in which the temporal
evolution of the studied systems is identical with the transmission of
information and where this does not hold. Finally, he discusses situ-
ations in which historical truth is over- or underdetermined by the evi-
dence. In the latter case, it depends on the context and the concrete
consequences whether we are willing to speak of historical truth. “It is
 20 Vittorio Hö sle

one thing to infer from a document that somebody was a spy for the
Habsburg monarchy in the nineteenth century; it is quite another to
prove that a living person was an informer for the Communist regime
a few decades ago on a comparable evidential basis.”
The volumes on beauty and goodness included Asian perspectives
(an Indian one on aesthetics, a Chinese one on ethics and politics); this
third volume contains an analysis of a Chinese approach to a form of
knowledge particularly important for a nation that gained statehood
earlier and has maintained its political unity for longer than any other
human culture. Nicola Di Cosmo belongs to the permanent faculty of
the Princeton Institute for Advanced Study and is a world-renowned ex-
pert in both Chinese and Mongol-Manchu history. In his essay, “Truth
and Unity in Chinese Traditional Historiography,” he gives a rich over-
view of the questions with which Chinese historians struggled. De-
spite enormous changes, the various strands of Chinese historiogra-
phy “produced what is arguably the foundation of China as a unified
cultural entity.” He distinguishes two basic types in the structuring of
the historical material: while the annalist proceeds diachronically year
after year, the historian tries to group his material thematically. For
the latter, history is not simply a series of events, but, for example, the
evolution of certain institutions. Sima Qian’s Shiji, written around
100 bc, is praised as the first truly comprehensive account of Chinese
history, “the largest repository of historical knowledge in human civi-
lization,” well corresponding in its unity to that of the empire itself
and even dealing with foreign peoples. The high political and intellec-
tual stakes of the historians’ work explain why, under the Tang dynasty,
a History Office was created: the state regarded it as both its duty and
its prerogative to write the history of the preceding dynasty. From this
time on, the tension between private and public historiography forms
one of the basic dichotomies of this tradition; others deal with the
contrast between philosophical foundations and the need for objective
records, with that between the dynastic and the comprehensive mod-
els, that between the annalistic procedure and the procedure of more
elaborate narration, and that between the separation and the inclusion
of the Confucian classics from and in the realm of historiography re-
spectively. Regarding truth, the reliability of sources is a major concern
for Chinese historians. But one may doubt that they ever aimed at a
 Introduction 21

slavish report of the facts; instead, their model was that of a “ritual re-
ality” in which only that was regarded as recordable that corresponded
to established codes of conduct.
Early on, there was an awareness that, due to “appropriate con-
cealment,” certain events may have vanished from historical accounts
even if they did in fact occur. This presupposes the implicit imperative
to be “true” to events, even those that were impalatable. Clearly, dif-
ferent rules hold when the historian is not simply the interpreter of
ancient sources, but is himself witness to the facts that he reports. In
opposition to the highly bureaucratized official historiography, a sys-
tematic critique of historical writing emerges; Liu Zhiji insists on rec-
titude and passion as remedies against the alteration of truth. At the
same time, the official historiography, while often unduly influenced
by those in power, had the advantage of access to more documents. The
aim of reliability does not mean that the historiographer is not ani-
mated by a moral purpose. The great problem of Western theories of
historiography, the relation between general laws of the social world
and individual facts, is not alien to Chinese historians, who wanted to
penetrate the universal moral essence of events. But the neo-Confucian
orthodoxy that began in the Song dynasty was increasingly challenged
by the philological turn that characterizes the scholarship of the last,
the Qing dynasty. Characteristic of the work of Zhang Xuecheng is the
combination of notions of cyclicality and evolution. While the tradi-
tional subdivision of knowledge in China had separated the Classics as
eternally valid from history, he subsumes them under history because
he recognizes their historical origin and context. One is tempted to
think of radical European historicism that, in the nineteenth century,
also threatened to absorb the traditional systematic disciplines of the
Western tradition into history.
Since the first conference on beauty did not include architecture
among the arts with which it dealt, an architect was invited to the third
conference on truth to explain in which sense architecture contributes
to the unity of knowledge. Michael Lykoudis is dean of the University
of Notre Dame School of Architecture, which is committed to classi-
cal and sustainable architecture. Lykoudis begins his essay, “The Ar-
chitecture School within the University,” with the argument that ar-
chitecture combines, perhaps more than any other art, almost all forms
 22 Vittorio Hö sle

of knowledge. For the architect needs a basic understanding of mathe-

matics, engineering, physics, chemistry, and biology—he must know,
for example, which chemical interactions might contribute to or threaten
the durability of a building. He needs, furthermore, knowledge of psy-
chology, law, and business. But, of course, “most important, architec-
ture is an art.” Lykoudis returns to the old comparison of architecture
and music, famously expressed in Friedrich Schlegel’s aphorism about
architecture as frozen music: in these two arts, formal structures play a
far more important role than any mimetic relation to reality. Lykoudis
interprets architecture as mirroring the fact that the visible universe
consists of solids and voids and explains why the voids are as impor-
tant as the solids. “The pattern of the openings or voids in a wall cre-
ates rhythms that make up the composition. The changing size of these
openings adds to the grammar and syntax of architectural composi-
tions, much like notes on a piece of music. . . . The rhythm of the build-
ing can be seen in the regular cadence of openings and the melody in
the punctuating changes in size and proportion of the openings.” Co-
lonnades are said to be as much about the space between the columns as
about the columns themselves. Regarding the different types of build-
ings, Lykoudis grounds his typology in the double nature of humans as
both private and social creatures. The first aspect leads to private resi-
dences and commercial centers, the second to town halls, libraries, and
other public buildings. He adds a third category by considering sa-
cred buildings. All these buildings must fit together, and so urbanism is
added as a third scale to construction—the assembly of the elements of
a building—and architecture—the design of an individual building as
a whole. Buildings have openings toward one another, and toward the
surrounding environment, that vary in size from the small windows of
a kitchen to the huge ones of a library—not to mention arcades, which
one might interpret as porous walls.
Lykoudis does not at all deny cultural variations between the vari-
ous architectures of the world, but he believes in architectural univer-
sals that have their ultimate justification in anthropological facts: for
example, “there is a clear distinction between public, private, and sa-
cred buildings irrespective of time.” The reason that Lykoudis favors
classical architecture over modern architecture is that classical archi-
 Introduction 23

tecture is committed to the principle of equilibrium: “balance and dura-

bility are harmonized.” He pleads for a combination of traditional aes-
thetic principles with the demand for sustainability, which will impose
itself upon us with the increasing depletion of fossil fuels. His tradi-
tionalism is open to improvements: after all, tradition is “the projection
of society’s highest aspirations into the future.” Good architecture has
to adapt to the natural world that surrounds it, as does the cavea of the
Epidaurus theater to the valley in which it sits, while at the same time
alluding to myths, which elevate mere craft to architecture. The essay
ends with a list of the requirements that individual buildings as well as
cities will have to satisfy in an age of energy limits. This return to sim-
plicity, paradoxically, will help us to regain beauty—for, even if human-
kind has built more buildings since 1950 than in its entire history be-
forehand, modern cities are much uglier than those of the past.
Celia Deane-Drummond is particularly qualified to answer the
question “How Is Theology Inspired by the Sciences?” as she received
a PhD in plant physiology before she became a theologian. She begins
by pointing to the different philosophical traditions behind theology
and the natural sciences: theology is “committed to metaphysical the-
ism as a starting point, while science generally eschews theism in favor
of methodological naturalism.” At the same time, Aquinas’s concept
of theology may convince us that theology may benefit from the
sciences—and even perform its specific task better as a result of its en-
gagement with them. Deane-Drummond supports “a self-conscious
search and aspiration for unity of knowledge rather than the bolder
claim that such unity can easily be achieved.” Shared practical concerns
can mediate the search for such unity, and she chooses the issue of sus-
tainability as an area the investigation of which must be multidiscipli-
nary: both science and ethics, for example, are relevant to issues of sus-
tainability. Deane-Drummond distinguishes such fruitful cooperation
from the reductionist absorptions of one field by the other. Though the
situation was once quite the opposite, today it is biology that threat-
ens to absorb theology: Deane-Drummond discusses several “expla-
nations” of religion that are couched in evolutionary terms. Since it is
not easy to point to its selective advantages, one may interpret religion
either as an exaptation (something which is beneficial now but was not
 24 Vittorio Hö sle

when it arose) or a spandrel (something lacking selective advantage

both today and in the past). Referring to the distinction between proxi-
mate and ultimate causes introduced by the great biologist Ernst Mayr,
Deane-Drummond insists with Bonhoeffer that all scientific expla-
nations are by nature penultimate rather than ultimate, since the latter
are about God. She does not deny that evolutionary biology may suc-
cessfully explain particular traits of religion. But even outside of the
theological framework, there are many other sciences, from cogni-
tive psychology to linguistics to anthropology, that can shed light on
other features of religion. Theological “God language,” finally, is about
second-order metaphysical claims that cannot be tested empirically.
According to Karl Rahner, “theology has a better chance than another
branch of science of acting as mediator between disciplines.”
What remains common to natural science and theology is their
sense of wonder in the face of nature’s beauty and complexity. Deane-
Drummond has an ambivalent stance toward the tradition of natural
theology, so powerful in Christian theology until it was destroyed by
Charles Darwin. Her two main objections relate to the following prob-
lems: how can humans be in relation with a God who manifests himself
in mathematical laws, and how is a good God compatible “with the ap-
parently cruel and wasteful processes found in the natural world”? The
second question particularly vexes biologists, “which may be one rea-
son that theism is less common among prominent contemporary biolo-
gists compared to physicists.” Clearly, natural theology cannot prove the
existence of God: it can only try to find God in nature based on the inde-
pendent conviction of his existence. In the context of ecological praxis—
which she distinguishes from the science of ecology—normative ques-
tions inevitably enter. Despite the rejection of teleology by modern
biology, sustainability is directed toward future generations, and this
introduces a form of teleology of which theology may help to make
sense. Even if we want to repristinate an ecosystem and biology is able
to tell us how it was before the advance of industrialization, biology as
such is never in the position “to adjudicate which historical period is
the one that should be aimed at today.” For that, prudence is neces-
sary. She ends with a plea for wisdom, “that classical sense of rediscov-
ering the relationship between fields of study and . . . inclusive of a re-
 Introduction 25

lationship with God,” for only wisdom can rightly prioritize some
fields of research above others.
This volume could not address the concept of truth in every
discipline — much less its articulation in the context of every major
culture. Some disciplines are painfully missing — such as physics and
sociology; but the reflections of the chemist and of the psychologist
respectively shed some light on them, too. Some disciplines are implic-
itly covered—even if Dutt focuses on literary interpretation, he uses ex-
amples from the visual arts, and his theory of interpretation can easily be
generalized over all humanities. Even if much work remains to be done,
the volume gives an overview of how practitioners of the various sci-
ences and humanities understand the methodological challenges they
have to face in order to achieve, or at least approach, truth in their own
work. Even if the methods they use and the problems they address are
very different, their commitment to truth is the same, and perhaps this
book can strengthen the conviction that the variety of methods does
not endanger the ultimate unity of knowledge, without which an in-
stitution such as the university would hardly make sense. Though the
branches of the disciplines vary from one another considerably, they
stretch from one stem— and the differences between logical inference,
the observation of external phenomena, introspection, and interpreta-
tion correspond to the inner articulation of the one world and the prin-
ciple that ultimately explains it. Perhaps the epistemological dilemma
between intuitionism and coherentism can be addressed more in depth
by considering the inner architecture of knowledge. This, at least, was
Hegel’s central conviction, and there is much to be learned from it.
I cannot end this introduction without thanking Don Stelluto, as-
sociate director of the Notre Dame Institute for Advanced Study; Caro-
lyn Sherman, its programs administrator; and Grant Osborn, its opera-
tions coordinator, for their excellent work in the organization of this
conference and their indefatigable help in support of this volume. With-
out their dedication, efficiency, and kindness, the Institute would not
have quickly become a place where scholars, so to speak, learn to look
and sometimes even to climb from one branch of knowledge to another.
 P A R T I

-
The Historical Development of
the Tree of Knowledge
 C h a p t e r O n e

How Did the Western Culture Subdivide Its

Various Forms of Knowledge and Justify Them?
Historical Reflections on the Metamorphoses
of the Tree of Knowledge

-
Vittori o H ö s l e

Even more amazing than the sheer amount of knowledge that human-
kind has acquired in the last millennia is the heterogeneity of its nature
and its sources. Aesthetic values, moral obligations, logical inferences,
regular polygons, laws of nature, events, states, and processes connected
with atoms and molecules, stars, organisms, mental states, technical ar-
tifacts, social institutions as diverse as the family and war, signs, lan-
guages, and artworks are some of the objects we may know something
about, and as their ontological features are strikingly different, it is
reasonable to presume that at least some of the cognitive operations
connected with them vary, too. On the one hand, the increase of knowl-
edge over time suggests the possibility that perhaps already in the near
future completely different types of objects and corresponding sciences
may be added to the list that we accept today and alter it in an unpre-
dictable way. On the other hand, reductionist programs betray the desire

29
 30 Vittorio Hö sle

to subsume all this variety under some simple concept, “natural selec-
tion” being a relatively recent and plausible candidate, for it can in-
deed be applied both to the evolution of organisms and cultures.1 The
pulls of expansion of research and unification of principles drive the
cosmos of knowledge in opposite directions and reshuffle it in every
generation, and the organizational changes that the social institution
that tries to capture this cosmos, the university, has experienced and
still experiences in its history clearly reflect these pulls. Probably most
convincing is an intermediate position that recognizes an irreducible
plurality of cognitive operations and disciplines but is committed to
trying to understand why the universe of disciplines is structured as it
is. The old metaphor of the tree of knowledge, which is still used when
a discipline tries to determine its own position in the system of knowl-
edge,2 suggests that, while new branches may always grow, there are
certain constraints in such a system. For all trees have certain parts in
common, such as roots and trunks, and thus the plausibility of the
metaphor rests on the assumption that there are some constraints in
the ordering of disciplines.
In this essay, I want to pursue the question of how the subdivision
of knowledge was conceived at certain crucial moments in Western in-
tellectual history. Even before we look at the most salient subdivisions
and try to find the causes and reasons that stand behind the major
changes, it is plausible to hazard the hypothesis that two major factors
determine the way a culture organizes its knowledge. First, there is the
rise of certain disciplines in given historical epochs. The scientific revo-
lution of the seventeenth century or the historicist turn of the nine-
teenth, for example, brought forth a wealth of new disciplines, and even
if some of the objects studied by them were already known to antiq-
uity and the Middle Ages, there are good reasons to speak of the cre-
ation of a new discipline when new principles and methods are used in
relation to the same objects. For what constitutes a discipline are the
logical links between its various tenets, and when the amount of these
links increases dramatically, we should speak of a new discipline even
if its object has been thought about by humans from age immemorial.
Edmund Husserl’s famous distinction, in § 64 of the “Prolegomena”
to the Logical Investigations, between abstract, or theoretical, and con-
 How Did Western Culture Subdivide Forms of Knowledge? 31

crete, or ontological, sciences points to what I have in mind: Geogra-

phy, and natural history in general, are concrete sciences, because their
unity is constituted by the common object they address, often with
completely different methods (think of the distinction between physi-
cal geography and anthropogeography), while theoretical sciences, such
as theoretical physics, are grounded in homogeneous principles of
explanation.3
The rise of a new discipline has to do with various causes, among
which two can be easily distinguished. First, there is the discovery of
new strata of reality, be it due to journeys of exploration to parts of
the earth, or the cosmos, still unknown, be it due to new instruments
of observation, such as microscopes or telescopes. Anthropology was
triggered by the conquest of the new world,4 microbiology by the de-
velopment of microscopes in the seventeenth century.5 Second, there
is the development of new theoretical tools, often mathematical ones,
that allow us to unify certain sets of assumptions and thus to grant
theoretical status to a discipline: Modern physics would not have suc-
ceeded without calculus, which permitted, among many other things,
determining instantaneous velocities; the social sciences originated in
the nineteenth century thanks to the application of statistics to social
data by the mathematician Adolphe Quetelet in his Sur l’homme et le
développement des ses facultés, ou Essai de physique sociale (Treatise on
Man) of 1835.
But the study of the scientific revolution shows us that even before
the great discoveries by physicists such as Newton, philosophers such
as Bacon and Descartes had developed the program of the new science
partly on epistemological, partly on moral and political grounds. Thus,
it is not simply scientific discoveries that determine the various shapes
that an epoch gives to the tree of knowledge. There are, second, philo-
sophical ideas that motivate the intellectuals of a certain time to look
out for, or even to found, a new discipline. Conceptual reasons gener-
ate the feeling that there is a gap in the tree of knowledge where, with
the help of some philosophical pruning, a new discipline may blossom.
Its absence up to now may have at least three different causes: people
may have ignored the objects theorized by such a discipline (think of
subatomic particles), they may have wrongly identified them with the
 32 Vittorio Hö sle

proper topic of another discipline (the study of religions was for a long
time a province of theology), or they may have recognized their dis-
tinct existence but not have regarded their study worth their time.
Long before Richard von Krafft-Ebing’s Psychopathia sexualis (Sexual
Psychopathy) of 1886, people knew about the vast variety of human
sexual behavior, but they avoided the trouble of systematizing and
even exploring it in detail, perhaps with the exception of poets such
as Ovid. In the following historical overview, dealing in four sections
with antiquity, the Middle Ages, early modernity, and the nineteenth
century, I will move back and forth between philosophical arguments
in favor of a certain structure of knowledge and the emergence of new
disciplines as an empirical phenomenon. I do not deny at all that there
are external causes, linked to power interests, behind the rise of certain
sciences: The development of economics, for example, was obviously
fostered by the successes of a capitalist economy, and it is well known
that the study of Oriental cultures was deeply linked to French and
British imperialism.6 Still, genesis and validity belong to two different
orders, and it remains an indispensable task for philosophy to find a
place for new disciplines within the whole of knowledge. Sometimes
this attempt leads to revolutions in the architecture of epistemology
that, although triggered by external factors, usher in a conceptually
more satisfying structuring of epistemic operations.

The first comprehensive attempt to systematize human knowledge

occurred in the Academy and the Peripatos. The prehistory of art and
religion demonstrates that early on humankind must have grasped the
peculiar nature of at least animals and stars; for their depiction and di-
vinization was a recognition that they have a special place in the uni-
verse. The artistic and religious focus on such objects goes beyond the
prescientific knowledge about many objects that humankind must have
gathered from its beginnings in order to survive (I name only plants
and the human body), and most probably out of religious cults civiliza-
tions such as Mesopotamia developed the first science of humankind.
 How Did Western Culture Subdivide Forms of Knowledge? 33

I have in mind astronomy, which must have already begun with the
Sumerians and which achieved the level of precise observations in the
Old and particularly the Neo-Babylonian epochs.7 But it was the Greeks
who first proposed a systematization of knowledge with a differenti-
ation not achieved by any other culture. This is due to the fact that the
fifth century saw the rise of a discipline of so peculiar a status that it
soon triggered complex ontological and epistemological reflections.
I mean, of course, pure mathematics. The discovery of incommen-
surable magnitudes by a rigorous indirect proof, no later than 450 bc,
lifted mathematics from a useful tool helping to solve practical prob-
lems, known also to other cultures, to the level of an autonomous sci-
ence, which at the same time was used for understanding features of
the physical world. Already at the end of the fifth century, the subdi-
vision of what was later called quadrivium into arithmetic, geometry,
music, and astronomy had become standard.8 It is not clear whether
Parmenides’s epistemological and ontological revolution had any-
thing to do with mathematics (it is only his pupil Zeno who obviously
struggles with the problem of the continuum and the issue of applying
mathematics to the empirical world); but there is little doubt that Par-
menides discovered that a metaphysical principle connected with the
law of noncontradiction has a different epistemic status than the ex-
planation of nature, which is categorized as δóξα or opinion. It is not
at all necessary to interpret this opinion as fallacious,9 but it clearly is
epistemically inferior to what Parmenides calls the way of truth, and it
deals with a distinct ontological realm.
The peculiar epistemic status of mathematics as well as the onto-
logical status of its objects are two of the problems that motivate
Plato’s philosophy. The subdivision of ontological realms and episte-
mological faculties in the analogy of the line in the Republic first dis-
tinguishes between the visible and the intelligible, each of which two
sections is then itself bisected. While belief (πίστις) and imagination
(είκασία) deal with physical objects and their images, understanding
(νóησις) and thought (διάνoια) deal with the intelligible realm. What
are the differences between the last two faculties, which underlie di-
alectic, the core of philosophy, on the one hand, and concrete disci-
plines (τέχναι), such as geometry, on the other? According to Plato’s
 34 Vittorio Hö sle

elusive comments they consist of two facts: First, thought accepts its
starting points as unquestioned first principles while understanding
tries to ascend to the unhypothetical first principle. Second, thought, for
example in geometry, is obliged to use images, even if mathematicians
do deal with squares and diagonals themselves, not the ones they draw.10
What Plato means by “dialectic” is notoriously difficult to grasp, but
two things are obvious from the Charmides as well as the Euthydemus
where it is called “the kingly art.” Dialectic is the highest metadisci-
pline that determines the status of the other sciences, including itself
(for it is a reflective science);11 the epistemological considerations of the
Republic are themselves an example of this function of dialectic. And
dialectic is also knowledge of good and evil. For even if someone had
all factual knowledge, including of all future events, but lacked ethical
knowledge, he could not become happy.12 A knowledge is needed that
combines making (πoι̃ειν) and knowing how to use the things made in
the appropriate manner.13 The distinction between theoretical and prac-
tical philosophy is alien to Plato, not because he could not imagine,
but because he deliberately rejects it. Ethics and political philosophy,
so the Republic tells us, can succeed only due to an epistemological and
metaphysical turn, and theoretical philosophy has its center in the grasp-
ing of the Form of the Good.
Within the mathematical disciplines, Plato, deviating from the
order that Euclid presents in the Elements, but followed by Nico-
machus of Gerasa in his Introduction to Arithmetic from the first or
second century ad, begins with arithmetic, continues with geometry,
and then introduces stereometry, to which astronomy and harmonics
are finally added. For Plato, these disciplines increase in concreteness:
Geometry deals with two dimensions, stereometry deals with three,
while astronomy and harmonics, which are considered closely akin,
add movement.14 But Plato does not seem to acknowledge that there is a
radical epistemological difference between arithmetic and geometry (in-
cluding stereometry), on the one hand, and astronomy and harmonics,
on the other; he seems to regard all these disciplines as equally non-
empirical. Still, Plato comes much closer than Aristotle to the modern
program of a mathematical description of nature, even if he clearly un-
derrates the necessity of experiments; his dualistic epistemology seems
 How Did Western Culture Subdivide Forms of Knowledge? 35

to make it impossible to have a true science of the world that we per-

ceive.15 The deliberate connection between the Timaeus and the Critias
points to the fact that Plato somehow wants to situate humans, their
political institutions and their historical development, within nature.
But his answer is complex, since, first, humans consist of an immortal
soul that is connected with material parts, and, second, the most di-
vine human organs, head, sight, and hearing, are directly shaped by
the gods, while the other parts of the body are a result of both divine
reason and the workings of necessity.16 Plato has no doubt that the
human capacity to grasp reasons is something that transcends any
naturalistic explanation of humans.17 At the same time, he sees in the
movement of the planets and the fixed stars the most divine feature of
nature, something which in its sublimity far outstrips humans in their
folly and vices.
We know that Plato’s second successor in the Academy, Xenoc-
rates, subdivided philosophy— and that means for his time knowl-
edge in general— in a way that remained canonical for the Hellenistic
age into physical, ethical, and logical philosophy.18 In this division,
physical philosophy probably was not limited to natural philosophy
but also included metaphysics.19 Since the same subdivision can be
found in an early work of Aristotle,20 it is not implausible that it goes
back to discussions in which Plato himself participated. Logic must
have contained much more than the syllogistic that Aristotle by the
time of the Topics had not yet developed, namely, analyses of the rela-
tions between fundamental concepts, thus being not very distant from
the discipline that Plato called “dialectic” (while Aristotle will devalu-
ate this term). But far more elaborate than this subdivision is an al-
ternative, which is also incidentally mentioned in the Topics and then
presented in chapters E 1 and K 7 of the Metaphysics.21 Here Aristotle
tries to determine the place of “physics,” that is, of natural science in
general, in the whole of knowledge. He states that every thought is ei-
ther theoretical, or practical, or poietical (productive). While the dis-
tinction between theoretical and practical disciplines is still used today
(but note that Aristotle’s concept of “practical” has nothing to do with
“applied”), the difference between poiesis and praxis is peculiar to Ar-
istotle: Praxis, action, is an end-in-itself and linked to moral choice,
 36 Vittorio Hö sle

while poiesis aims at producing an external object.22 Crafts are by

nature poietical, but there is only one craft with enough philosophi-
cal dignity that Aristotle studies it in depth: poetry and its products,
namely, poems.23 Practical philosophy does not aim at knowing the
world, as theoretical philosophy does, but at actions and acquisitions.24
Since humans are by nature the most political animals, whose moral
actions can only succeed in the context of a city-state, practical philoso-
phy is also called political philosophy in the broader sense of the term.25
In the narrower sense, however, politics is distinguished from ethics.26
Several disciplines are subordinate to politics, such as strategy, eco-
nomics (which is still the doctrine of administering a household), and
rhetoric.27 Of these three, Aristotle has dedicated a separate work only
to rhetoric, which he qualifies as “an offshoot of dialectic and also of
ethical studies,” which may also be called “political.”28 That rhetoric is
a counterpart of dialectic is claimed in the first sentence of the book;
the trait common to both disciplines is that neither deals with a defi-
nite topic but can be applied to all subjects.29 While dialectic is the doc-
trine of debating, rhetoric deals with continuous speeches. It does, how-
ever, not seem possible to give the logical works of Aristotle a clear
place in the triadic system of knowledge that he himself has set forth.30
This explains why they were later called Organon, that is, instrument
for the real sciences.31 With logic and rhetoric Aristotle already had two
of the three sciences of the later so-called trivium, and when gram-
mar was added from the third century bc on, the seven artes liberales
were ready to determine for fifteen centuries the structure of European
education.32
The theoretical sciences stand higher than the other sciences, for
contemplation is the ultimate source of happiness.33 In Metaphysics E 1
and K 7, Aristotle mentions three theoretical disciplines: physics, mathe-
matics, and First Philosophy, or theology. While Aristotle acknowl-
edges that mathematics considers some of its objects as immovable
and separable from matter, he does not want to commit himself to an
independent existence of mathematical objects and in fact denies it.
Physics, on the other hand, deals with objects that both exist separately
and are movable. First Philosophy, finally, studies objects that exist sepa-
rately and are immovable. It is also called theology, since only such ob-
 How Did Western Culture Subdivide Forms of Knowledge? 37

jects can truly be called divine. And since the value of a science de-
pends on the value of its object, this will be the noblest science. I can-
not discuss here how the various tasks that Aristotle ascribes to First
Philosophy— from theology to science of being qua being—can, ac-
cording to him, indeed be contained in one science.34
Aristotle’s peculiar connection between astronomy and theology,
so obvious in Metaphysics Λ 7– 9, is based on a radical separation be-
tween the sublunar and supralunar world. The regularity of the move-
ment of the stars and their presumed immutability is contrasted with
the generation and corruption of the sublunar world. Still, Aristotle
is the founder of zoology as a science, and he goes out of his way to jus-
tify the value of this subdiscipline of physics. True enough, only as-
tronomy deals with eternal objects, but we can know much less about
them than about plants and animals. Besides its greater certitude and
completeness, biology trumps astronomy also thanks to its greater
nearness to us humans. As the true object of architecture is not bricks
but the house, so natural philosophy must look not at the material ele-
ments but at their composition, that is, at the formal and final causes
that shape them; and thus biology is the highest science of sublunar
nature.35 Psychology, too, is grounded in biology, for the soul is the
principle of animal life, and so Aristotle can claim that it belongs to
physics. The greatest part, if not all, of the affects refer to the complex
of body and soul, and thus both physicist and dialectician can ap-
proach a phenomenon like anger. The human intellect, however, does
not belong to the province of natural science.36
For those who became familiar with it, the Aristotelian systemati-
zation of knowledge remained more or less convincing up to the sev-
enteenth century. The simpler tripartition of Xenocrates, however, be-
came authoritative in Stoicism and retained its plausibility much longer:
Kant still appreciates it at the beginning of the Preface to his Ground-
work of the Metaphysics of Morals. Both Zeno37 and Chrysippus make
use of it. The latter teaches that one has to begin with logic, then pro-
ceed with ethics, and finally conclude with physics. For logic strength-
ens the mind and secures it; ethics can improve one’s manners only if it
can rely on logic, and physics is the most divine discipline and in need
of deeper foundations. It is crucial to mention that for Chrysippus
 38 Vittorio Hö sle

physics contains as its final part theology, the teaching of which is com-
pared to an initiation; this follows from Stoic pantheism and corporal-
ism.38 Since the Neoplatonists reject this corporalism and the integra-
tion of theology into physics, they insist that even in the classification
of the Platonic dialogues a fourth group has to be added: While the
Timaeus can be described as a physical dialogue, the Parmenides has to
be called a theological one.39 As in Aristotle, theology must not be sub-
sumed under physics.

II

Christian medieval philosophy is dominated by the subordination of

all knowledge to theology, to be precise: a theology that is not simply
based on reason, such as the rational theology of ancient philosophy,
but accepts scripture as its ultimate basis. Of the various encyclope-
dias of the Middle Ages,40 the most influential was doubtless the Specu-
lum Maius (The Great Mirror) by Vincent of Beauvais in three parts
(Speculum Naturale, Doctrinale, Historiale/ Mirror of Nature, Mirror
of Doctrine, Mirror of History; the Speculum Morale/Mirror of Morals
was added later). But I want to focus on two more philosophical works
that deliberately structure the cosmos of knowledge of their time ac-
cording to abstract philosophical principles. The first is De reductione
artium ad theologiam (The Reduction of the Arts to Theology) by
Bonaventure, the second the Arbor scientiae (Tree of Science) by Ramon
Llull. In his short treatise,41 Bonaventure distinguishes between me-
chanical arts (chap. 2), sensual cognition (chap. 3), philosophical cog-
nition (chap. 4), and the light of grace and Holy Scripture (chap. 5) as
the four main steps of knowledge. The first deal with artificial figure,
the second with natural form, the third with intellectual truth, and the
last with salvific truth. Regarding the mechanical arts, Bonaventure
follows Hugh of Saint Victor, who in his Didascalicon (I 5, I 11) had
added them to theoretical philosophy, practical philosophy, and logic,
a fourth part still missing in Aristotle’s tripartition, while the mechani-
cal arts somehow continue his poietic disciplines but clearly enjoy a
greater attention and respect.42 The seven mechanical arts are quite
 How Did Western Culture Subdivide Forms of Knowledge? 39

heterogeneous: spinning and weaving, forging, agriculture, hunting,

navigation, medicine, and the theatrical arts. Bonaventure tries to re-
duce them to the satisfaction of basic human needs and acknowledges
the unique position of the theatrical arts, which will later join the fine
arts; the theatrical arts are pleasing, while the others are useful, relating
to the needs of covering, feeding, or supporting oneself. In the chapter
on sensual cognition, Bonaventure connects in an arbitrary way the
five senses with the five elements; and it is in general difficult to under-
stand why the senses can be intermingled with the arts. Philosophical
cognition has three parts, each of which is itself tripartite. Rational phi-
losophy deals with the truth of speeches, natural philosophy with the
truth of things, and moral philosophy with the truth of behavior. Ra-
tional philosophy covers the trivium, that is, grammar, logic, and rheto-
ric, which are understood as the disciplines of expressing, teaching,
and motivating. Bonaventure does not see that the traditional order
conflates disciplines of different structure: he has mainly logic in mind
when he claims that rational philosophy deals with the rationes intel-
ligendi, the reasons for grasping something, and he clearly refers to
grammar and rhetoric when he claims that the first part of philosophy
governs interpretation. Only the vague concept of teaching allows
him to correlate logic with the other disciplines, but it is obvious that
all disciplines want to teach, and that there is only one discipline of
teaching, namely, pedagogy, but certainly not logic. Natural philosophy
is subdivided like Aristotle’s theoretical sciences into physics proper,
mathematics, and metaphysics; they are supposed to correspond to
matter, the soul, and divine wisdom. Practical philosophy finally ob-
tains its third part by including between individual ethics (moralis phi-
losophia monastica) and political philosophy economics, which, as in
Aristotle’s remarks in the Politics and in most of the heterogeneous
work of the Corpus Aristotelicum with the title Oikonomika (Eco-
nomics), deals with the institution of the family, not with the object of
the modern discipline. Fourth, Bonaventure discusses scripture and its
literal, allegorical, moral, and anagogical interpretation.
That this is the ultimate end of knowledge becomes even more
manifest in the second part of the treatise where the reduction of the
arts to theology announced in the title occurs. Bonaventure begins by
 40 Vittorio Hö sle

counting the three steps of philosophical cognition as of equal ranking

with the mechanical arts, sensual cognition, and the light of Holy Scrip-
ture; thus, he can move from a fourfold to a sixfold subdivision. And
why is this desirable? Because it allows him to parallelize the steps of
knowledge with the hexaemeron, that is, the six days of the creation
story of the Genesis, the knowledge to be achieved in the eschatological
glory corresponding to God’s sanctification of the seventh day (chaps. 6
and 7). The last nineteen chapters try to detect in the cognitions of the
senses, the operations of the mechanical arts, and the three parts of phi-
losophy a reflection of the same Trinitarian structure, sometimes im-
manent, sometimes economic. The wisdom of God is hidden in all cog-
nition and all nature; and in each thing that is sensed and known God
himself is internally present. Therefore all cognitions serve theology:
“omnes cognitiones famulantur theologiae” (chap. 26). Bonaventure’s
interest in these cognitions independently of their pointing to the fun-
damental principles of Christian theology is limited.
It is controversial whether Ramon Llull’s Arbor scientiae of 1296/97
was first written in Latin or in Catalan.43 Although alluding to the tree
of Genesis 3:17 and influenced by the dichotomic subdivision of re-
ality in the famous arbor porphyriana, it is, as far as I can see, the first
attempt to use the tree metaphor for encyclopedic purposes. However,
Llull does not really present, as later authors will do, a tree branching
out into various disciplines. On the contrary, despite the two singulars
in the title, Llull’s book in reality deals with a small forest, namely, six-
teen trees, fourteen of which stand for different objects and the corre-
sponding sciences. Only the last two books, the longest ones that make
up half of the work, dedicated as they are to the arbor exemplificalis and
the arbor quaestionalis, the tree of examples and the tree of question, are
of a different nature; they partly exemplify the doctrines of the work in
a literary manner, partly give hints on how to answer questions con-
nected with the contents that were exposed. But the first fourteen sci-
ences are conceived as capturing different strata of reality, an exten-
sion of the nine subjects of his famous Art.44 The arbor elementalis,
vegetalis, sensualis, imaginalis, humanalis, moralis, imperialis, apostoli-
calis, caelestialis, angelicalis, aeviternalis, maternalis, divinalis et hu-
manalis, and divinalis are dedicated to the elements, the plants, organ-
 How Did Western Culture Subdivide Forms of Knowledge? 41

isms with sense perception, organisms with the capacity of imagina-

tion, human nature, virtues and vices, political structures, ecclesiastical
hierarchies, the heavenly bodies, the angels, paradise and hell, Maria,
Jesus Christ, and God. Llull unfolds the great chain of being,45 which
has to be grasped by the corresponding sciences and which extends
from inorganic nature through different types of organisms to humans,
their moral, political, and ecclesiastical organization, up to the heav-
ens. Also for him the supralunar bodies have a higher dignity than hu-
mans, for they are connected with the angels. The last five realms be-
long to theology proper, even if Llull is one of few medieval thinkers
to embrace a radical rationalism, that is, to insist that all articles of faith
have to be proven rationally. It is noteworthy that in this classification
the distinction between theoretical and practical sciences has disap-
peared; moral philosophy is integrated into the continuum of the tra-
ditional theoretical disciplines. Furthermore, mathematics is missing,
as are the disciplines of the trivium, even if Llull wrote texts on geom-
etry, logic, and rhetoric. One explanation may be that Llull had difficul-
ties integrating the formal nature of logic and mathematics into his on-
tological scheme. Less surprising, despite Vincent of Beauvais, is the
absence of history, which was not regarded as a science.
The tree metaphor, however, is applied to each single tree, which
consists of seven parts: radices, truncus, brancae, rami, folia, flores,
fructus, that is, roots, trunk, branches, twigs, leaves, flowers, fruits.
In the case of the first tree, to give only one example, the roots are
identified with the principles, the trunk with undifferentiated corpo-
rality, the branches with the powers, the twigs with the operations,
the leaves with the accidents, the flowers with the instruments, and
the fruits with the bodies consisting of the various elements. Llull both
insists on the fundamental isomorphism between the various trees
and renders justice to the fact that the later trees presuppose the ear-
lier ones. So he writes that the arbor vegetalis has each of its roots du-
plicated, since it has both an elemental and a vegetable nature. Analo-
gously, the sensitive nature is grafted upon the vegetable and the latter
upon the elementative — here metaphors relating to the cultivation
of trees are indeed used to clarify the relation between the various
sciences.46
 42 Vittorio Hö sle

III

The most impressive attempt at systematizing knowledge in early

modernity is due to Francis Bacon, who published Of the Proficience
and Advancement of Learning, Divine and Human in 1605.47 Until
the middle of the eighteenth century it remained authoritative: d’Alem-
bert’s famous “Discours préliminaire de l’encyclopédie” (Preliminary
Discourse to the Encyclopedia) of 1751 discusses it thoroughly. Still,
the seventeenth and eighteenth centuries are among the most impor-
tant epochs in the budding of new branches in the tree of knowledge;
thus, a comparison between Bacon and the Encyclopédie promises fruit-
ful insights into the way in which the rise of new disciplines obliges
philosophy to adapt to a new reality while trying to keep as much as
possible of the basic structures that have been developed so far.
The peculiar charm of Bacon’s system of knowledge results from
his intermediate position between Renaissance erudition and the sci-
entific revolution of the seventeenth century, which he anticipates with
his methodological ideas, even if he hardly contributed to its contents.48
While the first book of the Advancement defends the excellence of
learning and the merit in its augmentation against theological and po-
litical objections as well as “the errors and imperfections of learned
men themselves” (4), the second book offers “a small globe of the in-
tellectual world” (221). Inspired by the crossing of the Atlantic Ocean
(61), Bacon does not want to limit himself to “sciences already extant
and invented” but also tries to explore “parts of learning not sufficiently
laboured and prosecuted” (63). He does not simply exhibit “a great
muster-roll of sciences,” but looks for “some seed of proficience”
(140).49 I will ignore his institutional demands, such as investing in “fun-
damental knowledges,” appropriate salaries for lecturers, funding of
experiments, continuous control of academic institutions, and interna-
tional collaboration between the European universities, and focus on
his subdivision of knowledge. Since he is committed to the “continu-
ance and entireness of knowledge,” he presents his partitions “rather for
lines and veins than for sections and separations” (105); but he claims
that they follow nature and not simply use (152 f.). Bacon, this means,
is after natural kinds, and he also avers that no area of being is left out
 How Did Western Culture Subdivide Forms of Knowledge? 43

in his scheme, “that nothing be in the globe of matter, which should

not be likewise in the globe of crystal, or form” (188). The first cut is
between human and divine learning, and while the latter is, as in the
Middle Ages, still the “sabbath of all man’s contemplations” (90, 209),
thus examined at the conclusion of the book, it is striking that far less
space is dedicated to it than to human learning. Both learnings are ini-
tially divided according to the three parts of human understanding,
memory, imagination, and reason, into history, poetry, and philosophy
(69), even if the final chapter dedicated to divinity replaces this triparti-
tion by a dichotomy dedicated to the nature of revelation and the mat-
ters revealed, the latter being subdivided into faith, manners, liturgy,
and ecclesiastical government (213, 219), that is, what today we would
call dogmatics, moral theology, liturgical science, and canonical law.
Bacon’s correlation between disciplines and faculties has the prob-
lematic consequence that different realms of being appear twice, both
in history and in philosophy. This is true not only for what we today
call humanities but also for the knowledge of nature; for Bacon subdi-
vides history into natural, civil, ecclesiastical, and literary. “Knowledges
are as pyramids, whereof history is the basis. So of natural philosophy,
the basis is natural history” (95).50 While Bacon in my eyes is right in
avoiding Wilhelm Windelband’s famous characterization of the natu-
ral sciences as nomothetical sciences and the humanities as ideographi-
cal sciences (after all, the geology of the Great Lakes deals with par-
ticulars and general linguistics with universal features of languages),
it is easy to object to Bacon that also historiography presupposes in-
ferences and the other sciences memory. It is furthermore unsatisfying
that in the cosmos of learning only poetry has its place, not, however,
its theoretical analysis. And even more puzzling for our taste is the
separation of poetry from the other fine arts, which are mentioned
only in the Latin version, and there in connection with the sensual arts
relating to the human body.51 In the subdivision of history, Bacon’s
demand for a literary history is his most original contribution: “For
no man hath propounded to himself the general state of learning to be
described and represented from age to age . . . ; without which the his-
tory of the world seemeth to me to be as the statua of Polyphemus
with his eye out; that part being wanting which doth most show the
 44 Vittorio Hö sle

spirit and life of the person” (69). Only familiarity with the history of
learning “will make learned men wise in the use and administration of
learning” (70). Bacon subdivides natural history into history of nature
in course, nature erring, and nature altered—“that is, history of crea-
tures, history of marvels, and history of arts” (70; original emphasis).
The distinction between the first two twigs presupposes a metaphysics
of nature that is still more Aristotelian than Spinozian: Bacon does not
believe that all natural events have to be explained by appealing to gen-
eral laws and antecedent events, even if he encourages rejection of
fables. (Montaigne, who lacks a metaphysics of the laws of nature, at
least recognized that even monsters, such as conjoined twins, belong
to nature and cannot occur against it: Essais II 30.) Very important is
the place that he grants to the still largely deficient mechanical arts,
which “shall be operative to the endowment and benefit of man’s life”
(72). They are supposed to be an offshoot of the marvels of nature: There
is still no idea that the engineering sciences are best built upon the natural
sciences proper. The divide in the Aristotelian Corpus between Physics
and The Mechanical Problems, that is, the treatise, probably not by Aris-
totle himself, on what happens “contrary to nature, done through art for
the advantage of humanity” (847a10 ff.), continues in Bacon (see, how-
ever, Novum Organum II 28) and up to d’Alembert.52
Bacon’s subdivision of civil history into memorials, antiquity, and
perfect history has to do with different steps of approaching the same
material, not with different areas of history. What he calls “ruminated
history,” such as Machiavelli’s Discourses on Livy, “I think more fit
to place amongst books of policy . . . than amongst books of history”
(79). Here history appears to be a preparatory stage for what we call
today the social sciences. Natural and civil history are then combined
together with mathematics into cosmography. Ecclesiastical history fi-
nally consists of history of the church, of prophecy, and of providence.
Poetry, defined as feigned history, is subdivided into narrative, repre-
sentative, and allusive (or parabolical). While the first two categories
go back to Plato,53 the third category clearly does not belong to the
same type; for one could have allusive narrations or dramas, and Bacon
hardly wants his terms to overlap. It remains surprising how long it took
for lyrics to appear as a separate subgenre of poetry. But Bacon points
 How Did Western Culture Subdivide Forms of Knowledge? 45

to the revolution occurring in the reevaluation of poetry at the end of

the eighteenth century when he recognizes that poetry for wit and
eloquence is not much inferior to orators (85), even if rhetoric belongs
to a completely different branch.
Philosophy proper is then subdivided into divine philosophy, natu-
ral philosophy, and human philosophy (85, 105). But all three branch
out from a common stem, “which hath a dimension and quantity of
entireness and continuance, before it come to discontinue and break
itself into arms and boughs.” This common science is First Philoso-
phy, which, unlike in Aristotle, is separated from theology (92). It treats
principles that apply to all special parts of philosophy. It may surprise
that Bacon mentions divine philosophy, since we already saw that di-
vine learning comes at the end of the book. But what Bacon has now
in mind is a very different discipline, namely, natural theology, based
not on revelation, but on “the contemplation of His creatures” (88),
with angelology as a legitimate appendix, even if often treated in a
“fantastical” way. Natural philosophy is composed of natural science
and natural prudence, which organizes experiments. Bacon arranges
the first into physics and metaphysics (the latter no longer identical
with First Philosophy), the former dealing with material and efficient,
the second with formal and final causes of nature. Herein similar to
Leibniz, Bacon does not at all reject a teleological interpretation of na-
ture but only as long as this does not replace the search for the effi-
cient causes (97 f.). The connection of metaphysics with formal causes
explains why Bacon regards mathematics as a branch of metaphysics,
not, as Aristotle, as a third discipline added to physics and metaphys-
ics, and he articulates it in pure and mixed mathematics: geometry and
arithmetic belong to the first, perspective, music, astronomy, and oth-
ers to the second group, the further increase of which he predicts. But
the problem remains that, since metaphysics belongs to natural philoso-
phy, an application of mathematics to the sciences of humans is excluded
by this categorization. Even less does Bacon grasp the autonomy of
mathematics as a science.
Biology, so important in Aristotle’s conception of science, is widely
absent in Bacon, certainly due to the Christian gap between humans
and the other animals. His human philosophy investigates humans first
 46 Vittorio Hö sle

“segregate,” then “congregate,” that is, as individuals and in society.

Humans as individuals consist of body and mind, and so we have a
doctrine of the “sympathies and concordances” between the two (106),
a study of the body’s health, beauty, strength, and (in a pre-Cartesian
way) pleasure by medicine, cosmetic, athletic, and “art Voluptuary,”
and, finally, a knowledge of the mind, both its substance and its func-
tions. Appendices to the first deal with what we today call parapsycho-
logical phenomena, such as extrasensory perception and telekinesis. Far
more detailed are the discussions of intellectual and volitive functions,
the objects of rational and moral philosophy. In rational philosophy the
sources of knowledge, with which Bacon started, become themselves
objects of knowledge (121); their study is called “the art of arts” (122).
Bacon distinguishes four arts: of invention, of judgment, of memory, and
of elocution. The first is further subdivided into invention of sciences—
a knowledge that “should purchase all the rest” (122)—and invention,
or rather preparation and suggestion, of arguments, that is, topics. The
doctrine of judgment is made up of analytics and elenches, which ful-
fill the functions of direction and caution respectively.54 The custody
of knowledge occurs either through writing or through memory, the
latter being directed by prenotions and emblems. Regarding the trans-
ferring of knowledge to others, Bacon distinguishes the organ, the
method, and the illustration of tradition, which roughly correspond to
the three disciplines of the trivium. While Bacon acknowledges that
method “hath been placed, and that not amiss, in Logic,” he focuses
on the tradition of knowledge, which is material to its progress, since
knowledge is an intersubjective venture (140). Two appendices deal with
critical and “pedantical,” that is, pedagogical, issues. Moral philoso-
phy ought to deal both with the nature of the good and moral culture.
The moral good may concern the individual as such or in his social
functions; and the private good is divided into active and passive good,
which itself can be distinguished into conservative and perfective, while
the doctrine of duties addresses both general duties as member of a state
and special professional duties, trying to solve possible conflicts between
the various duties. The chapter on moral culture asks for what we today
would call moral psychology and insists on a doctrine of characters and
affections; without it, morality may “make men too precise, arrogant,
 How Did Western Culture Subdivide Forms of Knowledge? 47

incompatible” (175). Civil knowledge is finally subdivided into doc-

trine of conversation, negotiation, and government, the last of which in-
cludes jurisprudence. The major objection against this categorization is
that Bacon fails to grasp the difference between psychology of thinking
and logic as well as between psychology of volition and morals. There
is no place in his empiricist framework for a radical division between
Is and Ought.
Before we look at the changes that occur in the articulation of the
Encyclopédie compared with Bacon’s text, I have to mention the two
most far-reaching mutations in the outlook of the tree of sciences that
happen in the meantime. Clearly, the Cartesian revolt is the first one.
Descartes himself continues to use the metaphor of the tree of sciences:
in the preface to the French edition of Les Principes de la Philosophie
(Principles of Philosophy), he speaks of philosophy as a tree, whose
roots are made up by metaphysics, the trunk by physics, and the main
branches, from which alone fruits can be reaped, by mechanics, medi-
cine, and morals (the perfect one, not the provisional one he himself
sketches).55 But this simile hides more than it captures his own intel-
lectual revolution, as long as one does not realize the originality of his
concept of metaphysics. Its subjects are only God and the soul. The
Aristotelian and medieval vision of the cosmos, in which the supralu-
nar world mediated between the sublunar realm, in which we live, and
God, is completely replaced by a view that in the course of the seven-
teenth century, most impressively with the Newtonian synthesis of
Galilei’s terrestrial physics and Kepler’s astronomy, tears down all pre-
sumed differences between the two realms, as it does the ancient bar-
rier between physics and mechanics. There is no peculiar ontological
status of the stars, the concept of the scale of nature thus losing at least
its traditional structure that extended beyond humans into the uni-
verse, and our machines are as natural as the other events in the world.
Matter is extension; mathematical quantities, not qualities, consti-
tute its essence. While this invites human technological activities on a
large scale, the price to be paid is a feeling of loss: The eternal silence of
these infinite spaces scared Pascal.56 The subjective center that forms
the basis of Descartes’s project, the soul with its cogito, is no longer part
of nature but is accessible in a completely different way from the rest of
 48 Vittorio Hö sle

the world— through introspection, which is limited to oneself. Nei-

ther Descartes nor his successors embrace solipsism, but the sharp dis-
tinction between first- and third-person access threatens the traditional
idea that different intellectual operations apply to different ontological
realms. Epistemologically, introspection refers only to me, while on-
tologically I do not seem to be different from other persons who access
themselves in the first person, too, even if this access is ruled out for
anyone else, including me. While the Aristotelian subdivision drew a
horizontal separation line between those mental activities that we share
with animals and the human intellect, for Descartes the basic divide is,
so to speak, a vertical line between all our mental acts and the physical
states with which they interact. Since Descartes denies subjectivity to
any other organism except humans, this vertical divide appears at a
place in nature that coincides with one of the great cuts in the tradi-
tional scale of nature. Yet for his successors, who were more generous
to the animal realm and did not want to become pan-psychists either,
the situation grew more complicated.
While the Cartesian revolution suggests a subdivision of knowledge
into theology, natural science, and a psychology based on introspec-
tion, already the early eighteenth century felt the need to integrate what
Bacon called “congregate” human philosophy into a post-Cartesian
theory of sciences. The most original work in this direction is Giam-
battista Vico’s Principj di scienza nuova (Principles of a New Science)
of 1725 (with the definitive third edition in 1744). It presents a new sci-
ence of the “world of nations,” which is conceived as a third ontologi-
cal realm beside the world of God and mind and that of nature.57 (The
conflation of the world of God and that of mind probably originates
in Vico’s sympathy for occasionalism.) While it is wrong to interpret
Vico’s philosophy of human culture, which encompasses both the hu-
manities and the social sciences, primarily as a philosophy of history,
Vico raises the intellectual level of historiography by claiming to have
found an ideal eternal history: Being connected with a universal struc-
ture, history can itself become a science. The millennia-old distinction
between history and science thus begins to fade. Vico furthermore re-
flects on the fact that the humanities had to develop later than the natu-
ral sciences— like the eye needs a mirror before it can see itself.58 The
 How Did Western Culture Subdivide Forms of Knowledge? 49

largest second book of his main work is explicitly articulated according

to the tree of sciences: The stem is made up of metaphysics, one branch
of logic, morality, economics, politics (to which reflections on history
are added), the other of physics, its daughters, cosmography and as-
tronomy, and the latter’s daughters, chronology and geography.59 While
Vico probably believes that this is the right way of systematizing knowl-
edge also for his own enlightened age,60 the second book does not give
an account of the knowledge of Vico’s time. It deals with poetical wis-
dom, and all the disciplines just mentioned are qualified as “poetical.”
What does this mean? Vico is interested in reconstructing the meta-
physics, logic, moral values, family structure, political order, and vision
of nature peculiar to the archaic time; so he engages in a hermeneutical
effort even while dealing with the second branch of his tree. In the
New Science, Vico does not deal with natural philosophy; thus, he is
no longer an encyclopedist thinker. But he aims at understanding the
encyclopedic vision earlier cultures had of the world; and in this ab-
sorption of the idea of the tree of sciences by what in the earlier archi-
tectonics had been only one discipline of it, Vico points, albeit still from
afar, to the hermeneutic imperialism of the twentieth century, which is
based on the fallacy that since every theory is a product of culture the
theory of culture encompasses everything.
The greatest project of French Enlightenment, the Encyclopédie
ou dictionnaire raisonné des sciences, des arts et des métiers (Encyclo-
pedia, or a Systematic Dictionary of the Sciences, Arts, and Crafts),
mediates already in its title between the idea of a systematic unfolding
of all disciplines for a general public (this is the original meaning of
ὲγκúκλιoς παιδεία and particularly its humanist adaptation) and the
agglomeration of knowledge by alphabetically ordered entries on all
possible topics.61 Jean Le Rond d’Alembert’s introductory “Discours”
has often been considered the most eloquent manifesto of French En-
lightenment.62 In its third and last part it integrates Denis Diderot’s
“Prospectus” of 1750; it is followed by Diderot’s “Explication détaillée
du systeme des connoissances humaines” (Detailed Explanation of
the System of Human Knowledge), accompanied by a chart, and by an
analysis of the “Systeme général de la connoissance humaine suivant le
chancelier Bacon” (General System of Human Knowledge according
 50 Vittorio Hö sle

to Chancellor Bacon), before the single entries begin.63 I cannot dis-

cuss here its many ideas, such as its sensualist epistemology influenced
by Locke and Condillac, which rejects innate ideas but preserves the
distinction between sensation and reflection and on the basis of the
latter claims to get to the ideas of God and natural law (6 ff./72 ff.); its
analysis of the origin of the various sciences that start from basic needs
but lead to concepts more and more abstract (14 ff./77 ff.); its opposi-
tion of systematic spirit, which is affirmed, and spirit of system, which
is condemned (22 f./83, 94 f./135 f.), although d’Alembert at the same
time hopes for a monist explanation of the universe as “one great
truth” (29/88); its passionate defense of the dignity and superior use-
fulness of the mechanical arts (40 ff./96 ff., 122 ff./153 ff.), to which so
many of the plates of the encyclopedia are dedicated; and the profound
reflections on the price to be paid for the loss of Latin as a common
scientific language (92 f./134). I will focus on the “genealogical or en-
cyclopedic tree which will gather the various branches of knowledge
together under a single point of view and will serve to indicate their
origin and their relationships to one another” (45 f./99). Two aspects
are crucial. First, d’Alembert is much more aware than his predeces-
sors that his subdivision is only one among many possible alternatives.
Since for him only individual objects are real, there are many equally
legitimate ways of abstracting common properties, and the tree should
not divert attention from the study of the particulars (48 f./101 f.; 58
f./109 f.). D’Alembert does not have a theory of natural kinds; thus, he
cannot answer to his own or the reader’s satisfaction why some subdivi-
sions are more plausible than others. Second, d’Alembert nevertheless
deliberately rejects a classification based on the historical development
of the sciences. For this development is discontinuous and disorderly;
in it, a foundational discipline such as logic appears historically rela-
tively late (30/89; 46 f./100). And he wants to offer with his tree “a kind
of world map,” “une espece de Mappemonde,” while the articles would
be individual maps of the countries (47 f./101).
It is in this context that d’Alembert refers to Bacon (50/102), the
inaugural figure in the series of those geniuses who are later celebrated
as the creators of modernity (74 ff./120 ff.; 87 f./130 f.). But despite his
desire to follow “the immortal Chancellor of England,” d’Alembert’s
tree of knowledge is more different from Bacon’s than he is making
 How Did Western Culture Subdivide Forms of Knowledge? 51

explicit in the “Discours,” for, as Diderot says at the end, “it would
take too long to explain” all the reasons in detail why he deviated from
his model (159/183), reasons that only philosophers can judge (164/
187 f.). Let me try to name the main differences between the two trees.
First, in d’Alembert there is no version of divine learning equivalent
to human learning. This has to do with the eighteenth century’s in-
creasing doubts both with regard to the contents of revelation and with
the phenomenon of revelation as such, doubts partly due to historical-
biblical criticism. There is a submissive allusion to revealed religion
(26/85 f.), but its importance is, as we shall see, drastically reduced.
Second, while d’Alembert accepts Bacon’s subdivision of the arts and
sciences into those based on memory, imagination, and reason, he in-
verts the order of the last two. Imagination is declared to bring together
memory and reason, and reason in its most abstract operations, meta-
physics and geometry, to become imaginative itself (51 f./103 f.). Third,
while in Bacon only poetry correlates with imagination, in d’Alembert
it has been supplemented by the other fine arts (51/103; 55/106). This is
the result of a long process, starting in the sixteenth century and com-
pleted in the eighteenth, which finally led to the separation of archi-
tecture, sculpture, painting and engraving, music and poetry from both
the mechanical arts and the sciences and their collection into a separate
group (156 f./181).64 Narrative and dramatic poetry are further subdi-
vided (144); the opera was not yet known to Bacon, and the novel had
just begun its rise. Fourth, regarding the subdivision of history and phi-
losophy, d’Alembert upholds the great chain of being and therefore is
irritated that in Bacon’s subdivision of philosophy natural philosophy
mediates between divine and human. He inverts the order that we find
in Llull and begins with God, under whom revelation teaches the exis-
tence of spiritual beings (the sentence makes it clear that d’Alembert
does not really believe in angels), then deals with man, composed of
soul and body, and finally nature.65 He thus begins with sacred his-
tory, which, as based on revelation, is distinguished from ecclesiastical
history, as relying on tradition.66 Literary history is furthermore sub-
sumed, with civil history proper, under civil history in a broader sense,
which covers “the great nations and the great geniuses” (53/104). Natu-
ral history is, as in Bacon, subdivided into uniformity of nature, devi-
ations of nature, and arts, trades, and manufactures as uses of nature,
 52 Vittorio Hö sle

which, as in Bacon, are still connected with the errors and deviations
of nature (146/172). The third part is, fifth, far more extended; the chart
at the end does not even list all trades but ends with “etc.,” because there
are supposed to exist more than 250 (144; 147 f./173). The first two parts
deal with history celestial, of meteors, of the earth and sea, of miner-
als, of vegetables, of animals, and— oddly— of elements at the end. It
is worth mentioning that celestial history (in the section on uniformity
of nature) is typographically separated from the rest; perhaps only be-
cause in the French original the other qualifications all begin with “des,”
perhaps because d’Alembert, not unlike the Greeks, has a particular ven-
eration for the magnificent spectacle that astronomy offers us (21/82).
His order seems in any case to presuppose that stars are higher in the
scale of nature than organisms, even if man, who is higher than nature,
is itself an organism.
Rational philosophy has four parts: a general, namely, ontology
or metaphysics, which includes the science of possibility (149/174),
and three particular ones, theology,67 the science of man, and that of
nature. Note that d’Alembert, sixth, has a concept of metaphysics
much closer to Aristotle than Bacon (who subordinates that discipline
to the philosophy of nature) but that, unlike Aristotle, he does not
identify it with theology. D’Alembert recognizes, probably only de
forma, both natural and revealed theology and interprets revealed the-
ology as reason applied to the facts delivered by sacred history. His
criticism of Bacon that “to separate Theology from Philosophy would
be to cut the offshoot from the trunk to which it is united by its very
nature” (54/105) is weak, for in his own system of knowledge philoso-
phy is one of three branches, not the trunk. The chart at the end men-
tions only in this specific case an abuse, namely, “superstition” (144,
149/175). Since in d’Alembert’s system man comes before nature, his
body is not mentioned under the science of man; concerning this issue,
d’Alembert proves to be a true heir of Descartes, while Bacon comes
closer to Aristotle. Regarding the mind, d’Alembert follows Bacon by
recognizing a science of the soul (pneumatology) as well as a science of
its operations, subdivided into logic and ethics. Logic has only three
parts, because the art of invention has disappeared. The art of thinking
includes apprehension of ideas, judgment of propositions, reasoning,
 How Did Western Culture Subdivide Forms of Knowledge? 53

and method; the art of remembering is not very different from Bacon’s;
the art of transmitting wisely no longer contains logic but only gram-
mar and rhetoric, with philology, criticism, and pedagogy as parts of
grammar. Ethics falls into a general and a particular part, the latter
being the science of law or jurisprudence, which is further divided into
natural, economic, and political. It is fascinating how “economic” is
now acquiring a second meaning. First d’Alembert tells us that it is the
“science of the duties of a man as a member of a family,” but since so-
cieties should be no less virtuous than individuals, he mentions also
interior and exterior commerce by land and sea (151/176). Here eco-
nomics is no longer the doctrine of the household, as it still is in Vico,
but the discipline taught today as a social science, even if Adam Smith’s
synthesis The Wealth of Nations appeared only in 1776. The rise of eco-
nomics as science presupposed that people understood that there are
laws of the social world that are irreducible to individual intentions
and sometimes even run against them, an insight to which Vico con-
tributed considerably. But only in the nineteenth century will this lead
to an emancipation of the social sciences from ethics, to which they still
belong in d’Alembert’s tree. The science of nature begins with a gen-
eral part, subdivided into a metaphysics of bodies, dealing with prop-
erties such as impenetrability, and their measurement by mathematics,
which encompasses pure, mixed, and physicomathematics, quantity
being considered either alone or in real beings or in their effects. As far
Bacon, and so differently from Plato, mathematics is not an autonomous
science but a subdiscipline of the science of nature. Needless to say,
d’Alembert increases the number of mathematical disciplines (77/122)
by adding algebra and infinitesimal calculus (152 f./177 f.), but even mili-
tary architecture and tactics are listed under pure mathematics (145).
Mixed mathematics contains mechanics, geometric astronomy, optics,
acoustics, pneumatics, and the analysis of games of chance. Particular
physics finally is structured not according to the features of the objects
studied but according to the subjective criterion of what “is worthwhile
for us to know” (55/106). In the chart, zoology, under which medicine
is subsumed, comes first and is separated from botany by physical as-
tronomy, meteorology, and cosmology, with mineralogy and chemistry
following at the end. In the “Detailed explanation,” however, zoology
 54 Vittorio Hö sle

and medicine come after astronomy, meteorology, cosmology (which

includes, among other things, geology and hydrology), mineralogy, and
botany, of which agriculture and gardening are branches. Chemistry,
however, even in his ordering, follows suit because it is “the imitator
and rival of nature” (155/180).

IV

As far as I can see, the two last great endeavors to furnish an encyclo-
pedic overview of all the sciences stem from two very different philo-
sophers, Georg Wilhelm Friedrich Hegel and Auguste Comte. Their
philosophical positions are at variance on most points: Hegel devel-
ops an alternative to concept empiricism, whereas Comte’s positivism
is committed to experience as ultimate criterion. Still, they share not
simply an encyclopedic ambition but also a rejection of Bacon’s and
d’Alembert’s attempt to found the subdivision of sciences on human
faculties.68 Instead, they want to coordinate the various sciences with
different orders of phenomena, and they want to move from the sim-
pler to the more complex ones, which presuppose the earlier ones with-
out being presupposed by them. Hegel inherits his encyclopedic inter-
ests from Immanuel Kant’s philosophical revolution. True enough,
Kant’s interests were more foundational than encyclopedic, he thor-
oughly destroyed the subdivision of knowledge used in the Leibniz-
Wolff school, and he himself corrected his vision of how philosophy
had to be subdivided in the decade from 1781 to 1790 (after all, he did
not yet dream of the Third Critique when he published the First). Still,
the encyclopedic interests of German idealism are anticipated in the
chapter “The Architectonic of Pure Reason” at the end of the First
Critique.69 The intellectual changes due to Kant are fundamentally two:
Against the empiricism of the British tradition, Kant insists on syn-
thetic a priori features of our knowledge, and he particularly claims
that ethics must have an autonomous foundation in a formal law and
cannot be reduced to a prudential doctrine of how to become happy.
It is his desire to separate sharply a priori and a posteriori elements in
human knowledge that distinguishes his classification of sciences most
 How Did Western Culture Subdivide Forms of Knowledge? 55

obviously from those of Bacon and d’Alembert. Metaphysics, accord-

ing to Kant, has been wrongly understood as a science of the first prin-
ciples. This is misleading, not only because this definition does not
give us any criterion by which to separate the first and second prin-
ciples, but also because the proper definiens of metaphysics must be its
a priori origin, not the extent of its objects.70 The third chapter within
the “Transcendental Doctrine of Method” of the Critique of Pure Rea-
son defines architectonic as the art of constructing a system. A system
is opposed to an aggregate and compared with an organism. While it
can grow, the proportions must remain the same; the lack of a single
limb will be immediately felt, and the addition of a new one will be re-
jected. Kant distinguishes furthermore between technical and archi-
tectonic unity; the former delivers a merely empirical and thus acci-
dental scheme, the second is grounded in an idea.71 The founder of a
science need not have a conscious understanding of this idea, even if
in fact he is gravitating around it. But not only is each individual sci-
ence aiming at a systematic form, all the sciences together form a sys-
tem, too, that is, unfold an underlying idea. Kant is only interested in
the system of pure reason, that is, of one of the two stems originating
from the root of knowledge.72 He then distinguishes, quite similarly to
Bacon, between historical and rational knowledge. But for Kant this
distinction has only to do with the form of subjective appropriation:
Whoever learns a philosophical system by rote and can even repeat,
like a parrot, its proofs, possesses only historical knowledge of it. The
next subdivision is between philosophy and mathematics; the first deal-
ing with concepts, the second with their construction. Philosophy it-
self is differentiated according to its scholastic and its cosmic concept.
Philosophy in the latter sense has the task to relate all knowledge to
the essential aims of human reason, whose legislator it becomes— a
task reminiscent of the kingly art of Plato’s Euthydemus. Among the
essential aims there is only one final end, namely, the destination of
man; and the philosophy dedicated to it is moral philosophy. It deals
with what ought to be and is opposed to the philosophy of nature,
which deals with what is, even if both belong to a single philosophical
system. Still, the separation of normative and descriptive disciplines,
as we could call them, is something new and very characteristic of Kant.
 56 Vittorio Hö sle

Orthogonal to this division is the distinction between pure and empiri-

cal philosophy. Kant is interested in the former, which he further sub-
divides into propaedeutic, the work of the Critiques, and metaphysics
proper, which is then divided into metaphysics of nature and meta-
physics of ethics, which must not presuppose anthropology, since the
latter belongs to empirical philosophy. Metaphysics of nature has four
parts, of which, however, only the first two offer true knowledge; but
insofar as illusory knowledge is a necessary offshoot of reason, Kant
wants to include it in his tree of knowledge; he is in this respect quite
uncommon.73 The last two branches deal with the illegitimate transcen-
dent use of reason and lead to rational cosmology and rational theology,
the first looking at an internal connection of the objects of experience,
the second at an external connection. The two legitimate branches are
ontology, dealing with objects in general, and rational physiology, deal-
ing with given objects. It is split into rational physics and rational psy-
chology: The object of the first is accessed by the external senses, the
object of the second by the internal sense.
It is not difficult to trace the Hegelian system, as exposed in the
Encyclopedia of the Philosophical Sciences in its three editions of 1817,
1827, and 1830, back to this last subdivision. Hegel’s main work does
not offer a classification of all sciences in general but only of the philo-
sophical sciences, that is, subdisciplines. But since each philosophical
science grounds the principles of a science and there is no legitimate
science that lacks such a philosophical foundation, the difference can
be neglected for our purposes. Similarly to Kant, Hegel’s science of
logic deals, among other things, with properties of objects in general,
the philosophy of nature with nature, and the philosophy of spirit
with spirit. At the same time, the differences are so far-reaching that
the analogy between Kant and Hegel can easily be overlooked. First,
Hegel does not have anything comparable to Kant’s propaedeutic part,
for he regards the attempt to study knowledge beforehand, that is, be-
fore committing oneself to knowledge, as self-contradictory.74 Second,
Hegel uses trichotomic subdivisions, while Kant prefers dichotomic
ones. As is well known, in Hegel’s dialectic the object corresponding
to the third category is supposed to return from the alienation repre-
sented by the second one to the unity of the first one, and spirit’s pe-
 How Did Western Culture Subdivide Forms of Knowledge? 57

culiar role within being is exactly this: to grasp the ideal structures that
underlie nature (§ 18). This structure allows Hegel to understand hu-
mans as very special organisms. In his ordering of sciences, he is closer
to Aristotle and Bacon than to d’Alembert, whom we saw treating the
humanities before the natural sciences. Even if Hegel does not have
any theory of evolution, it is therefore not difficult to integrate it into
his system. But at the same time, Hegel is committed to a recognition
of the special role of spirit: humans have the capacity to rise to the level
of normativity, which means that they can explicitly connect them-
selves with the logical realm. Hegel’s tripartition has some affinity to
Gottlob Frege’s distinction of thoughts, external objects, and ideas in
his essay The Thought.
Third, Hegel’s science of logic combines several quite different
functions. It pretends to analyze, like the Aristotelian Metaphysics, de-
terminations of being as such, before the latter is differentiated into
nature and spirit, but it also contains a formal logic. Indeed, Hegel can
claim to offer one of the first alternatives to the theory of psycholo-
gism developed, under this name, only in the second half of the nine-
teenth century but in fact characterizing the placement of logic both in
Bacon and d’Alembert. (In Aristotle, as we have seen, its position re-
mained unclear.) Like Husserl in the Logical Investigations, Hegel re-
jects the idea that logic is a subdiscipline of psychology: Inferences,
for example, refer to being as such and must therefore be united with
metaphysics. At the same time, concept, judgment, and syllogism are
not only categories of the science of logic (§§ 163 ff.), but reappear in
the philosophy of spirit, now, however, as the intellectual faculties and
acts grasping those structures (§ 467). The science of logic has two fur-
ther functions. It is a theory about the conditions of possibility of all
theories, including itself; thus it is a reflexive transcendental work. And
it claims to be a doctrine of the attributes of the absolute (§ 86), thus a
form of rational theology. (Since Hegel is a panentheist, all his disci-
plines are somehow dealing with God. Therefore he writes that the
science of logic deals with God’s essence before creation.)75
It may surprise that Hegel combines these four different tasks in a
single discipline, and indeed one must harbor doubts that the Hegelian
science of logic is a viable discipline. But Hegel has both good reasons
 58 Vittorio Hö sle

and illustrious predecessors for his approach. Like d’Alembert and

Kant, he has no place for revealed theology; we will see, however, that
he has a philosophy of religion, which, among other things, deals with
belief in revelation. But this philosophy of religion belongs to the phi-
losophy of spirit and is sharply distinguished from philosophical the-
ology. While Bacon and d’Alembert distinguish between First Philoso-
phy and metaphysics respectively, on the one hand, and natural theology,
on the other, Hegel returns to the Aristotelian fusion of metaphysics
and theology, which was based on the idea that God is being in its
highest form. Hegel goes further, since for him it is misleading to be-
lieve that God is one particular object beside others. God is the reason
that manifests itself in reality. Logic and transcendental theory have to
be merged with theology, too, since they are necessary presupposi-
tions for thinking in general and thinking God in particular; and these
presuppositions cannot be something external to God but must con-
stitute part of his essence.
Despite the basic trichotomic division, Hegel has also a dichotomic
subdivision of his system of philosophical sciences, namely, into logic
and the so-called Realphilosophie. For the philosophy of nature and
the philosophy of spirit agree in dealing with real objects that are all
temporal (the natural ones also spatial), while the logic deals with ideal
structures without spatial or temporal properties. In recognizing such
an ideal realm, Hegel is closer to Plato than to Bacon and d’Alembert,
even if at the same time he rejects the radical separation of Is and
Ought characteristic of Kant, since he believes that the ideal structures
permeate the real world. His philosophy of nature is subdivided into
mechanics, physics, and organic physics. Interestingly, in the first edi-
tion of the Encyclopedia, Hegel calls the first part not “mechanics’”
but “mathematics,” probably in order to cover this discipline, too, in
his encyclopedia. The change of title has probably to do with the in-
sight that it is rather arithmetic and calculus, which are dealt with in the
science of logic, that are parts of mathematics, while geometry, which
according to Hegel presupposes space and thus the first category of
nature, is not pure mathematics. Hegel does not yet grasp the distinc-
tion between mathematical and physical geometry, but his connecting
arithmetic, and not geometry, with logic can be considered a distant
 How Did Western Culture Subdivide Forms of Knowledge? 59

precursor of Frege’s attitude.76 In the second and third edition, “me-

chanics” starts with space, time, movement, inertia, and gravitation
and ends with “absolute mechanics,” the basis of astronomy. Physics
deals with more concrete phenomena, such as light, sound, electricity,
chemical processes— no doubt the most heterogeneous and least con-
vincing part of the Encyclopedia, also because the corresponding sci-
entific theories were still very much in the making. The strongest part
of the philosophy of nature is the organic physics. Hegel recognizes,
following Aristotle and the Kant of the Critique of Judgment, the pe-
culiar dignity of organisms, which he ranks much higher than the stars.77
Not only botany and zoology, but surprisingly also geology, has its
place in Hegel’s organic physics.78
It would not be correct to identify the mental realm with the ob-
ject of Hegel’s philosophy of spirit. For, on the one hand, Hegel does
not deprive animals of mentality (§ 351), though he denies that they
have spirit, and, on the other hand, his philosophy of spirit includes
social institutions: While unfamiliar with Vico’s New Science, Hegel, a
great admirer of Montesquieu, wants to integrate the disciplines of the
social world, which began to emerge in the eighteenth century, into his
encyclopedia. Hegel subdivides his philosophy of spirit into three parts:
subjective, objective, and absolute spirit. The disciplines dealing with
the first are called anthropology, phenomenology, and psychology; they
move from the soul to spirit proper, which is able to know and to will.
Hegel locates language within psychology; striking in this context is
the absence of rhetoric, whose decline began in the late eighteenth
century, largely connected with the ascent of poetry, which from Aris-
totle on had been regarded as inferior to rhetoric. Spirit proper is sub-
divided into theoretical and practical spirit (only in the third edition
does Hegel add a third part, called “the free spirit,” to maintain his gen-
eral trichotomical procedure); and the social institutions generated by
the practical (or free) will form the objective spirit. Hegel’s doctrine
of the objective spirit tries to cover several disciplines: It is a norma-
tive theory of natural law, it has elements of a theory of moral behavior,
even if at the same time it criticizes the emancipation of Kantian mo-
rality from ethical institutions, and it offers a doctrine of the main so-
cial institutions, family, civil society, and state. Particularly important
 60 Vittorio Hö sle

is the recognition of political economics as a new discipline and of

economy as a driving force of civil society (§§ 524 ff.; see Philosophy of
Right § 189). The intermediate status between a normative theory and
a descriptive theory of the social world is doubtless the main problem
of this part of the Encyclopedia. It ends with a doctrine of the absolute
spirit. This is not at all an otherworldly God: absolute spirit for Hegel
are the activities in which humans try to make sense of the meaning of
the universe, that is, art, religion, and philosophy. Hegel thus covers
the modern humanities, which study the different arts from architec-
ture to poetry, the various religions, and philosophy in its historical de-
velopment and to which the German culture in the late eighteenth cen-
tury had been particularly dedicated: I only mention Johann Joachim
Winckelmann, Johann Gottfried Herder, and the brothers August Wil-
helm and Friedrich Schlegel. While the width of these intellectual in-
terests overcame the parochial narrowness of traditional Christianity,
what distinguishes Hegel’s approach from the value-free modern hu-
manities is, however, that the absolute spirit has its grounding in the
logical structures developed in the science of logic. On this basis, Hegel
claims that Christianity is the highest religion, even if the form of reli-
gion is itself inferior to that of philosophy. History is for Hegel not an
independent discipline, since it only deals with the temporal unfold-
ing of entities that have to be grasped from a systematic point of view
before their evolution can be studied. While the Encyclopedia locates
world history at the end of the objective spirit, Hegel’s lectures discuss
in detail also the historical development of art, religion, and philoso-
phy. Hegel’s familiarity with non-European cultures, which began to
be studied from the sixteenth century on, is impressive, even if he has
no place in his system for the foundational discipline of hermeneu-
tics, which was raised to a new level by his colleague Friedrich Daniel
Ernst Schleiermacher. He also lacked a specific interest in linguistics,
which, thanks to another colleague, Franz Bopp, became able to recon-
struct features of the verbal system of the Proto-Indo-European lan-
guage and with Wilhelm von Humboldt developed a powerful categori-
zation of all human languages.
In the same year in which Hegel published the last edition of his
Encyclopedia, the first volume of Comte’s Cours de philosophie positive
 How Did Western Culture Subdivide Forms of Knowledge? 61

(Course on Positive Philosophy) appeared, which included a “tableau

synoptique,” a synoptical table detailing the lectures planned for the
following volumes as well as the underlying system of sciences. In the
first lecture, Comte develops his famous law of the three stages, accord-
ing to which the positive stage, characterized by the scientific method
and the belief in invariable laws, follows the theological and the meta-
physical stage, both in the individual and the social development. But,
alas, this development does not occur with the same speed in all disci-
plines, and while the rise of the positive method has been irresistible in
the natural sciences since the seventeenth century, the social sciences are
still in the grip of the metaphysical or even the theological stage. The ul-
timate aim of Comte’s Course is to offer a scientific social physics, but
in order to get there, we must understand its place in the tree of knowl-
edge (the metaphors of stem and branch are used often: 61, 65, 67, 77,
85). Only by understanding philosophically the nature of the other
sciences, do we have a chance to develop a truly scientific “sociology”—
this is the term that Comte coins for the new discipline (50).79 Comte
recognizes that the division of labor, so necessary for the progress of
sciences, has had adverse effects, and he wants to counter them by in-
troducing the generalist positivist philosopher as a new specialist (69).
His overview of the results and methods of the various sciences is the
only way to discover the logical laws of the human mind, which have
to be founded on observation, and on observation only by the exter-
nal sense, since Comte rejects introspection, favored by Pierre Maine
de Biran, as an invalid method. For observation of one’s own thinking
is impossible, because subject and object would coincide, and observa-
tion of one’s passions is impeded by the passions themselves (71 ff.).
Comte wants to restructure the system of education (75 ff.), hopes that
the new generalists will foster the development of the single sciences
(77 ff.), and, from bringing all sciences into line with one another, ex-
pects greater social stability (80 ff.). But how does he order, in the sec-
ond lesson, the various sciences, whose peak is sociology? The first di-
vision is between theoretical and practical endeavors, sciences and arts
(89 ff.). Note that “practical” here has a very different meaning from
that in Aristotle; it is closer to his “poietical” and the later mechanical
arts. Comte insists that the existence of human technology does not
 62 Vittorio Hö sle

challenge the invariability of natural laws, as Bacon and somehow even

d’Alembert had suggested. While defending the utility of non–applied
research, Comte recognizes as a characteristic of his time the emergence
of an intermediary class between theoretical scientists and applied art-
ists, namely, engineers. Within the sciences, Comte distinguishes be-
tween general theories aiming at laws and descriptions of individual
objects, somehow continuing the old contrast between science and his-
tory. General physiology on the one hand, zoology and botany, on the
other, are his examples (94 ff.). The subject of the Course is only the
general sciences. But how should they be ordered? While recognizing
that each classification is partly artificial (77, 98), Comte suggests as a
criterion for the ordering of their mutual dependence. Such dogmatic
classification, so Comte in agreement with d’Alembert, is not the same
as a historical one, which follows the factual rise of the various disci-
plines, even if there are some parallelisms between the two approaches
(98 ff.). The dogmatic classification must start with those phenomena
that are simpler, more general, and more abstract—an idea that already
underlay the ordering of the quadrivium (or, better, quinquivium) in
Plato’s Republic. For Comte, the first basic subdivision in reality is the
one between dead matter and organisms; but the doctrine of the first is
itself subdivided into astronomy, as the more general discipline, and
terrestrial physics, which itself consists of physics proper and chem-
istry. The doctrine of organisms is then split into physiology and be-
havior relating to conspecifics; and this second part, social physics, is
particularly important for humans. Comte ascribes four advantages to
his own classification: It corresponds to the factual subdivision of the
sciences; it matches with their historical rise as rigorous disciplines; it
explains why the earlier sciences, which depend less on others than the
succeeding ones, are more precise; and it is a canon for an appropriate
education. Only at the end does Comte reveal the greatest lacuna of his
classification: He deliberately omitted mathematics, which is not part
but the basis of natural philosophy (120). So mathematics, astronomy,
physics, chemistry, physiology, and sociology constitute his encyclo-
pedic formula. In each of them, the positivist method is modified but
within the same fundamental science (117).
It is not difficult to detect the main limits of Comte’s sketch. First,
he does not really explain the peculiar difference between mathematics
 How Did Western Culture Subdivide Forms of Knowledge? 63

and the natural sciences; and even less does he have a place in his system
for logic or his own reflection on science. Second, putting astronomy
before physics and chemistry may still have had a certain plausibility
in the early nineteenth century but is untenable after science discov-
ered the chemical processes occurring in stars. Clearly, astronomy is
less general than physics and chemistry, as geology is less general than
astronomy. Most glaring is, third, the absence of psychology. We have
seen that it is connected to the difficult methodological status of intro-
spection; the experimental psychology that was created at the end of
the nineteenth century by thinkers such as Wilhelm Wundt and Wil-
liam James would probably have passed Comte’s test and happily be
integrated into his system, as long as it remained bound to physiology
(71). His claim that the social behavior of humans is only a particular
case of animal sociality has been triumphantly verified by sociobiology;
and doubtless Comte would have welcomed the Darwinian revolution
with the same enthusiasm as Herbert Spencer, had he not already died
in 1857. Fourth, Comte does not have an equivalent of Hegel’s absolute
spirit, and his generic commitment to the positivist method does not
allow him to grasp the peculiarities of the hermeneutical method with-
out which the humanities cannot operate, as Wilhelm Dilthey well un-
derstood in his polemic against positivism. True enough, Comte’s con-
cept of sociology encompasses also the sociology of art and particularly
of religion and philosophy; but it is reductionist and does not render
justice to the specific categorical novelties that appear in this realm. The
later sociology, particularly of Max Weber, explains much more of this
realm, now relegated from absolute spirit to culture; but the price of
this transformation has been that now the social sciences are detached
from both ethics and a view of the world that still has a place for the
validity of one’s own reflection.80
This essay is intended only to offer a story of our predicament, not
a solution to our quandaries. Comte’s and Hegel’s systematizations
of our knowledge remain the most advanced efforts by encyclopedic
philosophers to make sense of the cosmos of disciplines, and while
Comte’s approach is certainly simpler and less fraught with metaphysi-
cal assumptions than Hegel’s, in my eyes we can learn more from Hegel
if we want to inject normativity back into the tree of knowledge. For
an approach such as Comte’s is fundamentally unable to explain where
 64 Vittorio Hö sle

normativity comes from (even if he later added his ridiculous religion

of humanity to solve this problem). No arborist, however, should for-
get that all work of grafting is deeply flawed that deprives the final
tree of a recognizable connection to the tree of the knowledge of good
and evil, an interaction that, once upon a time, triggered the process of
humanization.

N ot e s

1. See, on a very high intellectual level, Edward O. Wilson, Consilience:

The Unity of Knowledge (New York: Knopf, 1998).
2. See, e.g., Gregg Henriques, “The Tree of Knowledge System and
the Theoretical Unification of Psychology,” Review of General Psychology 7
(2003): 150– 82.
3. Logische Untersuchungen. Erster Band: Prolegomena zur reinen Logik
(Hamburg: Felix Meiner, 1992), 235 ff.
4. Cf. Margaret T. Hodgen, Early Anthropology in the Sixteenth and
Seventeenth Centuries (Philadelphia: University of Pennsylvania Press, 1998).
5. See Edward G. Ruestow, The Microscope in the Dutch Republic: The
Shaping of Discovery (Cambridge: Cambridge University Press, 1996).
6. See Edward Said, Orientalism (New York: Vintage, 1978).
7. On the star catalog MUL. APIN, see Rita Watson and Wayne Horo-
witz, Writing Science before the Greeks (Leiden: Brill, 2011).
8. See Plato, Protagoras, 318e.
9. See the interpretation in the long introductory essay by Luigi Ruggiu
to Parmenide, Poema sulla natura: I frammenti e le testimonianze indirette
(Milan: Rusconi, 1991).
10. Republic 509d ff. Cf. 533b.
11. Charmides 166c, 168a. I cannot justify here why I think that Plato in-
deed believes in the existence of the reflective science hypothetically discussed
in the dialogue. See my analysis in Wahrheit und Geschichte: Studien zur Struk-
tur der Philosophiegeschichte unter paradigmatischer Analyse der Entwicklung
von Parmenides bis Platon (Stuttgart–Bad Cannstatt: Frommann-Holzboog,
1984), 424 ff.
12. Charmides 174a ff.
13. Euthydemus 289b.
14. Republic 528a f., 530c f.
15. See Alfred E. Taylor, A Commentary on Plato’s Timaeus (Oxford:
Clarendon Press, 1928), still the most impressive work of scholarship on
 How Did Western Culture Subdivide Forms of Knowledge? 65

Plato’s dialogue due to its familiarity with both ancient science and Aristotle’s
philosophy.
16. Timaeus 42d ff., 69c ff.
17. See the criticism of Anaxagoras in Phaedo 97b ff.
18. Sextus Empiricus, Adversus logicos I 16 = Xenocrates, frg. 1 Heinze =
Xenocrates frg. 82 Isnardi Parente.
19. See Senocrate Ermodoro, Frammenti, ed. Margherita Isnardi Parente
(Naples: Bibliopolis, 1982), 310. She mentions several Platonic passages where
φúσις includes the world of ideas but overlooks the most important one:
Laws 892c.
20. Topics 105b20 f. See also Cicero, Academica 1,4,19.
21. Topics 145a15 f.; Metaphysics 1025b25, 1063b35 ff.
22. See Nicomachean Ethics 1140b6 f. See also Politics 1254a1 ff. on tools
as instruments of production and slaves as tools of action.
23. Even if the Greeks did not have a specific term for the fine arts, Aris-
totle is aware of the fact that poetry is only one of several mimetic arts (Poetics
1447a18 ff.).
24. Eudemian Ethics 1214a11.
25. Nicomachean Ethics 1094b11.
26. The latter term is used, for example, in Politics 1261a31 to refer to the
Nicomachean Ethics.
27. Nicomachean Ethics 1094b3.
28. Rhetoric 1356a25 ff.
29. Rhetoric 1354a1 ff., 1359b2 ff.
30. Within reasoning, Aristotle distinguishes between demonstrative, di-
alectical, and contentious (Topics 100a25 ff.). The difference between the first
two has to do with the nature of the premises, not of the inference.
31. See Diogenes Laertius V 28.
32. While Varro’s Disciplinae encompassed nine books, touching also
upon medicine and architecture, the latter are excluded as mechanical arts
from the celestial ones in Martianus Capella’s De nuptiis Philologiae et Mer-
curii; at the allegorical wedding they are silent guests. Rome’s greatest contri-
bution to the universe of knowledge is jurisprudence. While almost all human
cultures know law, i.e., enforceable social norms, only the Romans developed
a science of law.
33. Already in the Protrepticus (frg. 6), Aristotle argues that theory is
higher than poiesis, since the latter can never be an end in itself.
34. Cf. Giovanni Reale, Il concetto di “filosofia prima” e l’unità della
Metafisica di Aristotele (Milan: Bompiani, 2008).
35. On the Parts of Animals 644b22 ff.
36. On the Soul 402a4 f., 403a3–b19, On the Parts of Animals 641a22 ff.
 66 Vittorio Hö sle

37. See Stoicorum veterum fragmenta, ed. I ab Arnim (Leipzig: Teubner,

1903– 5), I 45 f.
38. Ibid., II 42 ff.
39. See Anonymous, Prolegomena to Platonic Philosophy, ed. L. G. Wes-
terink (Amsterdam: North-Holland Publishing, 1962), 47, referring to a lost
work by Iamblichus.
40. On this genre, see the two essay collections by Bernard Ribémont, De
Natura Rerum: Études sur les encyclopédies médiévales (Orléans: Paradigme,
1995); Littérature et encyclopédies du Moyen Âge (Orléans: Paradigme, 2002).
41. I use the following bilingual edition: St. Bonaventure, On the Reduc-
tion of the Arts to Theology, ed. Z. Hayes (St. Bonaventure, NY: St. Bonaven-
ture University, 1996). I ignore here Gundassalinus’s De divisione philosophiae.
42. See Franz Schupp, Geschichte der Philosophie im Überblick, 3 vols.
(Hamburg: Meiner, 2003), II 212 ff.
43. Pere Villalba Varneda, the recent editor of the work in the Corpus
Christianorum Continuatio Mediaevalis (Raimundi Lulli Opera latina 65.
Arbor Scientiae, 3 vols. [ Turnhout: Brepols, 2000]) argues in his long “In-
troductio generalis” (I 5*–188*) for a Latin original. Lola Badia, however,
thinks that the Catalan text predates the Latin (“The Arbor Scientiae: A ‘New’
Encyclopedia in the Thirteenth-Century Occitan-Catalan Cultural Context,”
in Arbor Scientiae: Der Baum des Wissens von Ramon Lull, ed. F. Domín-
guez Reboiras, P. Villalba Varneda, and P. Walter [ Turnhout: Brepols, 2002],
1–18, 2).
44. On the complex numerology of the Arbor Scientiae, see Robert Pring-
Mill, “The Role of Numbers in the Structure of the Arbor Scientiae,” in Arbor
Scientiae: Der Baum des Wissens von Ramon Lull, 35– 63.
45. On the history of this concept, see the classical study by Arthur A.
Lovejoy, The Great Chain of Being: A Study of the History of an Idea (Cam-
bridge, MA: Harvard University Press, 1936).
46. I 117 and 131. See the excellent clarifications in Anthony Bonner,
“The Structure of the Arbor Scientiae,” in Arbor Scientiae: Der Baum des Wis-
sens von Ramon Lull, 21– 34.
47. I will focus on the English original, not the expanded Latin transla-
tion integrated into the Instauratio magna, which Bacon planned to have six
parts, again in correspondence with the six works of creation. For the English
text, I use the following edition: Francis Bacon, The Advancement of Learn-
ing, ed. G. W. Kitchin (London: J. M. Dent & Sons, 1973). Only occasionally,
I point to the Latin version.
48. On predecessors of Bacon’s classification, such as Petrus Ramus, see
Stephen Gaukroger, Francis Bacon and the Transformation of Early-Modern
 How Did Western Culture Subdivide Forms of Knowledge? 67

Philosophy (Cambridge: Cambridge University Press, 2001), 18 ff. On some

encyclopedists from the sixteenth century on, see T. Frängsmyr, ed., The Struc-
ture of Knowledge: Classifications of Science and Learning since the Renais-
sance (Berkeley: University of California Press, 2001).
49. Cf. Sachiko Kusukawa, “Bacon’s Classification of Knowledge,” in
The Cambridge Companion to Bacon, ed. Markku Peltonen (Cambridge: Cam-
bridge University Press, 1996), 46– 74, 70: “Bacon’s was a unique journey
through the intellectual globe, leading to unknown territories, not a meta-
physician’s guide to a spiritual Jerusalem, or a humanist’s guide to the ideal of
classical learning.”
50. See Aristotle, Prior Analytics 46a26, on the use of ίστoρία in the sense
of a prescientific collection of facts regarding nature.
51. Francis Bacon, The Works of Francis Bacon, ed. J. Spedding, R. L.
Ellis, and D. D. Heath, 7 vols. (Boston and Cambridge: Houghton Mifflin/
Riverside, 1857– 59), II 343, where Bacon mentions that the arts concerning
sight and hearing, more than the other sensual arts, have been considered lib-
eral: “Atque artes, quae ad visum aut auditum spectant, prae aliis praecipue
liberales habitae sunt.” In the English text, architecture is mentioned as a branch
of mixed mathematics (99) and once in a simile (163).
52. A philosophy of technology, as based on the deliberate use of rare
antecedent conditions but presupposing the same universal laws of nature,
can be found in Dieter Wandschneider, Technikphilosophie (Bamberg: Buch-
ner, 2004).
53. Republic 392c ff.
54. Bacon does not follow the interpretation of the Aristotelian Orga-
non, according to which the Topics mediate between the Analytics and On So-
phistical Refutations, an interpretation suggested by the ordering of the works
in the Andronicus edition and sometimes upheld in the Middle Ages. It was
definitively rejected when Christian Brandis showed in 1833 that the Topics is
older than the Analytics.
55. Œuvres de Descartes, ed. C. Adam and P. Tannery, 13 vols. (Paris: L.
Cerf , 1897–1913), IX 2, 14 f.
56. “Le silence éternel de ces espaces infinis m’effraie” (Les Pensées de
Pascal, ed. F. Kaplan [Paris: Les Éditions du Cerf, 1982], 152 = Nr. 130).
57. See Opere di Giambattista Vico, ed. F. Nicolini, 8 vols. (Bari: Lat-
erza, 1911– 41), IV 1, 5 f. and 34. (I give also the canonical paragraph number-
ing of Nicolini: 2, 42.)
58. IV 1 95, 118 (= par. 236, 331).
59. IV 1 140 (= par. 367).
60. IV 1 217 (= par. 502).
 68 Vittorio Hö sle

61. On the basic changes to the classifications of the Encyclopédie with

regard to Bacon and Ephraim Chambers (who was not a philosophical mind),
see Robert Darnton, The Great Cat Massacre and Other Episodes in French
Cultural History (New York: Basic Books, 2009 [1984]), 191– 213: “Philoso-
phers Trim the Tree of Knowledge: The Epistemological Strategy of the Ency-
clopédie.” I particularly appreciate Darnton’s cautionary remarks that the eclec-
tic (or, as he says, even inconsistent) epistemological statements of d’Alembert
should not be interpreted as a sign that d’Alembert wanted to discredit what
he was saying (203 f.). I miss, however, in this brilliant essay a discussion of the
remarkable philosophical differences between Diderot and d’Alembert, even
if they became more evident later in their lives, and I do not consider the com-
parison between Bacon’s and d’Alembert’s systems of knowledge sufficiently
detailed.
62. The French text can be found in the beginning of the first volume
of the reprint of Encyclopédie ou dictionnaire raisonné des sciences, des arts et
des métiers (New York: Readex, 1969) but is more easily accessible in Jean
d’Alembert, Discours préliminaire des éditeurs de 1751 et articles de l’Ency-
clopédie, ed. M. Groult (Paris: Honoré Champion, 1999), which, however, does
not contain the chart. An excellent annotated English translation is Jean Le
Rond d’Alembert, Preliminary Discourse to the Encyclopedia of Diderot, ed.
R. Schwab (Chicago: University of Chicago Press, 1995). (I have corrected one
error on p. 148; the chart on p. 144 furthermore dropped “architecture.”) I will
give first the translation’s page numbers and add afterward the page number of
the Groult edition. When I give only one number, I am referring to the chart.
63. In the English version, pp. 143– 64 contain Diderot’s text; in Groult’s
edition, pp. 171– 88.
64. Cf. Paul O. Kristeller, “The Modern System of the Arts: A Study in
the History of Aesthetics,” Journal of the History of Ideas 12 (1951): 496– 527;
13 (1952): 17– 46. Note that “practical” architecture and sculpture are still
listed under the arts, trades, and manufactures (145; 148/173).
65. The inversion, however, is not complete, because vegetables and ani-
mals as well as humans and stars hold the same order in both trees.
66. In the chart at the end, the two seem to hold equal rank to civil and
natural history; but in the “Detailed explanation” they are subsumed under
sacred history in a broad sense (143/171).
67. In the “Detailed explanation” theology belongs, together with the
doctrine of angels and of demons and the science of the soul, to pneumatology
(149/174 f.). It is curious that the chart, even if by Diderot, too, is closer to
the text of the “Discours” proper. Did he draw the chart before writing his
own text?
 How Did Western Culture Subdivide Forms of Knowledge? 69

68. Comte calls these attempts “par cela seul radicalement vicieuses,”
“for this reason alone radically faulty” (Auguste Comte, Philosophie des sci-
ences, ed. J. Grange [Paris: Gallimard, 1996], 86). This edition contains only
the first two lectures of the Cours de philosophie positive, but they are the
only ones that interest me in this context. While Comte rejects a priori classifi-
cations (88), he sees his own subdivision as an alternative to a purely empirical
one (112).
69. See also the beginning of § 79 of the Critique of Judgment (B 364).
70. Critique of Pure Reason B 870 ff./A 842 ff.
71. B 861/A 833. On the opposition between technical and architectonic,
cf. B 875/A 847 and Critique of Judgment B 305.
72. Critique of Pure Reason B 863/A 835. On the stem metaphor, see B 29.
73. Cf. B 869/A 841.
74. Encyclopedia § 10 Remark. If not otherwise noted, I refer to the last
edition of 1830.
75. Introduction to the Science of Logic (Georg Wilhelm Friedrich Hegel,
Werke in zwanzig Bänden, ed. E. Moldenhauer and K. M. Michel [Frankfurt:
Suhrkamp, 1969– 71], 5:44). On the tasks of Hegel’s logic, see Vittorio Hösle,
Hegels System (Hamburg: Felix Meiner, 1987), 61 ff.
76. I refer to § 14 of The Foundations of Arithmetic (Die Grundlagen der
Arithmetik, ed. C. Thiel [Hamburg: Felix Meiner, 1988], 28 f.).
77. See the oral additions to § 268 and § 341.
78. On Hegel’s philosophy of biology, see the excellent book by Chris-
tian Spahn, Lebendiger Begriff, begriffenes Leben: Zur Grundlegung der Phi-
losophie des Organischen bei G. W. F. Hegel (Würzburg: Königshausen &
Neumann, 2007).
79. We know today that the word had already been used half a century
before in an unpublished manuscript by Emmanuel Joseph Sieyès ( Jacques
Guilhaumou, “Sieyès et le non-dit de la sociologie: du mot à la chose,” Revue
d’histoire des sciences humaines 15 [2006]: 117– 34).
80. On the development of the value-free social sciences, see my essay
“Zur Philosophie der Geschichte der Sozialwissenschaften,” now in Vittorio
Hösle, Die Philosophie und die Wissenschaften (Munich: Beck, 1999), 125– 65
and 230– 33.
 P A R T I I

-
Epistemology, Logic, and Mathematics
 C h a p t e r T w o

Intuition and Coherence in the Keystone Loop

-
Keith L e h re r

There is a traditional controversy about the role of intuition and co-

herence in justification and knowledge. Some have thought that intu-
ition is itself a source of evidence. The character of intuition and co-
herence stand in need of clarification. I shall state my own conception
of intuition and coherence without arguing these are superior to oth-
ers. One must start somewhere in philosophical discussion, making
some assumptions, and, therefore, from the perspective of some oppo-
nents the starting point begs important questions. Perhaps the best
one can hope to do is to beg some important philosophical question in
adopting a starting point, thereby focusing attention on it. I make, in
short, no claim for starting where I do. So what about the alleged role
of intuition and coherence in epistemology? The minimal answer is
that intuition is a source of evidence for what one accepts that is im-
mediate and does not depend on other things one accepts, while co-
herence is a source of evidence for what one accepts that is mediated
by and depends on other things that one accepts. Put in this way, it
seems that intuition implies that not all evidence depends on coher-
ence, and the claim that all evidence results from coherence implies
that intuition is not a source of evidence. However, the matter stated

73
 74 Keith L eh rer

in this way leaves open the possibility that the kind of evidence re-
quired for knowledge is a combination of intuition and coherence. It
may be that intuition and coherence must be joined to yield the kind
of evidence required for knowledge.
There is another way to put the matter in terms of the metaphors
of the foundation and coherence theories of knowledge. Intuition as a
source of evidence can be the basis of a foundation of knowledge be-
cause the foundation is based on intuition, a source of evidence that
does not depend on other things one accepts. A coherence theory, by
contrast, affirms that evidence for what one accepts always depends
on other things one accepts. The standard objection to a foundation
theory is that evidence, since it does not depend on anything else one
accepts, cannot be explained in terms of what one accepts, and leaves
us with an explanatory surd. The standard objection to the coherence
theory is that it leaves us with a regress or circle of the evidence be-
cause the acceptance of something always depends on something else.
It is interesting to note historically that one famous foundationalist,
Thomas Reid (1785), affirmed the role of intuition very avidly while at
the same time affirming the dependence of the principles of intuition,
which he called first principles, on each other and, notably, on one spe-
cial first principle which he says has a priority in the order of evidence
over the others. Leaving aside a good deal of detail, he affirms the in-
tuitive character of principles of our faculties, which he says are original
capacities of the mind, to yield truth and knowledge, which include con-
victions concerning consciousness, perception, memory, and others. In
each case, he says that the principles yield convictions about things that
really do or did exist. So the principles appear to simply affirm the con-
nection between conviction and reality without saying anything about
evidence. But he holds that such principles are principles of evidence,
for he affirms that evidence is the ground of belief, and first principles
are the grounds of belief. What is more interesting is that he antici-
pates the objection to his first principles as a source of evidence, namely,
that we are deceived and they are fallacious. His reply, a natural one, is
principle 7 in his first principles, which I call the First First Principle
(Lehrer 1998, 2010) though listed after others, which affirms that it is a
first principle that our faculties by which we distinguish truth from
 Intuition and Coherence in the Keystone Loop 75

error are not fallacious. He remarks that this principle is a principle of

evidence that has priority over the others, noting, of course, that our
faculties must not be fallacious or else the claims on behalf of the other
first principles of our faculties fall victim to the fallaciousness of our fac-
ulties. He compares the evidence of principle 7 to light that reveals itself
as it reveals the illuminated object. The First First Principle is a first
principle of our faculties and, therefore, vouches for itself as it vouches
for the other first principles of our faculties.
I think that this is more than a historical oddity in the philosophy
of Reid. He is candid in noting that a foundationalist, someone who
holds that intuition is a source of evidence, is assuming, whether or not
he or she makes it explicit, that some convictions have immediate evi-
dence. Without that assumption how can one defend the claims of in-
tuition and the foundational character of evidence? Putting it another
way, the foundationalist is assuming that some convictions have the
evidence of intuition, and others do not. What is the source of the evi-
dence of that assumption? Those defending intuition and foundation
theories will usually claim that evidence of intuition does not depend
on general assumptions in order to avoid the conclusion that the evi-
dence of the foundation depends on such assumptions. They seek to
avoid the claim that the evidence of intuition depends on any general
assumption about which convictions have that evidence. But this is as if
the intuitionist, having specified in general terms which convictions
have the status of intuition, says, “Shh, we won’t mention this.”
Reid was more candid. However, he left us with a puzzle about the
nature of evidence that is as simple to state as it is puzzling to solve.
Reid argued that first principles, and particular instances of them as
well, both have immediate evidence and do not obtain their evidence
from reasoning. Even more strongly, he argued that reasoning could add
nothing to the evidence of them. The evidence is immediate and neither
requires nor admits of appeal to reasoning for their evidence. Now this
leaves us with a puzzle concerning the First First Principle. It is simply
that the evidence of intuition, the evidence of first principles of our fac-
ulties, seems to depend on the First First Principle, or, if that is too
strong, to admit reasoning in favor of such by appeal to that special
principle. Reid suggests that it has a priority, that it vouches for the
 76 Keith L eh rer

other principles, and, in so doing, appears to provide a premise for rea-

soning in favor of the evidence of other principles from the First First
Principle. Indeed, it seems, as Reid’s analogy to light suggests, that the
First First Principle explains the evidence of the other principles, to
wit, that they are evidence because they are first principles of our facul-
ties and, given the evidence of the First First Principle, those faculties
by which we distinguish truth from error are not fallacious.
Reid offers one answer to the question of why the first principles
do not admit of reasoning in their favor, namely, that first principles
have all the evidence they admit of, all that they can have, and, therefore,
do not admit of an increase in their evidence as the result of reasoning.
They come into the mind evident, as he puts it in another place: their evi-
dence is their birthright. But why can’t we add to that evidence? His
answer is that they have all the evidence they admit of. What he means
becomes clear when he remarks that the evidence they have is equal to
that of an axiom of Euclid. The point is that there is a maximum degree
of evidence that any conviction can have, and the first principles, as well
as their particular instances, come into existence with that degree of evi-
dence. You cannot add evidence to something that is maximally evident
to begin with. So reasoning from the First First Principle could not in-
crease the evidence of first principles and their instances because they
are maximally evident prior to such reasoning. However, an objection
remains. The First First Principle appears to explain why the other first
principles are evident and, thereby, explains why intuition is a source
of evidence. The illuminated object would not be visible without light,
after all, and the illumination of light reveals why it is visible. Evidence
admits of explanation by reasoning from the First First Principle.
Oddly, a famous coherence theorist, Wilfrid Sellars (1963), who de-
fended a coherence theory of knowledge, was equally candid in affirm-
ing that there were some beliefs whose evidence was noninferential.
Sellars was famous for his rejection of the myth of the given, which
sounds like a rejection of intuition. He argued that the mere existence
of some sensory state could not entail knowledge of the existence of it.
His argument was that language was required for a person to have any
conception of the sensory state and, therefore, for a belief that the state
exists. The conclusion that the entailment did not hold was the simple
 Intuition and Coherence in the Keystone Loop 77

consequence of the premise that the existence of a sensory state did not
entail the existence of language. Not everyone is willing to grant the
view that conception of something required the existence of language.
However, the argument has a cogency that does not depend on that as-
sumption about language being required for conception. The argument
against the given requires only the premise that the existence of sensory
state, a pain as a salient example, does not entail that the person who
has the state has a conception of it. Sensory stimuli are one thing, a con-
ception of them another, and the first does not logically entail the sec-
ond. Sometimes the argument is formulated in terms of representation
rather than conception. It would use the premise that knowledge re-
quires some representation of the state that is not logically entailed by
the existence of the state.
The reply of a Reidian intuitionist to this argument may grant that
the evidence of intuition may presuppose conception and conviction,
as Sellars avows. He may grant that the mere existence of the sensory
state does not entail any conception or representation of the state. But
he argues that the conception and conviction that accompanies the sen-
sory state, though not entailed by it, has evidence that is noninferential.
Reid and Sellars differed concerning the source of the explanation but
agreed that it was noninferential. Reid was a nativist who argued that
conception and belief arose, in the case of first principles, from the ex-
ercise of an original faculty, by which I suggested he meant an inborn
capacity, whereas Sellars was a behaviorist who argued that conception
and belief arose from stimulus-response (S-R) conditioning. However,
a conception and belief that arises from response to a stimulus, whether
as the result of an innate faculty of response or S-R conditioning, could,
in either case, be a response that was not inferred from anything else.
It is tempting to think that any background principle, whether an in-
nate principle or a conditioned principle, that explains the response is,
therefore, a premise from which one reasons to arrive at the response
of conception and conviction. That is a mistake. Moreover, neither
the foundationalist Reid nor the coherentist Sellars made that mistake.
They understood that inference from a principle is a causal process
of a special sort, a kind of transition from one representational state
to another, while other responses are transitions from states that are
 78 Keith L eh rer

nonrepresentational, input states in contemporary vocabulary, to ones

that are representational output states. Noninferential conception and
belief is of the second variety. Not every transition from one state to
another is an inference.
Sellars was, however, adamant that the evidence of even noninfer-
ential conception and conviction depended on a background system.
Though the noninferential conviction was not generated by inference
from premises of the system, the evidence of the conviction, or as he
would put it, the justification for the conviction, depended on the back-
ground system. Here is the basis of disagreement. Sellars thought that
the background system served the goal of maximizing explanation, and
it played a crucial role in what he called the justification game of evi-
dence. Thus, the belief that arises noninferentially is justified because of
the role it plays in the background system that explains why it should
be accepted. So there is an initial contrast between Reid and Sellars that
is characteristic of the contrast between intuitionist foundation theo-
ries and coherence theories concerning evidence of noninferential or
immediate conviction. Reid thought it was the way in which they origi-
nated that accounts for their evidence, the evidence of intuition, with-
out appeal to a background system. He says they are born justified; their
justification is their birthright of intuition. Sellars, by contrast, though
he might admit that how they came into existence enters into their jus-
tification, held that is only because of a systematic explanation of why
beliefs that come into existence in that way turn out to be true. For
Reid, evidence of their truth is a birthright. For Sellars, evidence of their
truth is explained by a system.
So who is right? Notice that Reid like Sellars holds that the first
principles constitute a system. Reid says that they are joined like links
in a chain and that you cannot have the links without the chain. His
meaning is disputed, but I believe that it is the First First Principle that
contains the answer to what he means. It tells us that all the first prin-
ciples by which we distinguish truth from error are not fallacious. So if
a person like Hume, Reid’s target of criticism, accepts a first principle
of consciousness but rejects a first principle of perception, he under-
mines his evidence for one by rejecting the second, for they are equally
first principles of our faculties. By the First First Principle they stand
 Intuition and Coherence in the Keystone Loop 79

or fall together. Moreover, Reid says that no one will note his convic-
tion of first principles, including the First First Principle, until they
are challenged. So Reid like Sellars appears committed to the view that
the First First Principle plays a systematic role in the justification
game, that is, in answering challenges to evidence of truth, including
intuition for our noninferential convictions.
We have reached an issue that is not merely a historical curiosity in
the study of Reid and Sellars. Any foundationalist insisting on the role
of intuition, whether Reid or his splendid twentieth-century follower,
Chisholm (1966), confronts a question about the claim that some con-
victions have intuitive evidence of truth. Why do those beliefs and not
others have this status? If you answer the question by appeal to some
principle, how do you avoid the problem that Reid faces in our account
of him? Why is the principle that accounts for the intuitive evidence of
the belief not a principle of evidence itself? And if it is a principle of
evidence itself, why is it not a premise from which the evidence of the
foundational belief is inferred? Harman (1973) suggested that all justifi-
cation, all evidence, is inference to the best explanation. But even Sell-
ars, to whom Harman is indebted, conceded that there are noninferen-
tially justified beliefs. How can justification of all beliefs be explained
when some beliefs are noninferentially justified?
The nature of the problem can be further illuminated when the in-
tuitionist and the coherence theorist confront a skeptic about justifica-
tion of the justified beliefs. How is either to answer the skeptic’s chal-
lenge that we are not justified? If he appeals to something he accepts to
justify the convictions, for example, Reid’s First First Principle, or any
other principle, he begs the question against the skeptic. Reid says that
when confronting the total skeptic he puts his hand over his mouth in
silence. Yet the First First Principle is a principle of evidence, he says,
though he cannot prove this to the skeptic without begging the ques-
tion. Reid does not take the path of his follower, Moore (1939), who
simply insisted that he had a proof he did know what the skeptic de-
nied he knew. That would beg the question. So can we prove that the
skeptic is wrong? If to prove he is wrong is to offer an argument such
that he ought to accept the conclusion, we cannot prove that he is
wrong. Must we concede, therefore, that the skeptic is right and we are
 80 Keith L eh rer

ignorant? No. We may know that the skeptic is wrong; we can know
what we suppose we know, even though we cannot prove this to the
skeptic.
Moreover, the explanation for why we find ourselves in a situation
in which we know something we cannot prove will provide us with an
answer to the question of how the justification of all beliefs can be ex-
plained though some beliefs are noninferentially justified. The point,
which is already implicit in what has been said above, is that what is
justified by proof or inference must be distinguished from the explana-
tion of why the belief is justified. A noninferential belief may be justi-
fied, as Reid avowed, because it is the First First Principle of a faculty,
or as Sellars says, because of the role it plays in what we might call the
Explanatory System, without appeal to the principle being a premise in
reasoning or inference by which we come to have the belief. The First
First Principle, for example, is not a premise in the justification of be-
liefs that are first principles. Appeal to the First First Principle or the
Explanatory System in defense of the beliefs only arises when the be-
liefs are called into question. Moreover, when they are called into ques-
tion, the explanation for why they are noninferentially justified may be
supplied. A foundationalist like Reid and a coherence theorist like Sel-
lars may agree that beliefs of this kind that arise from clear and distinct
perceptions of an object or quality, for example, have a security from
error in the way they come into existence that provides evidence of
their truth. Once we distinguish proof and inference from explanation
and defense, we may, without claiming to prove that a challenged belief
is true, explain why some beliefs come to exist in a way that makes
their truth evident. Proof and explanation separate to account for how
we know the total skeptic is wrong and how we know some things
without proof or inference from anything else.
There is an upshot and a remaining problem. Explanation requires
truth. If we offer an explanation from a false premise, we have ex-
plained nothing. Thus, the system that explains evidence and justifica-
tion, even intuitive evidence of noninferential beliefs, must provide us
with an explanation for why the evident beliefs are true. Moreover, the
explanation must include a defense of those beliefs against objections,
for objections against the beliefs call into doubt either the truth of the
beliefs or our reasons for thinking they are true.
 Intuition and Coherence in the Keystone Loop 81

Quine (1969) argued the overall system must contain a subsystem,

a truth system, telling us when our beliefs are true, at least those we ac-
cept when we aim at obtaining truth and avoiding error in a trustwor-
thy way. As I have argued elsewhere (2012), if we aim at maximizing
explanation, then the truth system, while it explains why other beliefs
we accept are true, must explain why it itself is true. Notice that the
First First Principle, which explains why other first principles are not
fallacious, explains at the same time why it is not fallacious, that is, why
it itself is true. I have called such principles that both explain the truth
of other principles and the truth of themselves keystone principles. For
such principles that explain the truth of other principles, and thereby
support the explanatory system, also require the support of the truth of
the other beliefs, or the system will collapse. I have argued against the
metaphors of a foundation and a bootstrap in favor of the keystone.
The truth system, or some principle thereof, loops back onto itself,
which may suggest a bootstrap or a foundation, but those metaphors
are misleading. Take the First First Principle as an example, or my for-
mulation of a principle of trustworthiness that we are trustworthy in
what we accept to distinguish truth from error. Such a principle cannot
pull itself up to the level of justification alone, nor can it serve as a foun-
dation for the justification of itself and all the rest. For the principle it-
self depends on our trustworthiness in distinguishing truth from error
in the other things we accept, in Reid’s terms, on the first principles not
being fallacious, or in Sellars’s terms on the structure of systematic ex-
planation, and with it the truth system. Otherwise, if we follow Reid’s
formulation, the First First Principle will collapse in the rubble of error.
A false truth system in a system of explanation is a snare and delusion.
It explains nothing.
Now it might seem as though the connection between truth and
the Explanatory System has become so tenuous that no one should
trust it. All justification, including the evidence of intuitive truth, de-
pends on a background system whose truth system loops back onto it-
self. So how can we ensure for ourselves that our truth system is con-
nected with experience and is not just a fanciful delusion? The first
step is to distinguish the kind of knowledge that I have so far discussed,
which I (Lehrer 2000) have called discursive knowledge, from another,
simpler kind of knowledge. This simpler knowledge may be possessed
 82 Keith L eh rer

by young children, in whom neither the capacity to distinguish truth

from error nor the capacity to reply to objections to explain arriving at
truth is present. I call (Lehrer 2000) this kind of knowledge primitive
knowledge.
Sosa (2007) later distinguished animal knowledge from reflective
knowledge, but that is a different distinction. Reflective knowledge is a
kind that engages reflection, and, though discursive knowledge may
presuppose a capacity to reflect, in the activity of the justification game,
for example, it does not presuppose the engagement of the activity of
reflection. Reid introduces principle 7, the First First Principle, after
other principles of consciousness, perception, and memory without
any mention of truth or falsity. I think the reason, though this is specu-
lation, is that Reid thought that young children were helplessly gullible
to the reports of their faculties and to the testimony of others because
they lacked the distinction between truth and error. So principle 7 may
find its place in the order of the principles because conception and con-
viction may exist in a person, in young children, perhaps in animals,
who lack an understanding of the conception of truth and the ability to
evaluate the truth of what they believe in the light of evidence. They
have primitive knowledge but lack discursive knowledge and the ca-
pacity to evaluate and play the justification game, even within oneself.
The question is, what is primitive knowledge? I would like to sug-
gest that it is based on a positive attitude, perhaps belief p, or perhaps
having the impression that p. The attitude is not inferentially articu-
late, though it may imitate inference by association of the attitude to-
ward p with other contents and ideas. Basically, I think that primitive
knowledge is an automatic response to stimuli in the form of output of
representation, perhaps encapsulated as Fodor (1983) suggested, per-
haps not, but not an attitude amenable to reflection on the distinction
between truth and error. Resilient illusions of sense, the bent stick in
water, the puddle of water on the sunbaked highway, are good examples.
They are the conversion of sense to representation. Neither entailment
nor logical necessity is involved in the conversion, for it is only the con-
version of input by an innate internal program or principle to represen-
tation. Will this serve as an anchor to connect conception with truth
in a way that ensures that the truth system is anchored in experience?
 Intuition and Coherence in the Keystone Loop 83

It is inadequate. The innate response system may be the result of pro-

cesses, evolutionary tales of survival, that protect us from danger by
overstating the danger we confront. Survival selection is not likely to
lead us to a truth-refined response system. It may be better to think a
beast is bigger and faster than he is to avoid getting eaten.
We need another way to get truth into the game of justification and
even intuitive evidence to take us to truth. I have proposed an idea, one
that Reid should have held but seems not to have adopted, though what
he says leads in that direction. For Reid (1785) noted that sometimes our
conception of a sensation, a sensory experience, may be a capital part of
our conception of some external quality, for example, in the case of
smell. A conception of sensation, an odor, may lead us to a conception
of a quality in the object that gives rise to the sensation, and, initially,
that may be all there is to our conception of the quality. We have a con-
ception of a quality in the object, a stink, for example, that occasions
the sensation in us. This is his view of our initial conception of second-
ary qualities. However, since the conception of the quality is based on
our conception of a sensation, the account implies that we have a con-
ception of a sensation. How? According to Reid, from our conscious-
ness of the sensation that occasions a conception of it. So the sensation
gives rise to a conception of it, and Reid says the sensation itself is a
capital part of the conception. Now this hypothesis reminds us of the
view of Hume (1739) according to which the particular sensation be-
comes general by being used to stand for a class of sensations. That
takes us to the proposal of Goodman (1968) according to which a par-
ticular, a sample, may be used to refer to a property that exemplifies it,
a color patch of paint, for example.
I have incorporated the view suggested by remarks of Reid, Hume,
and Goodman about the use of an individual experience and the indi-
vidual qualities thereof into a notion of exemplar representation (Lehrer
2012). The experience is used as an exemplar to exhibit what it is like as
well as the plurality of things, perhaps just experiences, perhaps some-
thing beyond experiences, that the exemplar represents by exhibiting
what the represented objects are like. Now an advantage of the ac-
count, not noticed by the other authors it seems, is that exemplar rep-
resentation gives us a minimal truth security as the exemplar, which
 84 Keith L eh rer

represents a plurality of objects by showing us what the represented

objects are like, loops back onto itself as one of the objects represented.
We may think of the exemplar as a term of representation, a sensation
of pain representing pains of the kind it exhibits, for example, and thus
true of sensations it is used to represent. But it is one of the sensations
it represents exhibiting what it is like as well as what they are like, and,
therefore, the sensation true of other sensations is true of itself. The
representational loop gives us some truth connection and security. Per-
haps that is why we speak of knowing what the sensation is like, what
kind of thing it is, as a result of simply experiencing it. We say, after all,
that we experience the sensation and, therefore, know what it is like.
Exemplarizing the sensation gives rise to a conception, a representation
of the sensation, as well as other things being used to exhibit what it
represents. It is true of what it represents, including, of course, itself.
True representation is not enough for discursive knowledge, for a per-
son may lack the capacity to defend the representation, but it may suf-
fice for primitive knowledge when affirmed by the mind.
A question for Sellars scholars is whether this argument is a rejec-
tion of Sellars’s argument against the myth of the given. Have we con-
verted the sensation into a representation of the sensation in such a
way that having the sensation entails knowledge of what the sensation
is like? The connection is not entailment. When one uses an exemplar
in exemplarization to represent a class, one adds, as a contingent mat-
ter of fact, a mental operation to the sensation. Some, Kriegel (2004),
for example, hold that consciousness involves a representational loop of
the sensation back onto itself as constitutive of the conscious sensation,
which may lead to the myth of the given. That view, though tempting,
seems wrong. In an initial state of waking and other disordered forms
of consciousness, the cognitive operations have not reacted yet, in spite
of the presence of the sensation. Exemplarizing, however automatic in
some cases, adds a mental operation to a sensation, to a conscious expe-
rience. Exemplarization explains why having the sensation adds a rep-
resentation of the state that is not available before the experience of it.
You have to have a sensation in order to exemplarize it, in order to en-
gage in exemplar representation using it as a token of representation,
but having the sensation does not entail that you exemplarize it or rep-
 Intuition and Coherence in the Keystone Loop 85

resent it in any other way. So exemplarization is consistent with Sell-

ars’s attack on the myth of the given. It allows us to explain how sen-
sory experience can give rise to representation without entailing the
existence of the representation. My conclusion is that a secure truth
connection can result from a contingent operation on a sensation creat-
ing a truth loop of the sensation back onto itself yielding primitive
knowledge. This is the simplest form of exemplarization, ostensive ex-
emplar representation.
We are now in a position to understand how the keystone princi-
ples of the truth system are connected to experience. Exemplarization
draws a truth loop into the truth system. There is a similarity in the
way in which the exemplar loops back onto itself and disquotation in
sentences like the following:

RT. “Red is a color” is true in English if and only if red is a color.

RM. “Red is a color” means in English that red is a color.

They quote a sentence that is used to formulate a truth condition or a

meaning condition. Sellars noted that the sentences use the unquoted
second occurrence of “red is a color” to exhibit the role or meaning
played by the quoted use in the sentences, in other words to exhibit
what sort of semantic roles they have. However, these sentences in-
volve two tokens, different particulars, to exhibit the truth condition
and the meaning condition. The relation affirmed between the two
particulars is subject to the hazard of error that can arise in the formu-
lation of any relation between two particulars. The exemplarization of
a state avoids that hazard as it exhibits what the state itself is like; it is
the particular state itself and not another particular that represents it-
self and is true of itself. To achieve this security, exemplarization is re-
quired, which differs from exemplification which brings in another
entity, whether a property or a predicate that introduces all the possi-
bilities of error that result from relating one thing, a property or predi-
cate, to another thing, an instance of it. In the case of exemplarization,
as Ismael (2007) noted followed by Tolliver and myself (Lehrer and
Tolliver 2011; Lehrer 2011), the exemplarized exemplar is an instance
of itself reflexively.
 86 Keith L eh rer

It is of some historical interest that both Sellars and Reid seem to

have held views close to exemplarization but did not embrace it, while
Hume did so. Sellars held the interesting view that in discursive thought
our description of mental states is best understood as a theory to explain
the behavior of others in terms of postulated inner episodes. These de-
scriptors then acquire a reporting role as the terms of the theory are
applied in the first person. As Stern and I (Lehrer and Stern 2000) have
argued, it appears that Sellars, when pressed in correspondence by
Hector-Neri Castaneda to explain the reporting role, argued that there
were physical states of a special kind, sensa, which we learn to use like
elements in the disquoted sentences to exhibit what the inner episodes
are like. His model, if we have understood him correctly, is that cer-
tain internal physical states have a special feature that allows us to use
them by disquoting them to exhibit what kind of state they are. Papineau
(2002) later held this view concerning conscious states much more ex-
plicitly. Reid failed to embrace the view that the sensations are represen-
tations of themselves, though he says that they occasion conceptions of
themselves in consciousness. When I ask why he did not hold the view
that sensations are exemplarized and, therefore, a reflexive sign of them-
selves, I can only conjecture that he thought a sign must signify some-
thing other than itself, or as he puts it otherwise, a sign must suggest the
existence of something else.
Forgive me this historical diversion, but it may help to explain the
special way that I am arguing that our truth system can incorporate
experience into the stones of the keystone arch of truth and knowl-
edge rather than leave us with a mystery of how experience is related
to representation. If experience is itself representational and represents
itself as well as other things, we obtain the truth connection within the
truth system of representation without the need to explain how the
representations of the system are related to what they represent. The
exemplars are what they represent. Exemplarization closes the truth
gap between representation and what is represented, at least in the
reflexive case. Moreover, as we know what the sensation is like as we
exemplarize it, so we know what the relation of representation is like
in that instance. The knowledge of what it is like may be in the first in-
stance primitive. I am seeking in this essay and elsewhere to make it
discursive.
 Intuition and Coherence in the Keystone Loop 87

I end with a caveat and qualification. I am not claiming that exem-

plarization is the historical or genetic starting point of our conception of
the world. Actually, I am inclined to agree with Fodor (1983) that the
starting point of representation is a form of representation that is auto-
matic, perhaps encapsulated or at least protected from reflection, where
we obtain a representation of the world, the output of an input system,
without knowing what the stimulus, the input, is like. Originally, at a
very young age, the response to stimuli may be remotely analogous to
the response of my computer to compression of the keys. However, we
acquire the ability to reverse the direction of attention from what is rep-
resented to the input state evoking the representation. We thereby be-
come aware of such states and gain the capacity to exemplarize them.
Once we have a representation of the input of our sensory experience
from exemplarization, we have made a great cognitive leap. We are now
in a position to evaluate the output of the input system, the representa-
tional meaning of the output, and revise it. Most simply, we can decide
that the appearance of the bent stick in water, a representation of the
output system, is erroneous and should be reinterpreted, re-represented.
But that requires that we know what the appearance is like. We cannot
reinterpret a term without knowing what it is like, even if we can un-
consciously change how we react. So, exemplarization is an ingredient
in our cognitive autonomy. It gives us plasticity to change how we rep-
resent our world, ourselves, ourselves in our world, and our world in
ourselves. The ongoing dynamic of semantic change, of revising our sys-
tematic view of the world, including our system of truth and evidence,
aggregates vectors of exemplarized experience.
I could say more. I could tell you how the background system is
used to meet objections to how we represent the world in terms of our
exemplarized experience of it and how this converts primitive knowl-
edge of what representation is like into discursive knowledge. But that is
what I have already attempted to do in this essay. I conclude by drawing
the account of this essay into the keystone loop of discursive knowl-
edge. It is an account of the loop of intuitive evidence because exemplari-
zation is immediate, truth-reflexive, and discursive. It explains how ex-
perience loops back into the truth system of the theory. Saying more
would only widen the keystone loop that maximizes explanation. Wel-
come to the power of the loop.
 88 Keith L eh rer

R e f e re nce s

Chisholm, R. M. 1966. A Theory of Knowledge. Englewood Cliffs, NJ: Prentice-

Hall.
Fodor, J. A. 1983. The Modularity of Mind. Cambridge, MA: MIT Press.
Goodman, N. 1968. Languages of Art: An Approach to a Theory of Symbols.
Indianapolis: Bobbs-Merrill.
Harman, G. 1973. Thought. Princeton: Princeton University Press.
Hume, D. 1739. A Treatise of Human Nature. London: John Noon.
Ismael, J. 2007. The Situated Self. Oxford: Oxford University Press.
Kriegel, U. 2004. “Moore’s Paradox and the Structure of Conscious Belief.”
Erkenntnis 61: 99–121.
Lehrer, K. 1998. “Reid, Hume and Common Sense.” Reid Studies 2 (1): 15– 26.
———. 2000. “Discursive Knowledge.” Philosophy and Phenomenological
Research 60 (3): 637– 54.
———. 2010. “Reid, the Moral Faculty and First Principles.” In Reid on Ethics,
ed. S. Roeser, 25– 44. London: Blackwell.
———. 2011. “What Intentionality Is Like.” Acta Analytica 26 (1): 3–14.
———. 2012. Art, Self and Knowledge. Oxford: Oxford University Press.
Lehrer, K., and D. G. Stern. 2000. “The ‘Denoument’ of ‘Empiricism and the
Philosophy of Mind.’” History of Philosophy Quarterly 17 (2): 201–16.
Lehrer, K., and J. Tolliver. 2011. “Truth and Tropes.” In Philosophical Papers
Dedicated to Kevin Mulligan, ed. A. Reboul. www.philosophie.ch/kevin
/festschrift.
Moore, G. E. 1939. “Proof of an External World.” Proceedings of the British
Academy 25: 273– 300.
Papineau, D. 2002. Thinking about Consciousness. Oxford: Oxford Univer-
sity Press.
Quine, W. V. O. 1969. “Epistemology Naturalized.” In Ontological Relativity
and Other Essays. New York: Columbia University Press.
Reid, T. 1785. Essays on the Intellectual Powers of Man. Edinburgh.
Sellars, W. S. 1963. “Empiricism and the Philosophy of Mind.” In Science, Per-
ception, and Reality. New York: Humanities Press.
Sosa, E. 2007. A Virtue Epistemology: Apt Belief and Reflective Knowledge.
Vol. 1. Oxford: Oxford University Press.
 C h a p t e r T h r e e

What Is the Nature of Inference?

-
Rober t H anna

A general but pure logic . . . has to do with strictly a priori principles,

and is a canon of the understanding and reason, but only with re-
gard to what is formal in their use, be the content what it may (em-
pirical or transcendental). A general logic, however, is then called
applied if it is directed to the rule of the use of the understanding under
the subjective empirical conditions that psychology teaches us. . . .
[G]eneral and pure logic is related to [applied logic] as pure morality,
which contains merely the necessary moral laws of a free will in gen-
eral, is related to the doctrine of virtue proper, which assesses these
laws under the hindrances of the feelings, inclinations, and passions
to which human beings are more or less subject, and which can never
yield a true and proven science, since it requires empirical and psy-
chological principles just as much as that applied logic does.
—I. Kant, Critique of Pure Reason

What is the nature of inference? In order to answer that question, we

must first have in front of us a commonsensical or intuitive, more or
less theoretically neutral, working definition of what it is we are talk-
ing about. By “inference,” then, I mean a cognitive process leading from
the mental representation of the premises of a deductive, inductive, or

89
 90 Rober t Hanna

abductive (abduction = inference-to-the-best-explanation) argument1 to

the mental representation of the conclusion of that argument, where the
cognitive transition from the representation of the premises to the rep-
resentation of the conclusion is governed by some rule-based stan-
dards of cogency, such that if all the premises are believed by a cognizer
or cognizers and if the cognitive transition from representing the prem-
ises to representing the conclusion is also believed by that cognizer or
those cognizers to be cogent, then, other things being equal,2 the con-
clusion will also be believed by that cognizer or those cognizers.
I am also assuming that an inference can have many further im-
portant properties, but I am also thinking that this will suffice as an in-
tuitive working definition of inference in a minimal sense.
The apparently univocal philosophical question, What is the na-
ture of inference?, however, is in fact four distinct yet intimately related
philosophical questions all rolled up into one:

(1) what are the metaphysics of inference?

(2) what is the goal or purpose of inference?
(3) what justifies an inference?
(4) what is the mechanism of inference?

In order to answer these questions, I will propose what I call “Con-

temporary Kantian Moralism about Inference,” or CKMI. In a nut-
shell, according to CKMI, inference is essentially an act of cognitive
free agency. By this, I do not mean that rational human cognizers can
choose to regard any inference whatsoever as cogent (inferential vol-
untarism): on the contrary, what I mean is that rational human cogniz-
ers can freely choose to make the inferences they rationally ought to
make; they can also freely fall short of inferring as they rationally ought
to, and thereby freely commit inferential errors; and in either case, they
are cognitively responsible for their inferences.
CKMI differs substantially from other theories of inference cur-
rently on offer, first, in its being robustly mentalistic about the meta-
physics of inference, and neither platonistic nor psychologistic; second,
in its being robustly reasons-internalist about the rational normativity
of inference, and neither emotivist nor instrumentalist; third, in its being
robustly practical about the justification of inference, although non-
 What Is the Nature of Inference? 91

voluntaristically practical, and neither noncognitivist nor holist nor in-

ferentialist; and fourth, in its being robustly cognitivist and phenome-
nological about the mechanism of inference, and neither functionalist
nor separatist.

Th e M e tap h y s ics of Inference:

Trans ce nde ntal Me ntalism, and
N e i t h e r Psy ch o l ogism nor Platonism

As I noted at the beginning, an inference in the minimal sense is a cog-

nitive process of a certain kind. More precisely, however, the cogni-
tive process of inference is an intentional or representationally object-
directed or propositional-directed process, with intrinsic “aboutness.”
The metaphysics of inference is therefore a subspecies of the meta-
physics of intentionality.
Psychologism about the metaphysics of inference says that infer-
ential facts and inferential laws are fully explicable by the natural sci-
ences together with the formal sciences, including mathematics, and
therefore explanatorily reducible to physical, chemical, or biological
facts, plus their mathematical properties.3
The basic problems with Psychologism about the metaphysics of
inference are that Psychologism cannot account for (i) the manifest ob-
jectivity, (ii) the manifest necessity, and (iii) the manifest apriority of in-
ferential facts. One could try to defend Psychologism by trying to re-
ject the very ideas of the objectivity, necessity, and apriority of logic,
but it is hard to see how one could do so without relying on deductive,
inductive, and abductive principles that are themselves tacitly assumed
to be objective, necessary, and a priori, without self-stultification.
Platonism about the metaphysics of inference says that inferential
facts and inferential laws are non-natural in the strong sense of being
abstract, non-spatiotemporal, nonexperiential, and noncontingent.4
The basic problems with Platonism about the metaphysics of in-
ference are (i) that the non-spatiotemporality and causal inertness of
Platonic facts and objects are clearly incompatible with the causal trig-
gering of all human knowledge (a.k.a. “Benacerraf’s Dilemma”),5 and
(ii) Platonism cannot explain how the truth-making connection between
its non-spatiotemporal, acausal truth makers and the truth of inferential
 92 Rober t Hanna

propositions and inferential beliefs is anything other than a mas-

sive coincidence (this is what I will call “the Problem of Cognitive-
Semantic Luck”).6
By sharp contrast to Psychologism and Platonism alike about the
metaphysics of inference, according to CKMI, the mentalistic meta-
physics of inference are those of what I call “weak or counterfactual
transcendental idealism,” or WCTI. Now, in a nutshell, WCTI about
the metaphysics of inference says that, first, necessarily, inferential facts
are also facts about rational animal minds whose cognition necessarily
begins in causally triggered sensory experience but without either its
content or its form being either reducible to or strictly determined by
causally triggered experience; second, necessarily, inferential facts in
the manifest natural world structurally conform to the nonempirical
or a priori forms of those mentalistic facts but without being identical
to those forms of mentalistic facts; and third, necessarily, if inferential
facts exist, then if there were to be rational human minds, they would
be able to cognize those worldly inferential facts directly, at least to
some extent.7
It follows from these three theses that, according to CKMI, an
inference is a naturalistically irreducible cognitive process in rational
animal agents or persons, whose cognition always begins in causally
triggered sensory experience but is strictly underdetermined by that or
any other actual or possible causal sensory triggering, such that there is
necessarily a nonaccidental or intrinsic truth-making relation between
the inferential facts in the manifest world, on the one hand, and the
truth of inferential propositions and inferential beliefs cognized by
those rational animals agents or persons, on the other. This view both
neatly avoids Benacerraf’s Dilemma and also solves the Problem of
Cognitive-Semantic Luck.

Th e Rat i o nal N o rm at iv it y of Inference:

S e l f- Trans ce ndi ng S econd-Order Caring
for t h e Sake o f t h e S um m um Bonum of Reasoning,
and N e i t h e r E m ot i v i s m nor Instrumentalism

Issues concerning the goal or purpose of inference are issues about the
normativity of inference. Normativity, in turn, has specifically to do
 What Is the Nature of Inference? 93

with motivation and action guidance, and with the ideals, standards,
rules, principles, and laws governing intentional choices, actions, pro-
cedures, and practices.
Emotivism about the normativity of inference says that the evalu-
ative content of inferences is not itself truth-apt, or truth-evaluable, and
consists instead exclusively in our pro attitudes and contra attitudes to-
ward inferences, and is strictly determined by those attitudes.
The basic problem with Emotivism about the normativity of infer-
ence is that it posits pro attitudes or contra attitudes that are essentially
unconstrained by rational norms of consistency, truth, logical conse-
quence, or soundness: in effect, anything goes, provided that everyone
shares the same feelings. So the problem is antirational arbitrariness. A
particularly pointed and reflexive version of the problem of antirational
arbitrariness arises when one applies Emotivism to one’s own inferen-
tial practices from the outside in: Do I really think that the cogency of
my own inferences should be held hostage to arbitrary pro attitudes or
contra attitudes?
Instrumentalism (a.k.a. “pragmatism”) about the normativity of
inference says that the evaluative content of inferences consists exclu-
sively in and is strictly determined by the good or bad results, from
the standpoint of human interests in either a narrowly self-oriented or
a larger social sense, that are produced by inferences.
The basic problem with Instrumentalism about the normativity of
inference is that it allows for the partial or total sacrifice of consistency,
truth, logical consequence, and soundness if good consequences will
ensue or bad consequences are avoided: in effect, anything goes, pro-
vided that good results are produced and bad consequences avoided
from the standpoint of human interests in either a narrowly self-oriented
or larger social sense. So, again, the problem is antirational arbitrariness.
As with Emotivism, a particularly pointed and reflexive version of the
problem of antirational arbitrariness arises when one applies Instru-
mentalism to one’s own inferential practices from the outside in: Do I
really think that the cogency of my own inferences should be held hos-
tage to the mere production of good or bad results?
By sharp contrast to Emotivism and Instrumentalism alike about
the normativity of inference, according to CKMI, an inference is aimed
at the classical soundness of an argument (soundness = true premises
 94 Rober t Hanna

and semantic consequence or validity, thereby guaranteeing the truth

of the conclusion) together with authentic knowledge (authentic
knowledge = justified true belief, such that there is an intrinsic connec-
tion between the evidence that justifies belief and the truth maker of the
propositional content of that belief ) of its conclusion, as the highest
or supreme good (summum bonum) of reasoning, but is also strictly
constrained by minimal consistency (minimal consistency = not every
proposition or statement is both true and false), in such a way that it is
underdetermined by any instrumental reasons that may also support it.
Now Internalism about reasons grounds motivation and action
guidance on desire-based conation and evaluation.8 In this way, ac-
cording to CKMI, the normativity of inference involves a specifically
Kantian version of reasons-internalism that posits an innate conative
capacity to desire self-transcendence, and in particular an innate ca-
pacity for having second-order desires to be moved by effective first-
order desires that are nonegoistic, nonselfish, nonhedonistic, and non-
consequentialist, according to and for the sake of the summum bonum
of reasoning in particular and the summum bonum of morality in gen-
eral.9 In short, the theory of inference properly falls under what Kant
calls the metaphysics of morals.

Th e Jus t i ficat i o n of Inference:

Cat e go ri cal L e gi t i m at ion, and Neither
Non- Co gni t i v i s m , no r H ol ism, nor Inferentialism

The justification of belief, choice, or action is appealing to (what pur-

port to be) sufficient reasons for holding some belief, for heeding or
following some principle, for making some choice, or for carrying out
some action.
A fundamental problem for any attempt to justify inference is that
some or all of the specific deductive, inductive, or abductive inferen-
tial principles that are being justified must also be presupposed and
used in the justification of those very principles. So, it seems, either
the inferential principles are unjustified or else the purported justifica-
tion fails because it is viciously circular.
 What Is the Nature of Inference? 95

One way out of this “logocentric predicament”10 with respect to

inference is to hold that the inferential principles have what Hartry
Field calls “default reasonableness,” in that we are rationally entitled
to presuppose and use them in the absence of any sufficient reason not
to.11 So the inferential principles do not need to be justified.
The entirely reasonable question then arises, What is the ground
or source of this nonjustificatory rational entitlement?
Field himself holds that the nonjustificatory default-reasonable
entitlement to inferential principles does not require a further ground
or source, that there is no deeper fact of the matter, and that the entitle-
ment merely reflects our strong pro attitudes toward the inferential
practices we are already engaged in. That is the noncognitivist (a.k.a.
“nonfactualist”) strategy.
Others hold that we are default-reasonably entitled to the pre-
supposition and use of these inferential principles by the smooth fit or
“reflective equilibrium” that gradually emerges over time between our
own inferences insofar as they are guided by these principles, our in-
tersubjective agreement about them, and other judgments about the
world made by ourselves and others.12 That is the holist strategy.
And still others hold that the concepts actually deployed in the
inferences guided by these principles themselves give rise to a priori
truths essentially involving these concepts, hence we are semantically
and default-reasonably entitled to the presupposition and use of these
principles.13 That is the inferentialist strategy.
The main problem with all three strategies is that there seems to be
no essential connection between rational entitlement and either pro at-
titudes, or coherence, or inferentialist semantics. For there could clearly
be pro attitudes, coherence, and inferentialist semantics in the absence of
the objectivity, necessity, and apriority of these inferential principles.
By sharp contrast to Non-Cognitivism, Holism, and Inferentialism
alike about the justification of inference, according to CKMI, an infer-
ence is inherently governed by categorically normative logical laws of
deduction, induction, or abduction. The justification of these specific
inferential principles then flows directly from rational obligations: Be-
cause you are a rational human animal, you categorically ought to reason
according to these principles. Hence you have an overriding practical
 96 Rober t Hanna

reason for carrying out that inference according to that inferential prin-
ciple. The ground or source of obligation, in turn, is rational human na-
ture and its absolute nondenumerable intrinsic value (a.k.a. “dignity”)
and, correspondingly, the specific constitution of our nature, namely,
our innate capacities for practical and theoretical reason. Furthermore,
if WCTI is also true, then the objectivity, necessity, and apriority of
the specific inferential principles is necessarily reflected in the mani-
fest world itself.
Given its robustly practical approach to the justification of infer-
ence, the CKMI account of the justification of deductive, inductive, and
abductive inference is quite similar to, and very much in the same spirit
as, what David Enoch and Joshua Schechter somewhat misleadingly
call “the pragmatic account” of justification—because it involves a di-
rect appeal to the overriding value of the ends and principles of human
rationality, over and above any merely instrumental motivation—
according to which

(i) there are certain projects that are rationally required for thinkers
like us and thereby rationally obligatory for thinkers like us, and
(ii) we are epistemically justified in employing a basic belief-forming
method that is indispensable for successfully engaging in one or
another of these rationally obligatory projects.14

The basic and important difference between the CKMI account and the
so-called pragmatic account, however, is that the CKMI account is ex-
plicitly grounded in a Kantian “metaphysics of morals,” and also fully
committed to WCTI, as well as to transcendental mentalism and—as I
argue in the next section— cognitive phenomenology. So the CKMI
account, in effect, includes the so-called pragmatic account and situates
it within a much broader and deeper epistemological and metaphysical
framework.

Th e M e ch ani s m of Inference:
Co gni t i v e Ph e no m e nol ogy without
Funct i o nal i s m or S e paratism

An inference in the minimal sense, as I spelled it out in the opening dis-

cussion, is a certain kind of cognitive process. But what is the mecha-
 What Is the Nature of Inference? 97

nism of the cognitive process of inference: that is, how does inference
actually happen? A “mechanism” in the sense I mean here is simply
an ordered spatiotemporal process with causal efficacy, perhaps goal-
directed or teleological, or perhaps not. Consequently, I do not mean
to imply that this ordered causally efficacious spatiotemporal pro-
cess has to be mechanical or mechanistic or in some narrower, non-
teleological sense of the notion of a mechanism—for example, Turing
computability— although that narrower sense is not ruled out by my
use of this notion either.
Is the mechanism of inference naturalizable, in the sense that it is
fully explicable by the natural sciences together with mathematics, in
the way that a digital computer (hardware) and its program (software)
are both fully explicable by the natural sciences together with mathe-
matics? Let us call the thesis that the mechanism of inference is natu-
ralizable in essentially the same way that computers are naturalizable
Functionalism about the mechanism of inference. The main problems
with Functionalism about the mechanism of inference are that either

(i) it is reductively naturalist and a version of Psychologism, and

thereby subject to the basic problems of Psychologism,
or else
(ii) it is nonreductively naturalist, but then inference has no causal
powers of its own to produce beliefs or intentional actions because
this production power is causally and explanatorily excluded by
the causal powers of its physical underpinnings,15 hence inference
is epiphenomenal, which seems clearly false.

For example, I believe that if P then Q, & P; so I believe that Q. It seems

to me clear that nothing other than my intentional act of inference itself
causally produced that belief,16 for otherwise I would not have been the
agential source of that inference, and consequently I could not have been
the agential source of that belief. On the contrary, something or some-
one else would have been the causal source of that inference and that be-
lief instead, so it would have been out of my hands and not up-to-me,
and thus not truly either my inference or my belief.17
And is the mechanism of inference intentional or representational
without necessarily being phenomenally conscious? Let us call the
 98 Rober t Hanna

thesis that the mechanism of inference is intentional or representa-

tional without necessarily being phenomenally conscious Separatism
about the mechanism of inference.18 More specifically, according to
Separatism about the mechanism of inference, the intentional or rep-
resentational content of inference is logically and metaphysically dis-
tinct from, and also independently variable in its specific character
with respect to, any phenomenal content that is associated with infer-
ence. Therefore, not only can the intentional or representational con-
tent of inference vary unrestrictedly its specific character with respect
to the phenomenal content of inference, but also the intentional or
representational content of inference can even occur altogether with-
out any phenomenal content, as in “nonconscious inference” or infer-
ential “zombiehood.”
The main problem with Separatism about inference mechanisms
is that the normativity of inference clearly requires both consciousness
and consciously free willing — otherwise it cannot be “in the space of
reasons” and is merely in the space of causes and brute valueless facts.
Randomly variable specific characters simply cannot be normative spe-
cific characters, because normativity is a certain special kind of regu-
larity; and zombies do not have norms, because, by hypothesis, zom-
bies cannot care about anything, and also because they are naturally
determined either by a distal or local causal source of choice and ac-
tion in such a way that they cannot be agential sources and so cannot
be free. So if Separatism with respect to inference mechanisms is true,
then the normativity of inference, and especially the categorical nor-
mativity of inference, but also even the instrumental normativity of
inference, cannot be explained.
According to CKMI, an inference is an intentional process that is
internally and/or externally consciously represented linguistically, with
more or less clear and distinct inferential phenomenology, and is car-
ried out by means of a prereflectively conscious or self-conscious free
choice of the will, under categorically normative metalogical laws and
specific inferential principles, for the sake of classical soundness and
minimal consistency. So the mechanism of inference is freely willed con-
scious linguistic cognition, inherently constrained by a special logico-
practical ought. Or in other words, as I put it right at the beginning of
this essay, inference is essentially an act of cognitive free agency. As I
 What Is the Nature of Inference? 99

also mentioned, this view does not imply voluntarism about inference
but also adequately explains our cognitive responsibility for the infer-
ences we make.

Co ncl usion

If I am correct, then Contemporary Kantian Moralism about Infer-

ence, or CKMI, outperforms all the relevant competing theories of the
nature of inference along all the basic dimensions of that problem. So
I conclude, by abduction, that CKMI is true.
That CKMI requires a weak form of transcendental idealism should
not be regarded as cause for philosophical alarm. For weak or counter-
factual transcendental idealism, or WCTI, is not scary idealism—that is,
it is not either subjective or Berkeleyan idealism, according to which
only individual minds and the contents of such minds exist, or absolute
or Hegelian idealism, according to which reality is just ideal rationality,
both of which seem highly implausible. All WCTI says, at bottom, is
that fundamental mental properties—for example, those instantiated in
your conscious understanding of this essay—are essentially co-basic in
nature with fundamental physical properties. Nature is not only essen-
tially physical; it is also essentially alive, purposive, and mental. This is
not to say that every part of nature is actually alive, purposive, con-
scious, caring, desiring, thinking, or rational but rather only to say some
parts of nature are actually alive, purposive, conscious, caring, desiring,
thinking, and rational (i.e., we are), and therefore that, necessarily, na-
ture as a whole always inherently included within itself, from the Big
Bang forward, the metaphysical ground of the potentiality of the ac-
tual existence of living, purposive, conscious, caring, desiring, thinking,
and/or rational animals. Call that thesis liberal naturalism.19 This is natu-
ral idealism, not scary idealism.

Not e s

1. For the purposes of this essay, I won’t attempt to define deduc-

tion, induction, or abduction. See, e.g., S. Shapiro, “Classical Logic,” in The
Stanford Encyclopedia of Philosophy (Winter 2009 Edition), ed. E. N. Zalta,
available at http://plato.stanford.edu/archives/win2009/entries/logic-classical/;
 100 Rober t Hanna

J. Hawthorne, “Inductive Logic,” in The Stanford Encyclopedia of Philosophy

(Winter 2011 Edition), http://plato.stanford.edu/archives/win2009/entries
/logic-inductive/; and I. Douven, “Abduction,” in The Stanford Encyclopedia
of Philosophy (Spring 2011 Edition), ed. E. N. Zalta, available at http://plato
.stanford.edu/archives/spr2011/entries/abduction/.
2. As always, the devil is in the “other things being equal” (a.k.a. ceteris
paribus) clause. I won’t attempt to spell it out here, but certainly special al-
lowances would have to be made for the distinction between monotonic (=
adding premises cannot reduce the set of logical consequences of the original
set of premises) and nonmonotonic (= adding premises can reduce the set of
logical consequences of the original set) reasoning, etc.
3. See R. Hanna, Rationality and Logic (Cambridge, MA: MIT Press,
2006), chap. 1.
4. For a modern version of Platonism, see B. Russell, The Problems of
Philosophy (Oxford: Oxford University Press, [1912] 1980), esp. chaps. 7–11.
5. See P. Benacerraf, “Mathematical Truth,” Journal of Philosophy 70
(1973): 661– 79; Hanna, Rationality and Logic, chap. 6; and R. Hanna, Cogni-
tion, Content, and the A Priori (Oxford: Oxford University Press, forthcom-
ing), chaps. 6– 8.
6. The explicit recognition of this problem goes back to Kant—see, e.g.,
Kant’s famous letter to Marcus Herz of 21 February 1771; and also the Critique
of Pure Reason Bxv–xviii and B166– 68. But it has also been recently rediscov-
ered by Hartry Field, in connection with the Benacerraf Dilemma, in “Recent
Debates about the A Priori,” in Oxford Studies in Epistemology, ed. T. Szábo
Gendler and J. Hawthorne (Oxford: Oxford University Press, 2005), 69– 88.
7. For a more explicit formulation and defense of WCTI, see Hanna,
Cognition, Content, and the A Priori, sec. 7.3.
8. See, e.g., B. Williams, “Internal and External Reasons,” in Moral Luck
(Cambridge: Cambridge University Press, 1981), 101–13.
9. See also Hanna, Rationality and Logic, chap.7; R. Hanna and M.
Maiese, Embodied Minds in Action (Oxford: Oxford University Press, 2009),
sec. 3.4; and Hanna, Cognition, Content, and the A Priori, chap. 5.
10. See Hanna, Rationality and Logic, chap. 3.
11. H. Field, “Apriority as an Evaluative Notion,” in New Essays on the
A Priori, ed. P. Boghossian and C. Peacocke (Oxford: Clarendon Press, 2000),
117– 49.
12. See, e.g., N. Goodman, “The New Riddle of Induction,” in Fact, Fore-
cast, and Fiction, 4th ed. (Cambridge, MA: Harvard University Press, 1983),
59– 83; and J. Rawls, A Theory of Justice (Cambridge, MA: Harvard University
Press, 1971).
 What Is the Nature of Inference? 101

13. See, e.g., P. Boghossian, “Knowledge of Logic,” in Boghossian and

Peacocke, New Essays on the A Priori, 229– 54; and C. Peacocke, “Explaining
the A Priori: The Programme of Moderate Rationalism,” in Boghossian and
Peacocke, New Essays on the A Priori, 255– 85.
14. See, e.g., D. Enoch and J. Shechter, “Meaning and Justification: The
Case of Modus Ponens,” Noûs 40 (2006): 687– 715; and D. Enoch and J. Schech-
ter, “How Are Basic Belief-Forming Methods Justified?,” Philosophy and Phe-
nomenological Research 76 (2008): 547– 79.
15. See J. Kim, Supervenience and Mind (Cambridge: Cambridge Univer-
sity Press, 1993), esp. chaps. 13, 14, and 17; J. Kim, Physicalism, or Something
Near Enough (Princeton: Princeton University Press, 2005), esp. chaps. 1 and 2;
and also J. Kim, Philosophy of Mind, 3rd ed. (Boulder, CO: Westview Press,
2011), chap. 7.
16. See also R. Wedgwood, “The Normative Force of Reasoning,” Noûs
40 (2006): 660– 86. Wedgwood’s nonreductive physicalist causal theory of in-
ferential basic intentional action is, as regards its intentional-action-oriented
and robustly normative approach to reasoning, similar to the CKMI account.
But Wedgwood’s theory also falls short of CKMI in two other important re-
spects: first, classical causal theories of action, no matter how sophisticated,
cannot ultimately solve the deviant causal chains problem; and second, nonre-
ductive physicalism, no matter how sophisticated, cannot ultimately avoid the
causal-explanatory exclusion problem and epiphenomenalism. See Hanna and
Maiese, Embodied Minds in Action, esp. chaps. 3, 4, 6, and 7.
17. The argument strategy I have just used is a version of what, in the
contemporary debate about free will, is known as “source incompatibilism.”
18. On Separatism vs. Anti-Separatism, see T. Horgan and J. Tienson, “The
Intentionality of Phenomenology and the Phenomenology of Intentionality,” in
Philosophy of Mind: Classical and Contemporary Readings, ed. D. Chalmers
(Oxford: Oxford University Press, 2002), 520– 33.
19. See, e.g., G. Rosenberg, A Place for Consciousness (Oxford: Oxford
University Press, 2005); Hanna and Maise, Embodied Minds in Action, esp. 11
and 312–13; R. Hanna, Kant, Science, and Human Nature (Oxford: Clarendon
Press, 2005), esp. 16–17, 310–11; T. Nagel, Mind and Cosmos (Oxford: Oxford
University Press, 2012); and R. Hanna, “Nagel & Me: Beyond the Scientific
Conception of the World” (unpublished MS), available at http://academia.edu
/3769311/Nagel_and_Me_Beyond_the_Scientific_Conception_of_the_World.
 C h a p t e r F o u r

Speculation and Narration in Mathematics

-
Laurent Laf f o rgue

The inspiration for the topic of this essay came from two pages by the
young French philosopher Fabrice Hadjadj, published in a recent book
of interviews, L’Héritage et la promesse (Legacy and Promise).1 Asked
about the key question on the articulation between theology and sci-
ences, Hadjadj replied:

If science seems to us so far removed from faith, it is not so much

because of the supposedly insurmountable obstacle between
experiment and faith, as the fact that ‘truth’, as most often repre-
sented by sciences, blanks out the proper noun, rejects actual exis-
tence, ignores the relationship. Truth tends to be thought of as see-
ing, and not living. Knowing truth is understood as being in the
position of a spectator dominating his object. In that sense, con-
temporary technology, television, the virtual world, would all ap-
pear to emanate from this concept of truth, where seeing takes over
from living. You could say that this is a concept of truth that is not
nuptial but pornographic: we want to see love made, possibly make
love, but not live it. For truth to be nuptial, it needs to resemble an
embrace: I must give up being an invulnerable spectator to enter
into a relationship with someone.

102
 Speculation and Narration in Mathematics 103

Christ’s extraordinary words to Thomas, “I am the way, the truth

and the life” (John 14:6), do indeed reveal the fact that “truth will not be
found in general terms, in what could be called anonymous clauses; no,
Truth is a person, with his uniqueness, with a proper Name of which
Peter says: ‘there is no other name under heaven given to mankind by
which we must be saved’” (Acts 4:12).
Which is why, says Hadjadj, “in theology, exploring truth always
requires both a speculative and a narrative approach, it has elements of
both science and biography. Yet, for us, biography and science have been
separated, speculation and narration have been perceived as opposites.”
The intuition we have of the existence of truth and of its unity
therefore prompts us to think of sciences from the point of view offered
by Hadjadj: are they purely speculative and in accordance with the way
we currently represent them, or are they also narrative and therefore
closer than is usually acknowledged to the model of the Revelation?
Are they purely theoretical—that is, etymologically, visual—or are they
also life experiences? Are they completely impersonal, or do they also
include at least in part relationships with people?
It seems perfectly natural to consider mathematics, as it is in the
first analysis seemingly the most speculative among the sciences, the
most theoretical and the most impersonal of all, as well as one that pro-
vides the models for modern physics and, to a lesser extent, to other sci-
ences. The mathematician that I am therefore believes it quite justifiable
for me to speak in front of philosophers and representatives from other
fields of learning and—because that is the crux of the matter—address
the link between this unique science and truth.
Let us first ask ourselves what thoughts would go through the
mind of a person who was a complete stranger to mathematics and who,
for the first time in his life, was invited to visit a university department
or research center dedicated to this discipline.
The first thought to take hold of his mind after this visit would
surely be that, in a mathematics center, there is “nothing to see”: in
other words, there is no spectacular experimentation equipment, no
out of the ordinary object that might spark some curiosity or interest,
just corridors, offices, rooms, and amphitheaters furnished only with
chairs, tables and seats, chalkboards or sometimes wipe boards, the sort
of computers and keyboards that you see everywhere nowadays, books
 104 Laurent L affo rgue

stacked in shelves, a few printers and photocopiers, some printed and

some blank paper, and pens.
Thinking about it a little more, the visitor might also reflect that,
as well as being very common and uninteresting, the objects seen are
few. In comparison with places of social activity— shops, transporta-
tion hubs, industry, construction, craft, hospitals, police stations, the
military— places of mathematics would seem to comprise an unusu-
ally far higher proportion of people to material objects. Only teaching
places—schools and high schools—and those other places of abstract
and speculative activity— banks and insurance companies— seem to
offer a similar appearance: everyday objects, few of them in fact, and
proportionally many people.
It is not just the case that places of mathematics look like teaching
places: they are, mostly. Our visitor could attend lectures, and if he
did, would first and foremost be struck by the fact that he understood
nothing, that all the sentences were woven with words the meaning of
which escapes him, or more intriguingly still, words that are familiar
in the common language but that become incomprehensible in the con-
text of the sentences spoken or written on the board. He could also at-
tend seminars, that is, sessions similar to lectures, but where, as he
could see, mathematicians would be teaching other mathematicians,
and not students, another cause for surprise and questions.
If, for want of anything interesting to look at, the visitor tried to
catch the people who inhabit places of mathematics in all their activities,
he could also overhear informal conversations among mathematicians—
and find then, not perhaps without feeling a degree of terror, that it
seems natural for these people to speak in a casual way with words no
less obscure than those used in lectures and seminar presentations. He
could also observe many people in offices busy reading—reading books
as well as printed texts or texts bound into what these people quaintly
call journal fascicules (in French, the same word is used for a journal as
for a newspaper)—or busy writing on a keyboard or with a pen. In the
last instance, he could catch a few people apparently lost in thoughts
more mysterious than everything else.
In short, the visitor would conclude that places of mathematics
are inhabited mainly by people who are busy talking and listening in
large and small groups, or reading, writing, and thinking alone.
 Speculation and Narration in Mathematics 105

So, are mathematicians related to writers, journalists, orators, or

activists, who commit to the cause presented in the talks they hear, or
to friends putting the world to rights, chatting over a café counter?
To attempt a reply to these questions, there is no other way than to
explore the contents of mathematicians’ lectures, presentations, con-
versations, journals, and books. What are they in fact talking about in
their prolific speech and writing, what are they expressing to one an-
other, and what are they expressing to themselves when they think?
For this reason, our visitor would need to venture into opening ar-
ticles and books written and read by mathematicians. Even if not one
word made sense, he would be struck by the extremely structured and
organized appearance the texts display: they are broken down into parts,
chapters, paragraphs, and subparagraphs, each with a title or at the very
least a sequencing and identification number. In addition, they com-
prise clearly identified and themselves numbered statements, with names
such as “definition,” “lemma,” “proposition,” “theorem,” together with
well-framed expositions introduced with words like “notation,” “dem-
onstration,” “remark,” and so on. The visitor would also notice that
these texts are riddled with internal and external references and that
these references obey a simple rule: they refer practically always to parts
of the text situated above, that is, read beforehand, and to articles or
books published previously.
Our mathematics visitor would therefore be led to observe that
each mathematical text is like a step on a journey, or a path, as if it told
a story in chronological order and, what is more, as if the specific story
it appears to tell were part of a general step in mathematical science and
in the history of that onward march.
However, the visitor would necessarily make a comment that seems
to contradict the previous one: whereas any text is structured like a story
and is explicitly linked to the wider story that mathematics as a whole
appears to form, it is written in the present tense. As if it were a story
without events, or rather a story telling events outside or beyond time.
Our visitor would then understand that mathematical texts belong
to the writing and reading time frame, as well as to the publications
time frame, but that the temporal structure of their neatly ordered sto-
rytelling mirrors the intemporal structure of what they are telling the
story about. The chain of logical implications that weaves the fabric of
 106 Laurent L affo rgue

deductive reasoning is displayed in the layers of linear reading over

time and in carefully dated publications.
Through the linear progression of single stories they each expound,
and in the vast story they all make up together, mathematical texts re-
veal logical implications as if there were causes and effects linked over
time, as if logic were a form of the principle of causality, as if logical
and structural structures identified one another. But the use of the pres-
ent tense in the texts’ verbs means that, in its very essence, mathemati-
cal logic does not depend on time.
And yet our visitor could wonder, any mathematical text is well
and truly anchored in time: it not only includes a logical structure in
the time frame of a story; it is itself included in the time frame of man-
kind as a moment and stage in the long history of the science it belongs
to. Most concepts, results, and methods that are part of it or that make
it possible are borrowed from other articles and books, or from other
mathematicians, long gone or contemporaries. Any mathematical text
is both a story, that is, a temporal form, of an intemporal mathematical
story and the hinge — between past and present—of the history of
mathematical science in its own time frame. This duality in the identity
of mathematical texts would in fact be particularly manifest in our visi-
tor’s eyes in the fact that very many mathematicians’ names are used to
designate mathematical objects and results, which is another way of
linking the intemporal substance of mathematics to their history made
by mankind.
Our observant visitor could also discover that mathematical texts
veer between, on the one hand, the use of prior methods, concepts, and
results in new analyses in order to solve problems set since a specific
time and, on the other, the development of new methods, the introduc-
tion of new concepts, the exploration of new territories, the formula-
tion of new statements the demonstration of which follows or are sug-
gested as conjectures. Each of these two models could then appear to
be both narrative and speculative, but in reverse: the model using estab-
lished theories in new analyses to prove well-known conjectures fits
more easily in the history of mathematics, in the illustration, summary,
and, in a way, the renewed telling of one of its pages. It is however
more speculative from the point of view of the substance of mathemat-
 Speculation and Narration in Mathematics 107

ics, as it consists in solving given problems by working back in the

chain of their causes, to find the implications that derive from them.
On the contrary, the model of discovery and development of new the-
ories is more speculative from the point of view of mathematical his-
tory, as it equates to going back to the source of ancient theories so
that, from those sources, another path is followed, but more narrative
from the point of view of mathematics, as it flows down the causal
chain, just as water flows down a riverbed.
Mathematical texts that expound or introduce a theory are indeed
similar to narratives, more specifically, to travel tales. The entirely dis-
cursive nature of speech or writing prevents an entire theory from
being presented in one scene: its paths need to be traveled slowly, its
crossroads negotiated, its towns, lands, and buildings need to be vis-
ited one by one.
Our still uninitiated visitor could finally ask himself if mathemati-
cal texts comprise events in the stories they unfold. He would conclude
that, from the history of mathematics angle, this is certainly the case:
the solution to an old problem or the emergence of a new concept rep-
resents historical events. But, from the substance of mathematics angle,
do events exist? If it is true that mathematics are outside of time, can
anything happen there? Our visitor would doubtless avoid answering
such a strange question too quickly. He would simply note that when
mathematicians talk through this or that theory or demonstration they
often use such expressions as “Here is where something happens.” As
if striking statements, an unexpected simplification, a computation re-
sult, gave these mathematicians the feeling they were moving from one
landscape to another, opening a new vista.
But if it is true that mathematicians’ texts and speeches, each indi-
vidually and all together, have the form of narratives, our visitor might
well ask for whom are mathematicians telling their stories?
However paradoxical it may seem, given the manifestly unenter-
taining nature of these texts and the difficulty of understanding them,
an objective fact would immediately spring to our visitor’s mind: lec-
tures are given so that students can follow them, seminar presentations
are made for colleagues to listen to them, conversations are held so that
mathematicians can enter into a dialogue with one another, articles are
 108 Laurent L affo rgue

written to be published, and, finally, journals and specialized books are

commercialized so that they can be bought—especially by university
libraries—and read by mathematicians.
Our visitor would then ask what inclines mathematicians to teach
students, give presentations for the attention of their colleagues or
speak to them informally, to write articles or books intended for publi-
cation. He would then learn that what those institutions which employ
mathematicians expect from them is that they give mathematics lec-
tures and presentations and publish articles or books in exchange for
the means of subsistence provided to them. This information would,
however, doubtless only displace the question in his mind, and make
him address those formative years, when longing to become a mathe-
matician appears and takes shape: why would young people ever en-
gage in careers of which the objective reality for them will consist of
speaking to be listened to and writing to be read? It is obvious that this
fact echoes far deeper human aspirations that dwell within them. So
what can these aspirations be?
Our visitor could first recognize in most researchers in mathemat-
ics—as in all academic fields— the avatar of a desire, highly visible in
children and probably still present in adults, although in a more con-
cealed manner: the desire to obtain confirmation from others, to draw
their attention, and to receive their approval of what they do, which is
felt in part to be approval of who they are. In other words, the desire to
be loved, more or less corrupted and confused by the tenacious illusion
that it is possible to conquer love by merit. Young children seek their
parents’ approval of what they do every day; to grow, they need to feel
their parents’ benevolent and supportive gaze on them. Later and in
parallel, the desire to receive positive judgments from their teachers
and professors represents one of students’ most powerful reasons for
working diligently. Our visitor, discovering the world of mathematics,
would not fail to liken this well-known characteristic of the human
heart to the importance afforded to small and great honors in univer-
sity life and the prestige attached to certain places, names, institutions,
and journals, and so on.
Acknowledging this could lead our visitor to interpreting the ob-
ligation, which mathematicians and other academic researchers have
 Speculation and Narration in Mathematics 109

placed upon themselves, to try to be listened to and read, as a sign of

immaturity. Unless he realized that another very deep human longing
dwelled in researchers side by side with the previous one, and in the
purest hearts, dominated it: the longing to share. In other words, the
need to offer others a form of love, by sharing with them objectively
precious things, which we possess or hope to be able to discover and
make known. People know deep in their hearts that giving brings
more happiness than receiving. For mathematicians, there is no other
possible gift than speech, and therefore, our visitor would think, no
true happiness is possible unless the narrative content of mathematics
is liable to be of considerable value. Mathematicians necessarily com-
pose their articles and books— that is, the tales of their mathematical
journeys— thinking that the regions these tales describe are beautiful
and worthy of being known and visited by others. They are propelled
by their desire to make known and loved their mathematics pathways
and landscapes, the wealth and splendor of which they discover; in the
same way are we happy to invite friends to a beloved country and try
to share with them all the knowledge that this country gives us.
But our visitor would soon observe that the efforts deployed by
mathematicians to share with others the truth and beauty they discover
encounter obstacles and disappointments similar to those we suffer from
when we invite dear friends to a familiar region, where everything speaks
to our hearts but not to theirs. Nowadays especially, mathematicians
seem to be moving away from one another at great speed, and whereas
the number of published articles and books never ceases to increase,
that of actual readers of almost all these articles and books is undoubt-
edly constantly decreasing.
Mathematicians expound their narratives in the hope of sharing,
that is, of experiencing a form of love with their human brethren
through mathematics; but most will know the secret misery of the fruit
of their mind’s labor awakening no particular interest, as if they were
talking and writing in a desert without ears or eyes. Even the most
highly regarded mathematicians, those whose presentations and writ-
ings are awaited, listened to, and read, generating other work, are not
sheltered from self-doubt and despair; this can arise if they become con-
scious that, in the tokens of respect and admiration they are surrounded
 110 Laurent L affo rgue

by, fascination for social glory and strength, or even jealousy, domi-
nates and supplants the shared love for truth.
Our visitor could also ask mathematicians the difficult question of
whether they prefer to read or write mathematics. Judging by the ever
increasing flow of articles sent to the editorial boards of journals and
by the matching increasing difficulty of finding reviewers for the ar-
ticles submitted, our visitor would probably hear most mathematicians
reply that they prefer to write mathematics. And this is humanly quite
normal, the visitor would reflect, as there is more joy in giving than in
receiving.
This clearly identified fact would then bring the visitor to ask him-
self whether writing a mathematical narrative intended to be read is in-
deed a form of gift to other mathematicians, if the latter do not in fact
appear to be very interested in reading them. And if writing mathe-
matical narratives represents a gift of questionable value to others, then
why write?
Perhaps to write to one’s self?
Our visitor would observe in any case that, as in any text, mathe-
matical texts are prepared and drafted by their author before being
read by others: at least chronologically, writing precedes reading, with
the result that a mathematician who writes has no other witness than
the narrative itself. This general nature of writing is particularly strongly
marked in mathematics, because, when compared to other sciences,
time flows in it slowly: the time taken to craft books and articles is mea-
sured in months, even years, and articles are generally published long
after they have been submitted to a specialized journal. Today most
articles are made available beforehand on the web; they are easier to
download, but time is still needed to read them, and most are probably
never read.
If he were to ask mathematicians, our visitor would in fact learn
that when they write up their work mathematicians are not thinking
about their future readers or anybody. They even forget themselves.
They think of nothing but the narrative, from which they are trying,
through their writing, to reveal in their minds the clearest possible
picture. A mathematical narrative is like a travel story in which, in fact,
the journey consists of the narrative itself, the attentive traveler is its
 Speculation and Narration in Mathematics 111

author, and the pen, pencil, or keyboard are both instruments of the
tale and instruments of the journey featured in the story. The paper or
screen, a proxy for paper, is an ocean, and the pen or keyboard, the
stereotyped form of a pen, is a boat on that ocean.
But if it is true that mathematicians first themselves tell their tales
for themselves, our visitor could ask, what is the point of this?
The question deserves all the more to be asked that mathematical
writing requires considerable efforts of concentration and attention,
at the very limits of the possibilities offered by the human mind. Our
visitor might then reflect that telling one’s self a story is only of true
value if it is apparent that the story is not one’s own, in other words,
that the story is inspired. The term inspiration is often used by widely
differing mathematicians.
But most go no further. They talk of inspiration, without the idea
entering their mind that it could come from an unknown person or
author, from God, whose marvels every mathematician would praise
and celebrate in writing and in speech.
Some mathematicians even say in the same breath that, to write
mathematics, hearing needs to be finely tuned to the subtle, and very
discreet, almost inaudible voice of mathematical truths that are waiting
to be told, to receive form in language; most, however, do not think that
anyone is speaking, and yet, in laying down on paper their mathematical
narratives, they show that they are paying attention to a certain kind of
words from this someone and that they find their only joy in becoming
servants to these words, following them as faithfully as they can.

Translation by Hélène Wilkinson,

Institut des Hautes Études Scientifiques

Not e

1. Mgr. Dominique Rey and Fabrice Hadjadj, L’Héritage et la promesse

(n.p.: Éditions RCF Méditerranée, 2011).
 P A R T I I I

-
Explanation in the Natural Sciences
 C h a p t e r F i v e

A Molecular Glimpse of How Mother Nature

Can Regulate Our Being

-
Thomas N owak

The concept of truth has been a philosophical topic of discussion and

study for centuries. Truth can be attributed to various forms of ques-
tions, focus, and presentations that include truth emanating from rea-
son, rhetoric, and religion. In these cases, “truth” arises from what
might be logic, argument, or beliefs dictated by a formal religious con-
viction. Truth in the arena of science comes from different sources;
those of research, observation, and discovery [1]. Even in this venue,
the concept of truth has various meanings. With this in mind, two dif-
ferent types of truth are defined, objective truth and subjective truth.
Objective truth is relatively easy to recognize; it appears absolute. If
someone walks into the room and the light is on, we all recognize that
is true. You are drinking coffee from a cup. It is clearly a fact. Subjec-
tive truth is much less distinct. It is how we perceive ideas, beliefs, and
concepts. In general, subjective truth emanates from our perception
based on arguments, prejudice, feelings, logic, and whatever means we
use to foster our belief that our ideas are correct. Truth as “common-
sense knowledge” is an example of subjective truth. It is rarely proven

115
 116 Thomas Nowak

but often believed. Our individual belief in a political philosophy is

subjective truth. In the Christian religion, one believes in the virgin
birth. We cannot prove this to be true, but it is inherent in the belief.
This truth or concept has been passed on through tradition over the
centuries. Many would believe that a particular actor or actress is good-
looking. Again, this is clearly subjective.
In science, objective truth normally takes the form of observation
and experimental data. Objective truth is usually obtained by infor-
mation collected: the observation of how fast a process occurs, the for-
mation of a precipitate, the measurement of how tightly a drug binds
to a receptor, the nuclear magnetic resonance (NMR) spectrum of a
synthesized compound, and so on. The validity of experimental data is
a critical part of science, and its truth or correctness is fundamental to
attempting to answer the scientific question addressed. The basis of sci-
entific ethics requires a clear description of experimental procedures
and a correct representation of the data collected. This information,
objective truth, is fundamental to the understanding of the mysteries
of nature. This approach is clearly the opposite approach to understand-
ing the universe suggested by Plato, who reputedly argued that “reason
alone was the superior approach to unravel the mysteries of nature.”
How far we have come in understanding Mother Nature!
In the realm of science, subjective truth is usually the interpreta-
tion of observed data. Such interpretations are referred to as hypothe-
ses or theories. The interpretation is based, in large part, on what sci-
entists already know relative to the question addressed and how the
data are consistent with conclusions proposed. Virtually all credible
scientists believe in the phenomenon of evolution. This hypothesis, or
theory, is consistent with biochemical, biological, and paleontological
studies. The data collected and observed make scientific sense. Nearly
all those who investigate this area of science believe this phenomenon
to be true, but clearly evolution has not been proven beyond a shadow
of a doubt. This might not even be possible. Interpretation of data is
subjective truth, and this is a normal process, especially in mechanistic
science. Mechanistic science attempts to determine how processes func-
tion and are controlled. The explanation of a process is based on the
hypothesis being consistent with the data obtained. This is a proces-
sive process to obtain absolute scientific truth. In this scientific pro-
 How Mother Nature Can Regulate Our Being 117

cess, truth is almost always an asymptotic search. The normal search

for scientific truth begins with an observation and the subsequent at-
tempts to explain it. A hypothesis or theory is generated to explain the
observation. There are often a number of reasonable theories that can
explain the observation, and scientists may disagree on which theory
is more correct. Subsequent experiments are performed to challenge
the hypothesis, and if the consequent data are consistent with this hy-
pothesis, it is retained. What is most often true is that hypotheses that
are inconsistent with the data are clearly incorrect. One description of
the scientific process is that it is the attempt to overthrow our cur-
rent solution to the problem addressed. The choice of experiments to
challenge the hypothesis depends, in part, on the technology available.
Many refinements of theories have been made because newer tech-
nology has enabled the experimental scientist to address the question
with greater speed, precision, detail, and accuracy. An important facet of
this scientific process is the imagination of the experiment performed.
Einstein commented that in science, “imagination is more important
than knowledge. For knowledge is limited, whereas imagination em-
braces the whole world, stimulating progress, giving birth to evolution”
[2]. In science, a hypothesis is rarely proven but is either supported or
demonstrated to be inconsistent with the data. Continuing studies rule
out other potential theories that are inconsistent with the data, and hy-
potheses that are consistent are retained or refined. A simple reflection
of this scientific process comes from an appropriate quote by Sher-
lock Holmes to his colleague Watson, “It is a capital mistake to theorize
before one has data. Insensibly one begins to twist facts to suit theories
instead of theories to suit facts” [3]. When hypotheses have been chal-
lenged over an extended period of time and always found to be con-
sistent, these then become “laws.” Such scientific laws include the
conservation of mass, the law of gravity, Boyle’s law, and the laws of
thermodynamics, among others.
The evolution of many theories has come from refinements of what
were the current theories. There are many striking examples, however,
where the challenge of current theory has led to total paradigm shifts
in understanding nature and often given rise to new applications of
the science, technology. Humans believed for centuries that physically
we were the center of the universe until observations performed and
 118 Thomas Nowak

analyzed first by N. Kopernik (1473–1543) and subsequently by Galileo

Galilei (1564 –1642) demonstrated that the sun was actually the center
of our cosmos. Of course, this was highly controversial and both suf-
fered from the authorities of the time as they expressed their interpre-
tations of their observations. The Big Bang theory drastically changed
our concepts of how the entire galaxy began and is still evolving. Signifi-
cant cosmology data support this theory. Very early belief, expressed
by the Greeks, was that there were four elements in nature: air, earth,
fire, and water. This changed when experiments initially performed by
Henry Cavendish (1731–1810) clearly described several elements that
were found to be fundamental structures in nature. This work has
continued so that we now can attribute the presence of 118 elements,
even though several have been synthetically prepared. The infamous
phlogiston theory of the late seventeenth century, described by G. Stahl
(1660–1734), explained combustion of different fuels containing an im-
measurable substance, phlogiston, which was emitted when combus-
tion occurred. The more phlogiston a substance contained, the better
the fuel. This theory held for a century before A. Lavoisier (1743– 94)
carefully quantified matter and demonstrated the presence of oxygen
that was utilized in the combustive process. Although a creative chemist/
alchemist, his career was cut short by the guillotine during the revolu-
tion in France. Atomic and molecular theory developed in the later part
of the eighteenth century, giving rise to chemical synthesis that has en-
abled us to devise and synthesize new compounds such as drugs, plas-
tics, fibers, fertilizers, explosives, and even computer chips. In the late
nineteenth century, the germ theory of disease was described from the
observations made by J. Lister and L. Pasteur who, in 1890, argued the
practice of antisepsis. This work emanated from the observations of
microorganisms that are present and correlated their presence with spe-
cific diseases. The ability to see these microorganisms was possible with
the development of microscopes by A. van Leeuwenhoek (1632–1723).
The attack on microorganisms to treat disease was initially performed
by the early experiments of P. Ehrlich, who showed that certain arsenic-
containing compounds could destroy the microorganisms that caused
syphilis (1880), hence, chemotherapy. Although most of these concepts
were slow to be accepted, the observations by A. Fleming in 1929 that
mold prevented the growth of pus-producing bacteria in the labora-
 How Mother Nature Can Regulate Our Being 119

tory gave rise to the recognition that the compound penicillin, found
in this mold, cured a large number of infections. It took about ten
years to implement this treatment as medical practice. These microor-
ganisms, and subsequently human cells, were believed to have magical
powers that existed only when cells were intact and living. This mys-
ticism was used to explain the power and chemistry of life in orga-
nisms and required that cells be intact. Experiments performed in 1899
by the Büchner brothers showed that following lysis of yeast cells, the
extract could also ferment grape juice to form wine, although at a
slower rate than intact yeast. These observations gave rise to the disci-
pline of biochemistry. This has allowed many new developments that
include our current “logical approach” in drug design. We can target
specific molecules, proteins, and nucleic acids that reside in cells as we
now believe there is an understanding of the function of many of these
molecules in living organisms. Molecular treatment of disease is cur-
rently a rapidly growing process. If light is a waveform, then it appears
logical that something must be waving similar to water molecules mov-
ing up and down in (water) waves. It was believed that something im-
mobile is present in space that allows light and gravity to be transmit-
ted. This something, believed to be a rarefied gas, did not interfere with
the motions of celestial bodies. It was an immobile constant of nature.
This rarefied gas was called ether, a term used by Aristotle to describe
the substance that makes up the heavens and heavenly bodies. It wasn’t
until July 1887 that the experiments of A. Michelson and E. Morley,
who measured the speed of light, demonstrated by interferometry, that
there was no “ether wind” that would affect the speed of light and con-
sequently no absolute motion in the universe. In the early twentieth cen-
tury, a number of observations by physicists and chemists could not be
explained by the classical theories of Newton and others in physics.
This required a revision of theory. M. Planck (1900) and N. Bohr (1914)
developed quantum theory to explain a number of molecular events.
This included the ability to describe light not only as a particle but also
as a wave. Many of our most recent electronic devices and computers
that are now common have evolved via applications of these recent the-
ories. For centuries we understood the concept of genetics in that in-
formation about what we call genetic traits was passed down through
progeny, whether plants (corn is a great example) or humans. These
 120 Thom as Nowak

unknown materials were called genes. Early in the twentieth century it

was thought that proteins were genes, but experiments in 1944 and
shortly thereafter by F. Griffith and O. Avery demonstrated that com-
pounds called nucleic acids carried genetic material. The characteriza-
tion of genetic material as DNA gave rise to molecular biology where
genetic analyses are now common and genetic engineering in plants
and microorganisms is a fruitful applied technology. We have dreams
of genetic treatment of diseases and possibilities of “curing” many dis-
eases genetically in the near future.
The specific example to demonstrate the evolution of scientific
truth that I develop in this essay is the attempt to clarify the biochemi-
cal mechanism for a key step in the regulation of the metabolic process
of glycolysis in greater detail. This study reflects the asymptotic pro-
cess of better understanding the details of how a specific regulatory
process occurs. Glycolysis is the metabolic pathway where the carbo-
hydrate metabolite glucose (and some other related carbohydrates) is
degraded under anaerobic conditions to generate pyruvic acid. In this
sequential metabolic process, chemical energy in the form of adenosine
triphosphate (ATP) is generated. This process occurs in ten sequential
steps, each step being catalyzed by a separate enzyme. This metabolic
pathway is virtually ubiquitous in all living systems, is critical to life,
and is one of the most well studied pathways in metabolism. Each of
the individual enzymes in the pathway has now been isolated, purified,
and characterized and the structures of each enzyme determined [4].
Most mechanistic details have been well investigated. Figure 5.1 out-
lines this pathway. The fate of the end product, pyruvate (pyruvic acid),
depends upon the metabolic state of the organism. Pyruvate can un-
dergo subsequent oxidative degradation that yields significantly more
chemical energy (ATP), can go on to form fats to conserve and store
chemical energy, can serve as a scaffold in the biosynthesis of some
amino acids, and can be used in the biosynthesis of glucose, or tem-
porarily form lactate (lactic acid) under anaerobic conditions. It should
be clear that these, and virtually all other metabolic pathways, must be
biochemically regulated to yield the most rapid, economical, and effi-
cient set of processes for the living organism and to respond to the meta-
bolic needs of the organisms in a rapid manner.
 Glucose
ATP
ADP
Glucose – 6 – P

Fructose – 6 – P
ATP
Phosphofructokinase (PFK)
ADP
Fructose – 1,6 – P2

GAP DHAP

1,3 – BPG (+) (–)

ADP
ATP
3 – PGA

2 – PGA

ADP, H+ PEP
M2+, M+ Pyruvate Kinase (PK)
ATP Pyruvate
Figure 5.1. The Glycolytic Pathway
The glycolytic pathway, the primary, anaerobic path of carbohydrate metabolism.
Abbreviations: ATP, adenosine-5’-triphosphate; ADP, adenosine-5’-diphosphate;
GAP, glyceraldehyde-3-phosphate; DHAP, dihydroxyacetone-phosphate;
1,3-BPG, 1,3 bisphosphoglycerate; 3-PGA, 3-phosphoglycerate; 2-PGA, 2-
phosphoglycerate; PEP, phosphoenolpyruvate. Glucose and fructose are 6 carbon
carbohydrates and each of the subsequent metabolites along the pathway are
3 carbon metabolites. The (+) arrow demonstrates where the product FBP serves
as a positive activator of PK and the (-) arrow demonstrates that the substrate PEP
serves as an inhibitor of PFK.
 122 Thom as Nowak

Mother Nature has devised a vast array of regulatory mechanisms

so that our biochemical ability to function allows us to rapidly respond
to physiological challenges accordingly. Eating, fasting, exercise, stress,
and so on, all require different biochemical responses. Control and regu-
lation of key metabolic steps is critical. A number of methods for regu-
lation and details of each control process often vary for each enzyme
that is being modulated. The two general methods of control are an
on/off process where the catalytic reaction is active or inactive (much
like a light switch) or modulation of the reaction process so that the re-
action rate catalyzed by the enzyme can proceed at a faster or slower
rate to produce the reaction product. The analogy here is a rheostat. In
the first case, the on/off process usually occurs by a covalent modifica-
tion of the enzyme that can be subsequently reversed when required.
This chemical modification is catalyzed by a specific enzyme, unrelated
to the target enzyme, which results in the alteration of the target. The
target enzyme might be phosphorylated, acetylated, glycosylated, and
so on, and this modification either turns on or turns off its active form.
Another separate enzyme can catalyze the reverse of this modification.
The latter case of modulation often occurs via the reversible binding of
some small molecule, usually another metabolite. As more of the regu-
lator molecule binds the target protein, the greater the effect on the tar-
get enzyme. Some of these metabolic regulatory processes can be initi-
ated by the secretion of metabolites or hormones that trigger a cascade
of regulatory steps for the process to begin or to finish. Understand-
ing how these mechanisms work is also the basis for a significant num-
ber of drug development studies.
The example chosen to demonstrate truth as a subjective scientific
understanding and how the approach to obtaining this truth is asymp-
totic is the understanding of the mechanism of regulation of the en-
zyme pyruvate kinase (PK). PK catalyzes the nearly irreversible reac-
tion of phosphoenolpyruvate (PEP) and adenosine diphosphate (ADP),
requiring the presence of two divalent cations (Mg2+ or Mn2+ can serve
this purpose) and a monovalent cation (K+), to yield pyruvate and ATP
[5]. The chemical process is shown in figure 5.2. This reaction results in
a net formation of 2 moles of ATP per mole of glucose utilized under
anaerobic conditions. Humans synthesize four different forms (iso-
zymes) of this enzyme: fetal muscle (fPK), red blood cell (rPK), liver
 How Mother Nature Can Regulate Our Being 123

Figure 5.2. The Chemical Mechanism Catalyzed by Pyruvate Kinase

The figure shows that this enzymatic reaction occurs at least in two distinct, known
steps on the enzyme. In the first step, the phosphoryl group of PEP is transferred
to ADP resulting in the formation of ATP and the very unstable, enzyme-bound
enolate form of pyruvate. In the second step, the enzyme catalyzes the addition
of a proton to the enolate of pyruvate to yield the stable keto form of pyruvate
[14]. The net reaction is the sum of the two partial reactions. Both steps require
the presence of the divalent and monovalent cations.

(lPK), and skeletal muscle (mPK) pyruvate kinases. Each of these en-
zymes catalyzes the identical reaction, but their kinetic behavior differs.
These differences in kinetic responses reflect specific metabolic require-
ments of the expressing tissues. These isozymes have very similar but
distinct amino acid sequences that account for these differences. The
first three forms of the enzyme display allosteric or sigmoidal kinetic
behavior in product formation as a function of substrate (PEP) con-
centration, whereas the skeletal muscle enzyme displays “classical”
hyperbolic kinetics (fig. 5.3). Hyperbolic kinetic responses are nor-
mally expected for a response of the initial velocity of the enzymatic
 124 Thom as Nowak

Figure 5.3. The Kinetic Responses of Allosteric Pyruvate Kinase in the Absence
and Presence of the Heterotropic Activator, FBP
The response of the initial velocity of the enzyme-catalyzed formation of pyruvate
as a function of the concentration of the substrate PEP in the absence (▲) and pres-
ence (●) of the heterotrotropic activator FBP. The hyperbolic curve (●) is the same
form as seen with the mPK.

reaction as a function of the concentration of the substrate, PEP (as

demonstrated, fig. 5.3). In some tissue cancers such as prostate or breast,
the PK that is normally present is mPK, whereas in tumors, the mPK
that would normally be present is replaced by fPK [6]. In part, the an-
aerobic metabolic process of generating energy as ATP differs. It was
observed many years ago that the rate of the glycolytic process increases
in tumor cells [7].
The key facet to the control mechanism of the allosteric form of
PK from most species is that the kinetic response of the enzyme to the
concentration of PEP is altered from the sigmoidal response to a hyper-
bolic response by the interaction of the enzyme with the product of an
earlier reaction in the pathway. Sigmoidal kinetics and/or binding is de-
scribed as positive cooperativity. The explanation for this phenomenon
 How Mother Nature Can Regulate Our Being 125

is that initial binding of the substrate to the enzyme is weak, but subse-
quent binding, once some substrate has bound, is enhanced. This im-
plies multiple binding sites on the enzyme. Since PK from most species
is a tetramer, there are four binding sites in the active enzyme complex.
In most species of PK, the metabolite that induces a hyperbolic kinetic
response is fructose-1,6-bisphosphate (FBP, or sometimes known as
FDP, fructose-1,6-diphosphate) that is formed by the reaction cata-
lyzed by the enzyme phosphofructokinase (PFK), the third reaction in
the pathway (see fig. 5.1). This activation process is known as “feed-
forward activation,” where a metabolite formed early in the pathway
activates an enzyme farther down the pathway so the rate of the path-
way increases. Of course, the change from sigmoidal to hyperbolic re-
sponse is not an all-or-none phenomenon but a gradual process as the
enzyme PK becomes saturated with the activator FBP. There are four
FBP binding sites on tetrameric PK. From these observations, it ap-
pears clear that the activation process of allosteric PK arises by the
binding of FBP as the transition of the kinetic effect takes place. Earlier
studies demonstrated that the binding of FBP to the enzyme causes
some alterations in the three-dimensional structure of PK [8]. Recent
crystallographic studies of PK from yeast, very homologous to lPK,
have demonstrated that FBP binds at a site that is about 40Å from the
active site of the enzyme where the catalytic reaction occurs [9]. This is
consistent with the general idea that allosteric modulation of activity
by a ligand that is not the substrate (a heterotropic modulator) is an
“action at a distance,” an indirect effect on kinetic responses. The cur-
rent hypothesis for the action of FBP on PK activation is that the bind-
ing of FBP to allosteric PK induces a subtle structural change at the
reactive site of the enzyme that results in enhanced binding of the sub-
strate PEP to the enzyme that is identical for each of the four bind-
ing sites on the enzyme. This results in a hyperbolic kinetic response.
This description is given in most biochemistry textbooks. This kinetic
response is then the same as found for mPK. This activation process is
considered to be due entirely to FBP binding. As the activation process
occurs, the KM, apparent value, the concentration of substrate that gives
half maximal activity, decreases. This transformation results in metabolic
activation because the physiological concentration of PEP in the cell is
roughly 25mM where enzymatic activity is very slow for the sigmoidal
 126 Thom as Nowak

form of the enzyme. Activity increases by as much as an order of mag-

nitude when hyperbolic kinetics occur (see fig. 5.3).
It has been observed that one divalent cation (either Mg2+ or Mn2+)
plays a key role in the binding of the substrate PEP to PK by serving
as a bridge for the substrate to interact with the enzyme. This metal
ion also assists in the activation process for catalysis [10, 11].
A more detailed investigation was initiated on the impact of both
the activator FBP on the interaction of PEP to PK and the role of the di-
valent cation in these interactions. Two approaches to these interactions
have been taken. One approach was kinetic where the effects of varying
concentrations of the divalent metal, FBP, and PEP on the initial rates
of the enzyme reaction are measured. A second approach is thermody-
namic in nature where the effects of the divalent metal, FBP, and PEP on
the mutual binding of each ligand are determined. Both sets of studies
take what were three-dimensional experiments (e.g., measurement of
kinetics vs. variable concentrations of PEP and metal) and extend them
to four dimensions (measurements of kinetics vs. variable concentrations
of PEP, metal, and FBP). The theoretical treatments of both the thermo-
dynamic and the kinetic studies and subsequent mathematical treat-
ments to fit the data are given in appendixes A and B.
In the first series of studies, a three-dimensional approach to the
putative synergism between ligands that bind the enzyme might occur
as shown in scheme I. In this example, the interaction of the divalent
metal (M) and the substrate PEP (S) can independently interact with
the enzyme (E) to form the resultant EM and ES binary complexes, re-
spectively. Each interaction is characterized by a kinetic or thermo-
dynamic constant, depending on the experiments performed. The terms
Km and Ks are used here to designate the strength of the formation of
complexes with metal and substrate, respectively. (The nomenclature
for thermodynamic and kinetic constants differs, but the fundamental
concept for synergism is the same.) In this formalism, the enzyme is ac-
tive only in the E-M-S complex since the metal is absolutely necessary
for catalysis to form the product of the reaction, P. In this scheme, one
can easily observe the identity Km ⋅ Km,s = Ks ⋅ Ks,m. Any synergism in
the interaction of substrate to the enzyme can be quantified via the
ratio Q where Qs,m = Ks/Km,s. In thermodynamic nomenclature, the
free energy change in binding, D(DGo′s,m) = -RT lnQs,m.
 How Mother Nature Can Regulate Our Being 127

SCHEME I

The results of a kinetic analysis of the effect of the divalent metal

Mn2+ on the kinetic response of the enzyme to variable concentrations
of PEP and the converse is shown in figure 5.4. The data demonstrate
that an increase in the concentration of the metal decreases the value
of Km,s and an increase in the concentration of PEP decreases the value
of Ks,m. Synergy occurs.
Fluorescence measurements were used to determine the effect of
the metal on direct binding of PEP to the enzyme and its converse. The
binding of either the metal ion to the enzyme or of PEP to the enzyme
results in a change in the intrinsic fluorescence of the enzyme. The data
reflect binding and are shown in figure 5.5. From both kinetic and ther-
modynamic studies, the data clearly demonstrate that the metal ion has
a substantial positive synergistic effect on the interaction of PEP to the
enzyme. The expected converse of the effect of PEP on the interaction
of the metal to the enzyme is also observed. This fits the model out-
lined in scheme I, and the extent of synergism is quantified. It is clear
from the kinetic studies that sigmoidal kinetic responses are observed
throughout these experiments since FBP is not present in these studies.
These studies have been extended to another dimension by the in-
vestigation of the interaction of FBP to the formation of the E-M-
FBP-PEP complex for catalysis [8, 12]. The studies have been extended
Figure 5.4. The Results of the Kinetic Study to Analyze the Effect of Increasing
Concentrations of Mn2+ on the Response of Enzyme Activity as a Function of
Variable Concentrations of PEP

A. The steady state kinetic rate profiles of PK as a function of variable concen-

trations of PEP at fixed, variable concentrations of Mn2+ [8]. The curves are best
fits to the data using the last equation given in Appendix B.

B. The calculated values

of the Km of PEP (●)
(from fig. A), FBP (■)
(data not shown), and
Q ′PEP-FBP as a function of
free Mn2+ concentration.
Note that the data are
plotted as a log-log plot.
 How Mother Nature Can Regulate Our Being 129

to analyze the respective effects of both the metal and the heterotropic
activator (an activator that is not the substrate) FBP on both ligand
binding and kinetic responses. In this case, scheme I is extended to give
scheme II. In scheme II, the activator, FBP is designated as F.
The analysis of this study is partially simplified kinetically since
kcat ≈ kcat′. The experiments were performed by a series of three-
dimensional studies with fixed variable concentrations of the free metal
ion resulting in the fourth dimension. The results of one such set of ex-
periments are shown in figure 5.6. In these studies, it is clear that both
FBP and M2+ have an influence on both the kinetics and the binding
(similar to the kinetic data but not shown) of PEP to the enzyme.
The analysis of both the kinetic and the thermodynamic results
were analyzed in a sequential manner to determine values for the in-
teraction of a single ligand to the enzyme in the absence of the second
ligand X (KAo ) and the value in the presence of various concentrations
of the second ligand, X, (KA). These studies were extended to where
both other ligands, X and Y, were present. The binding of each ligand
in the absence of any other is designated as KPEPo, KFBPo, and KMno.
The extent of couplings for the ligand pairs, synergistic effects, are
measured as QPEP-FBP, QPEP-Mn, and QFBP-Mn. The free energy change
in binding is calculated as DGPEP-FBP, and so on. These results are
shown in table 5.1.

Table 5.1 Kinetic and thermodynamic properties of ligand

interactions to pyruvate kinase [8]

Kinetic Values Thermodynamic Values

Parameter (mM) (mM)

KPEPo (mM) 2440 ± 540 1106 ± 187

KFBPo (mM) 150 ± 56 321 ± 7
KMno (mM) 860 ± 333 7180 ± 930
DGoPEP-FBP (kcal/mol) 0 -0.22 ± 0.03
DGoPEP-Mn (kcal/mol) -2.75 ± 0.14 -3.88 ± 0.08
DGoFBP-Mn (kcal/mol) -1.55 ± 1.03 -1.09 ± 0.02
 5.5A
Relative Flourescence

Wavelength (nm)

Figure 5.5. The Fluorescence Spectra of Yeast PK (YPK) and YPK in the Pres-
ence of Various Ligands
In figure 5.5A (above), the emission spectra of free YPK and the enzyme in the
presence of various ligands and combination of ligands are shown. In figure 5.5B
(right), the emission spectra for the enzyme with additional ligands are shown. In
these cases, the emission spectra of the enzyme in the presence of PEP, PEP + FBP
and PEP + FBP + Mn2+ all give virtually the same emission spectra [12].
 Relative Flourescence
5.5B

Wavelength (nm)
 132 Thomas Nowak

SCHEME II

The values for the synergy among the three ligands that interact
with the enzyme are similar whether measured by kinetic methods or
by direct binding, reflected in thermodynamic values. The values for the
interactions of Mn2+ to the free enzyme vary significantly, but weaker
binding and the inability to perform “simple” kinetic analyses at low
concentrations of Mn2+ give rise to these variations. It appears that the
activator FBP itself does not have a significant effect on the interaction
of PEP to the enzyme. The primary synergistic effect is the mutual ef-
fect of the substrate PEP and the metal on their respective interactions
to the enzyme. From a brief survey of the literature, this is the largest
synergistic effect measured for two ligands interacting to an enzyme.
Such a qualitative effect is not surprising since structural information
demonstrates that the metal interacts with the enzyme and PEP also in-
teracts with the enzyme-metal complex to form the E-M-S complex
[9]. From the level of the synergy, binding for both metal and substrate
is enhanced by approximately two orders of magnitude. The synergy
between the activator FBP and the metal is also relatively large. This is
not obviously expected since the binding of these two ligands occurs in
different segments of the protein, approximately 40Å apart [9]. Syner-
gism is not an obvious result. These data demonstrate that ligand bind-
ing to a protein may certainly have long-distance structural effects that
can alter the activity or behavior of the protein.
Figure 5.6
The apparent four-dimensional plot of the steady state kinetic rate profiles of pyru-
vate kinase as a function of variable concentrations of the substrate PEP at fixed
variable concentrations of FBP and various fixed, variable concentrations of Mn2+.
The series of plots on the left show the full set of data, and the plots on the right
are expanded to demonstrate, in greater detail, the changes in kinetics at the lower
concentrations of both PEP and of FBP prior to saturation effects occurring.
The measured steady state kinetic results are shown in the vertical axis as a
function of both variable PEP concentrations at fixed variable concentrations of
FBP. These experiments were all performed at fixed variable concentrations of free
Mn2+ as indicated at the far left. Concentrations of free Mn2+ below 29 mM were
too difficult to calculate with any degree of sensitivity and accuracy [12]. All data
showing sigmoidal behavior were fit to the Hill equation, whereas the data that
showed hyperbolic behavior were fit to the “classical” Michaelis-Menten equation,
both given in Appendix B.
 134 Thom as Nowak

The global analysis of the four-dimensional experiments of ligand

interactions by both kinetic and thermodynamic studies was also ana-
lyzed. In this treatment, all the data were utilized together in a general,
four-dimensional analysis. In this case KA was the value for PEP, Kx
for Mn2+, and Ky for FBP. One additional value that is obtained is the
combined coupling, QAXY for the ligands together. A comparison of
the kinetic values determined by either using a sequential analysis of
three-dimensional studies (1) or a global analysis in four dimensions
(mathematically) (2) is shown in table 5.2.

Table 5.2 Steady state kinetic parameters describing the influence

of Mn2+ and FBP on the interaction of PEP to pyruvate kinase [8]

Parameter 1 2

KPEPo (mM) 2466 ± 1352 2440 ± 541

KFBPo (mM) 129 ± 15 150 ± 56
KMno (mM) 880 ± 685 860 ± 33
QPEP-FBP 1.0 1.0
DGPEP-FBP (kcal/mol) 0 0
QMn-PEP 105 ± 56 105 ± 25
DGMn-PEP (kcal/mol) -2.75 ± 0.36 -2.75 ± 0.14
QMn-FBP 20 ± 4 14 ± 1.4
DGMn-FBP (kcal/mol) -1.77 ± 0.12 -1.55 ± 1.03
QPEP-Mn-FBP 1278 ± 154 1520 ± 880
DGPEP-Mn-FBP (kcal/mol) -4.22 ± 0.07 -4.32 ± 0.40

The results demonstrate that if one uses either the simpler (1) or
the more global treatment (2) of the data, virtually identical results are
obtained. The major synergistic effect on binding is between the diva-
lent metal and the substrate PEP. A smaller but significant synergistic
effect is observed between the metal ion and the allosteric activator FBP.
There is no effect of FBP on the interaction of the substrate PEP, unlike
that initially assumed from simpler experiments. A major enhancement
of the synergism among these three ligands for binding to the enzyme is
 How Mother Nature Can Regulate Our Being 135

observed when each of these three ligands is present. The free energy of
interaction for the presence of all ligands is increased by about another
-1.5 kcal/mol or another order of magnitude in binding to form the ac-
tivated form of the enzyme.
The results of this investigation demonstrate that when a more de-
tailed study of the effects of both the activating divalent metal ion and
the allosteric activator on the interactions of the substrate PEP with
PK is measured, it is the metal ion that serves as the key to enhancing
the interactions of PEP and FBP to the enzyme. The activator FBP is
not solely responsible for the activation process. The theory behind the
mechanism of how PK is activated by FBP has been refined.
This study is a great example of how truth in science progresses.
We believe in a particular hypothesis that serves as an understanding of
how and why particular processes occur. This theory is believed be-
cause it is consistent with the data that have been obtained relative to
the question raised. That belief is held as it continues to be consistent
with the information provided. As those beliefs are challenged by fur-
ther experimentation, they continue to be held unless the results of fur-
ther studies become inconsistent with the current theory. If that oc-
curs, the theory is subsequently expunged or refined such that the new,
revised truth is consistent with all the data. Truth in science often ap-
pears ephemeral as we continue to explore Mother Nature. Science is
an ongoing search for clear, objective truth, the goal of those who con-
tinue to pursue scientific truth.

Ap p e ndix A:
A B ri e f I nt ro duct i on to Thermodynamics
as Ap p l i e d to L i gand Binding to Proteins

The reversible binding of a ligand X to an enzyme E to form the EX

complex can be described as follows:
៝
E + X ៞ EX

The equilibrium for this binding is quantified via an equilibrium con-

stant Keq where
 136 Thom as Nowak

and [Ef] and [Xf] represent the free concentrations of E and X respec-
tively and [EX] is the concentration of the complex formed at equilib-
rium. The concentration of total enzyme and total ligand are determined
prior to the experiment then [Ef] = [Et] – [EX] and [Xf] = [Xt] – [EX].
In the experiment, either the concentration of the EX complex or ei-
ther [Ef] or [ Xf] are measured, giving sufficient information to mea-
sure Keq.
To understand the meaning and quantitative value of Keq, if a fixed
value of ET is used in the experiment and [EX] is measured at various
concentrations of XT, the Keq value is the concentration of free X that
gives half saturation of the enzyme ([Ef] = [EX]). Mathematically this
can be shown as follows:

This relationship demonstrates that when [Ef] = [EX], then the ratio
[Ef]/[EX] = 1 and Keq = [ Xf]. The equilibrium constant, also called
dissociation constant, Kd can be related to the free energy of binding
(DG°′ ) by the relationship

DG°′ = – RT ln Keq

R is the gas constant (R = 1.987 cal/mol •°K = 0.001987 kcal/ mol •°K).
°
K is the temperature in degrees Kelvin. If the temperature is 37° C
(98.7° F) then T = 310°K. Thus, DG°′ = - 0.616 ln Keq (kcal/mol).
To give a quantitative feel for the data, if an alteration in binding
occurs and is enhanced by an order of magnitude, the change in free
energy change is -1.4 kcal/mol.

D(DG°′ ) = DG°′ 1 – DG°′ 2

If D(DG°′ ) = 0 there is no alteration in binding.

If D(DG°′ ) > 0 weaker binding, negative synergy, occurs
If D(DG°′ ) < 0 stronger binding, positive synergy, occurs.
 How Mother Nature Can Regulate Our Being 137

Thus we can quantify the extent of ligand binding and any alteration
thereof via thermodynamic measurements and units.
For most of the measurements made in this study, the binding of a
ligand, PEP, FBP, or the metal ion has an influence on the fluorescence
(F) of the enzyme PK. The measure of a change in fluorescence as a
function of ligand added gives rise to a change in fluorescence. From
these data, the equilibrium constant can be quantified via the mathe-
matical model

Fobs = Fmax [L] / (Kd + [L]) eq. a

where [L] is the concentration of ligand added, Fobs is the observed

fluorescence at a given ligand concentration and Fmax is the maximum
fluorescence observed when the enzyme is saturated with L. If Fobs is
observed as a function of variable [L], this gives rise to a hyperbolic
response and Fmax is the asymptote. If the binding is cooperative, giv-
ing rise to a sigmoidal response, the data are then fit to a variant of the
above mathematical model

Fobs = Fmax [L]n /Kd + [L]n eq. b

where the coefficient n is a measure of the cooperativity in binding.

Equation b is the same form as the Hill equation [13] used to fit oxygen
binding to hemoglobin.

Ap p e ndi x B : Enz y me Kinetics

The measure of the initial rate of an enzyme-catalyzed reaction is a criti-

cal tool for examining how a catalytic process works. The treatment of
the data is based on a fundamental model where chemistry only occurs
when the substrate is bound to the enzyme ([ES]), the concentration
of substrate is significantly greater than the concentration of enzyme
([S] >> [E]) and the steady state occurs. In the steady state, the rate of
change of the concentration of the ES complex (d[ES]/dt) is 0. The rate
of the initial velocity of the reaction at steady state, vi is measured as a
function of the concentration of the substrate. The following model and
mathematical treatment is obtained.
 138 Thomas Nowak

The final equation shown, known as the Michaelis-Menten equation, is

the same form as for a rectangular hyperbola if vi is measured as a func-
tion of the variable concentration of substrate, [S]. The rate of the re-
action (d[P]/dt) is measured as a function of [S] and the concentration
of total E and S are both known to a high degree of accuracy. The con-
centration of ES is rarely known via normal measurements, and that
value is very small and difficult to directly measure. Although individ-
ual rate constants k1, k-1 and k2 are not determined by these methods, a
combination of these constants gives rise to a functional constant, re-
ferred to as a Michaelis Constant KM. This constant has units of concen-
tration (as does [S]) and has the functional meaning that it is the sub-
strate concentration that gives ½ maximal velocity of the reaction. Most
enzymes follow this general model. This is the identical mathematical
form used to analyze classic measured binding data. If “nonclassical”
kinetics are observed that reflect cooperativity, a variant of this equation
normally fits that data:

where, as with cooperative binding, the coefficient n is a quantitative

value of the extent of cooperativity. When n > 1, positive cooperativity,
 How Mother Nature Can Regulate Our Being 139

sigmoidal kinetics, are observed. The data are well fit via the Hill equa-
tion. Even in the case of cooperative kinetics, the KM value, regardless
of its microscopic meaning, is the concentration of substrate that gives
half maximal activity, Vmax.

R e f e re nce s

1. Gardener, W. M. (2012) Handling Truth. Fairhope, AL: Logica Books.

2. Einstein, Albert. (1931) Cosmic Religion: With Other Opinions and Apho-
risms. New York: Covici-Friede, 97.
3. Doyle, A. C. (1892) “A Scandal in Bohemia.” The Adventures of Sherlock
Holmes. London: George Newnes, 163.
4. Voet, D. & Voet, J. G. (2011) Biochemistry. 3rd ed. New York: John Wiley
& Sons.
5. Nowak, T. & Suelter, C. (1981) Pyruvate Kinase: Activation by and Cata-
lytic Role of the Monovalent and Divalent Cations. Mol. Cell. Biochem. 35,
65– 75.
6. Imamura, K., Noguchi, T. & Tanaka, T. (1986) Markers of Human Neuro-
ectodermal Tumors (Staal, G. E. & van Veelen, C.W. M.). New York: CRC
Press, 191– 222.
7. Warburg, O. (1956) On the Origin of Cancer Cells. Science 123, 309–14.
8. Mesecar, A. D. & Nowak, T. (1997) Metal Ion Mediated Allosteric Trigger-
ing of Yeast Pyruvate Kinase II. A Multidimensional Thermodynamic
Linked Function Analysis. Biochemistry 36, 6803–13.
9. Jurica, M. S., Mesecar, A. D., Heath, P. J., Shi, W., Nowak, T. & Stoddard,
B. L. (1998) The Allosteric Regulation of Pyruvate Kinase by Fructose-
1,6-bisphosphate. Structure 6, 195– 210.
10. Mildvan, A. S. & Cohn, M. (1966) Kinetic and Magnetic Resonance Stud-
ies of the Pyruvate Kinase Reaction: II. Complexes of Enzyme, Metal and
Substrates. J. Biol. Chem. 241, 1176– 93.
11. Nowak, T. & Lee, M. J. (1977) Reciprocal Cooperative Effects of Multiple
Ligand Binding to Pyruvate Kinase. Biochemistry 16, 1343– 50.
12. Mesecar, A. D. & Nowak, T. (1997) Metal Ion Mediated Allosteric Trig-
gering of Yeast Pyruvate Kinase I. A Multidimensional Kinetic Linked
Function Analysis. Biochemistry 36, 6792– 6802.
13. Hill, A. V. (1913) XLVII. The Combinations of Haemoglobin with Oxy-
gen and with Carbon Monoxide. I. Biochemistry 7, 471– 80.
14. Kuo, D. J. & Rose, I. (1978) Stereochemistry of Ketonization of Enol-
pyruvate by Pyruvate Kinase. Evidence for Its Role as an Intermediate.
J. Amer. Chem. Soc. 100, 6288– 89.
 C h a p t e r S i x

What Light Does Biology Shed on

the Social Sciences and the Humanities?

-
Francisco J. Ayal a

Humans are animals but a unique and very distinct kind of animal.
Our anatomical differences include bipedal gait and enormous brains.
But we are notably different also, and more important, in our individ-
ual and social behaviors and in the products of those behaviors. With
the advent of humankind, biological evolution transcended itself and
ushered in cultural evolution, a more rapid and effective mode of evo-
lution than the biological mode. Products of cultural evolution include
science and technology; complex social and political institutions; reli-
gious and ethical traditions; language, literature, and art; and electronic
communication.
Here, I explore ethics and ethical behavior as a model case to illu-
minate the interplay between biology and culture. I propose that our
exalted intelligence — a product of biological evolution— predisposes
us to form ethical judgments, that is, to evaluate actions as either good
or evil. I further argue that the moral codes that guide our ethical be-
havior transcend biology in that they are not biologically determined;
rather, by and large, they are products of human history, including so-
cial and religious traditions.

140
 What Light Does Biology Shed? 141

H um an Origins

Humankind is a biological species that has evolved from species that

were not human. Our closest biological relatives are the great apes and,
among them, the chimpanzees and bonobos, who are more closely re-
lated to us than they are to the gorillas, and much more than they are to
the orangutans. The hominid lineage diverged from the chimpanzee line-
age 6 million to 8 million years ago (Mya) and evolved exclusively in the
African continent until the emergence of Homo erectus, somewhere
around 1.8 Mya. Shortly after their emergence in tropical or subtropi-
cal eastern Africa, H. erectus dispersed to other continents of the Old
World, where their descendants eventually became extinct. Modern hu-
mans first arose in Africa (or in the Middle East) between 150,000 years
ago (kya) and 100 kya. Starting around 60 kya modern humans spread
throughout the world, eventually replacing in the Old World any sur-
viving Homo populations, including Homo neanderthalensis.

H um an Di s t inct iv e Traits

Erect posture and large brain are the two most conspicuous human
anatomical traits. We are the only vertebrate species with a bipedal gait
and erect posture; birds are bipedal, but their backbone stands hori-
zontal rather than vertical (penguins are a minor exception). Brain size
is generally proportional to body size; relative to body mass, humans
have the largest (and most complex) brain. The chimpanzee’s brain
weighs less than a pound; a gorilla’s, slightly more. The human male
adult brain has a volume of 1,400 cubic centimeters (cc), about three
pounds in weight.
Until recently, evolutionists raised the question whether bipedal
gait or large brain came first, or whether they evolved consonantly. The
issue is now resolved. Our Australopithecus ancestors had, since four
million years ago, a bipedal gait, but a small brain, about 450 cc, a pound
in weight. Brain size starts to increase notably with our Homo habilis
ancestors, about 2.5 Mya, who had a brain about 650 cc and also were
prolific toolmakers (hence the name habilis). Between one and two
 142 Francisco J. Ayala

million years afterward, Homo erectus had adult brains about 1,200 cc.
Our species, Homo sapiens, has a brain about three times as large as
that of Australopithecus, 1,300 to 1,400 cc, or some three pounds of gray
matter. Our brain is not only much larger than that of chimpanzees or
gorillas but also much more complex. The cerebral cortex, where the
higher cognitive functions are processed, is in humans disproportion-
ally much greater than the rest of the brain when compared to apes.
Erect posture and large brain are not the only anatomical traits
that distinguish us from nonhuman primates, even if they may be the
most obvious. A list of our most distinctive anatomical features in-
cludes the following:

• Erect posture and bipedal gait (which entail changes of the back-
bone, hipbone, and feet)
• Opposing thumbs and arm and hand changes (which make pos-
sible precise manipulation)
• Large brain
• Reduction of jaws and remodeling of face
• Cryptic ovulation (and extended female sexual receptivity)
• Slow development
• Modification of vocal tract and larynx

Humans are notably different from other animals not only in

anatomy but also, and no less important, in their behavior, both as in-
dividuals and socially. A list of distinctive human behavioral traits in-
cludes the following:

• Intelligence: abstract thinking, categorizing, and reasoning

• Symbolic (creative) language
• Self-awareness and death awareness
• Toolmaking and technology
• Science, literature, and art
• Ethics and religion
• Social organization and cooperation (division of labor)
• Legal codes and political institutions
 What Light Does Biology Shed? 143

Humans live in groups that are socially organized, and so do

other primates. But primate societies do not approach the complexity of
human social organization. A distinctive human social trait is culture,
which may be understood as the set of human activities and creations
that are not strictly biological. Culture includes social and political insti-
tutions, ways of doing things, religious and ethical traditions, language,
common sense and scientific knowledge, art and literature, technology,
and in general all the creations of the human mind. The advent of culture
has brought with it cultural evolution, a superorganic mode of evolution
superimposed on the organic mode, which has, in the last few millennia,
become the dominant mode of human evolution. Cultural evolution has
come about because of cultural change and inheritance, a distinctively
human mode of achieving adaptation to the environment and transmit-
ting it through the generations.

B i o l o gi cal Ro ot s of Et hical Behavior

I want to now discuss ethics and ethical behavior as a model case of how
we may seek the evolutionary explanation of a distinctively human trait.
The objective is to ascertain whether an account can be advanced of ethi-
cal behavior as an outcome of biological evolution and, if such is the
case, whether ethical behavior was directly promoted by natural selec-
tion or has rather come about as an epigenetic manifestation of some
other trait that was the target of natural selection.
People have moral values; that is, they accept standards according
to which their conduct is judged either right or wrong, good or evil. The
particular norms by which moral actions are judged vary to some ex-
tent from individual to individual and from culture to culture (although
some norms, like not to kill, not to steal, and to honor one’s parents,
are widespread and perhaps universal), but value judgments concerning
human behavior are passed in all cultures. This universality raises two
related questions: whether the moral sense is part of human nature, one
more dimension of our biological makeup; and whether ethical values
may be products of biological evolution rather than being given by
religious and other cultural traditions.
 144 Francisco J. Ayala

When philosophers consider theories of morality they distinguish

between metaethics, normative ethics, and practical ethics (Copp 2006).
Theories of metaethics seek to justify why we ought to do what we
ought to do. They are the primary concern of philosophers, who favor
different theories, such as “divine command” (God’s commanding is
what makes a particular kind of action moral); “moral realism” (there
are moral facts; our moral judgments are made valid or not by the moral
facts); “utilitarianism” (the moral value of an action is determined by the
expected benefit to the largest number of people); “positivism” (there
are no objective rational foundations for morality, but rather moral
norms are determined by social agreement or, in the individual, by emo-
tional decisions); “libertarianism” (moral values are measured by the ex-
tent to which they maximize personal freedom and limit the role of the
state to the protection of individual freedoms); and several others.
Normative ethics refers to the rules or laws that determine what we
ought to do. Practical ethics considers the application of moral norms to
particular situations, which often involve conflicting values: will abor-
tion be justified in order to save the life of the mother?
Aristotle and other philosophers of classical Greece and Rome, as
well as many other philosophers throughout the centuries, held that
humans hold moral values by nature. A human is not only Homo sapi-
ens but also Homo moralis. For the past twenty centuries, the founda-
tions of morality were an important subject for Christian theologians,
as in the case of Thomas Aquinas, but also for philosophers, such as,
in the eighteenth and nineteenth centuries, Hume, Kant, and others
familiar to Darwin, including notably William Paley (The Principles of
Moral and Political Philosophy, 1785) and Harriet Martineau (Illustra-
tions of Political Economy, 1832– 34).
The theory of evolution brought about the need to reconsider the
foundations of morality. We do not attribute ethical behavior to ani-
mals (surely, not to all animals and not to the same extent as to hu-
mans, in any case). Therefore, evolution raises distinctive questions
about the origins and tenets of moral behavior. Is the moral sense de-
termined by biological evolution? If so, when did ethical behavior
come about in human evolution? Did modern humans have an ethical
sense from the beginning? Did Neandertals hold moral values? What
 What Light Does Biology Shed? 145

about H. erectus and H. habilis? And how did the moral sense evolve?
Was it directly promoted by natural selection? Or did it come about as
a by-product of some other attribute (such as rationality, for example)
that was the direct target of selection? Alternatively, is the moral sense
an outcome of cultural evolution rather than of biological evolution?

Darw i n and t he Moral Sense

Darwin’s most sustained discussion of morality is in chapter 3 of The

Descent of Man (Darwin 1871, 67–102). The keystone significance of
morality in human distinctness is clearly asserted by Darwin in that
chapter’s first sentence: “I fully subscribe to the judgment of those writ-
ers who maintain that of all the differences between man and the lower
animals the moral sense or conscience is by far the most important” (67).
Darwin’s two most significant points concerning the evolution of
morality are stated early in chapter 3 of The Descent of Man. The two
points are (1) that moral behavior is a necessary attribute of advanced in-
telligence as it occurs in humans, and thus that moral behavior is biologi-
cally determined; and (2) that the norms of morality are not biologically
determined but a result of human collective experience, or human cul-
ture as we would now call it.
After the two initial paragraphs of chapter 3, which assert that
the moral sense is the most important difference “between man and
the lower animals” (see quotation above), Darwin states his view that
moral behavior is strictly associated with advanced intelligence: “The
following proposition seems to me in a high degree probable—namely,
that any animal whatever, endowed with well-marked social instincts,
would inevitably acquire a moral sense or conscience, as soon as its in-
tellectual powers had become as well developed, or nearly as well de-
veloped, as in man” (68– 69). Darwin is affirming that the moral sense,
or conscience, is a necessary consequence of high intellectual powers,
such as exist in modern humans. Therefore, if our intelligence is an out-
come of natural selection, the moral sense would be as well an outcome
of natural selection. Darwin’s statement further implies that the moral
sense is not by itself directly promoted by natural selection but only
 146 Francisco J. Ayala

indirectly as a necessary consequence of high intellectual powers, which

are the attributes that natural selection is directly promoting.
In the ensuing paragraph of chapter 3, before proceeding to a dis-
cussion of how morality might evolve, Darwin makes an important dis-
tinction: “It may be well first to premise that I do not wish to maintain
that any strictly social animal, if its intellectual faculties were to become
as active and as highly developed as in man, would acquire exactly the
same moral sense as ours. . . . [T]hey might have a sense of right and
wrong, though led by it to follow widely different lines of conduct”
(70). According to Darwin, having a moral sense does not by itself de-
termine what the moral norms would be: which sorts of actions might
be sanctioned and which ones would be condemned.
Darwin’s distinction between the moral sense or conscience, on
the one hand, and the moral norms that guide the moral sense or con-
science, on the other, is fundamental. It is a distinction I will elaborate.
Much of the post-Darwin historical controversy, particularly between
scientists and philosophers, as to whether the moral sense is or is not
biologically determined, has arisen owing to a failure to make that dis-
tinction. Scientists often affirm that morality is a human biological at-
tribute because they are thinking of the predisposition to make moral
judgments: that is, to judge some actions as good and others as evil.
Some philosophers argue that morality is not biologically determined
but rather comes from cultural traditions or from religious beliefs, be-
cause they are thinking about moral codes, the sets of norms that de-
termine which actions are judged to be good and which are evil. They
point out that moral codes vary from culture to culture and, therefore,
are not biologically predetermined.

Darw i ni an Aft e rmath

Herbert Spencer (1820 –1903) was among the first philosophers seek-
ing to find the grounds of morality in biological evolution. In The Prin-
ciples of Ethics (1893), Spencer seeks to discover values that have a natu-
ral foundation. He argues that the theory of organic evolution implies
certain ethical principles. Human conduct must be evaluated, like any
biological activity whatsoever, according to whether it conforms to the
 What Light Does Biology Shed? 147

life process; therefore, any acceptable moral code must be based on natu-
ral selection, the law of struggle for existence. According to Spencer, the
most exalted form of conduct is that which leads to a greater duration,
extension, and perfection of life; the morality of all human actions must
be measured by that standard. Spencer proposes that, although excep-
tions exist, the general rule is that pleasure goes with that which is bi-
ologically useful, whereas pain marks what is biologically harmful.
This is an outcome of natural selection: thus, while doing what brings
them pleasure and avoiding what is painful, organisms improve their
chances for survival. With respect to human behavior, we see that we
derive pleasure from virtuous behavior and pain from evil actions, as-
sociations which indicate that the morality of human actions is also
founded on biological nature.
Spencer proposes as the general rule of human behavior that people
should be free to do anything that they want, so long as it does not in-
terfere with the similar freedom to which others are entitled. The
justification of this rule is found in organic evolution: the success of an
individual, plant or animal, depends on its ability to obtain that which
it needs. Consequently, Spencer reduces the role of the state to protect-
ing the collective freedom of individuals so that they can do as they
please. This laissez-faire form of government may seem ruthless, be-
cause individuals would seek their own welfare without any considera-
tion for others’ (except for respecting their freedom), but Spencer be-
lieves that it is consistent with traditional Christian values. It may be
added that, although Spencer sets the grounds of morality on biological
nature and on nothing else, he admits that certain moral norms go be-
yond that which is biologically determined; these are rules formulated
by society and accepted by tradition.
Social Darwinism, in Spencer’s version or in some variant form,
was fashionable in European and American circles during the latter part
of the nineteenth century and the early years of the twentieth century,
but it has few or no distinguished intellectual followers at present. Spen-
cer’s critics include the evolutionists J. S. Huxley and C. H. Wadding-
ton who, nevertheless, maintain that organic evolution provides grounds
for a rational justification of ethical codes. For Huxley (1953; Huxley
and Huxley 1947), the standard of morality is the contribution that ac-
tions make to evolutionary progress, which goes from less to more
 148 Francisco J. Ayala

‘‘advanced’’ organisms. For Waddington (1960), the morality of actions

must be evaluated by their contribution to human evolution.
Huxley’s and Waddington’s views are based on value judgments
about what is or is not progressive in evolution. But, contrary to Hux-
ley’s claim, there is nothing objective in the evolutionary process itself
(i.e., outside human considerations) that makes the success of bacteria,
which have persisted as such for more than two billion years and which
consist of a huge diversity of species and astronomical numbers of indi-
viduals, less valuable than that of the vertebrates, even though the latter
are more complex. The same objection can be raised against Wadding-
ton’s human evolution standard of biological progress. Are the insects,
of which more than one million species exist, less successful or less
valuable from a purely biological perspective than humans or any other
mammal species? Waddington fails to demonstrate why the promotion
of human biological evolution by itself should be the standard to mea-
sure what is morally good.
More recently, numerous philosophers as well as scientists have
sought to give accounts of moral behavior as an evolutionary out-
come (e.g., Blackmore 1999; Hauser 2006; Maienschein and Ruse 1999;
Ruse 1995; Sober and Wilson 1998; Wilson 2012). Particularly notable
are the early contributions of Edward O. Wilson (1975, 1978, 1998),
founder of sociobiology as an independent discipline engaged in discov-
ering the biological foundations of all social behavior. Sociobiologists,
as well as the derivative subdisciplines of evolutionary psychology (e.g.,
Barkow, Cosmides, and Tooby 1992) and memetics (Blackmore 1999),
have sought to solve the naturalistic fallacy by turning it on its head.
They assert that moral behavior does not exist as something distinct
from biological, or biologically determined, behavior. As Ruse and Wil-
son (1985) have asserted, “Ethics is an illusion put in place by natural se-
lection to make us good cooperators” (emphasis added).

M o ral Judgm e nt v e rsus Moral Norms

The question whether ethical behavior is biologically determined may,

indeed, refer to either one of the following two issues. First, is the
 What Light Does Biology Shed? 149

capacity for ethics— the proclivity to judge human actions as either

right or wrong—determined by the biological nature of human beings?
Second, are the systems or codes of ethical norms accepted by human
beings biologically determined? A similar distinction can be made with
respect to language. The question whether the capacity for symbolic
creative language is determined by our biological nature is different
from the question whether the particular language we speak—English,
Spanish, Chinese, and so on—is biologically determined, which in the
case of language obviously it is not.
I propose that the moral evaluation of actions emerges from human
rationality or, in Darwin’s terms, from our highly developed intellec-
tual powers. Our high intelligence allows us to anticipate the conse-
quences of our actions with respect to other people and, thus, to judge
them as good or evil in terms of their consequences for others. But I
will argue that the norms according to which we decide which actions
are good and which actions are evil are largely culturally determined,
although conditioned by biological predispositions, such as parental
care to give an obvious example.
The moral sense refers first and foremost to our predisposition to
evaluate some actions as virtuous, or morally good, and others as evil, or
morally bad. Morality, thus, consists of the urge or predisposition to
judge human actions as either right or wrong in terms of their conse-
quences for other human beings. In this sense, humans are moral beings
by nature because their biological constitution determines the presence
in them of the three necessary conditions for ethical behavior. These
conditions are (i) the ability to anticipate the consequences of one’s own
actions; (ii) the ability to make value judgments; and (iii) the ability to
choose between alternative courses of action. These abilities exist as a
consequence of the eminent intellectual capacity of human beings.
The ability to anticipate the consequences of one’s own actions is
the most fundamental of the three conditions required for ethical behav-
ior. Only if I can anticipate that pulling the trigger will shoot the bullet,
which in turn will strike and kill my enemy, can the action of pulling the
trigger be evaluated as nefarious. Pulling a trigger is not in itself a moral
action; it becomes so by virtue of its relevant consequences. My action
has an ethical dimension only if I do anticipate these consequences.
 150 Francisco J. Ayala

The ability to anticipate the consequences of one’s actions is closely

related to the ability to establish the connection between means and
ends, that is, of seeing a means precisely as a means, as something that
serves a particular end or purpose. This ability to establish the connec-
tion between means and their ends requires the ability to anticipate
the future and to form mental images of realities not present or not yet
in existence.
The ability to establish the connection between means and ends
happens to be the fundamental intellectual capacity that has made possi-
ble the development of human culture and technology. An evolutionary
scenario, seemingly the best hypothesis available, proposes that the re-
mote evolutionary roots of this capacity to connect means with ends
may be found in the evolution of bipedalism, which transformed the an-
terior limbs of our ancestors from organs of locomotion into organs of
manipulation. The hands thereby gradually became organs adept for the
construction and use of objects for hunting and other activities that im-
proved survival and reproduction, that is, which increased the reproduc-
tive fitness of their carriers. The construction of tools depends not only
on manual dexterity, but on perceiving them precisely as tools, as objects
that help to perform certain actions; that is, as means that serve certain
ends or purposes: a knife for cutting, an arrow for hunting, an animal
skin for protecting the body from the cold. According to this evolution-
ary scenario, natural selection promoted the intellectual capacity of our
bipedal ancestors because increased intelligence facilitated the percep-
tion of tools as tools, and therefore their construction and use, with the
ensuing improvement of biological survival and reproduction.
The development of the intellectual abilities of our ancestors took
place over several million years, gradually increasing the ability to con-
nect means with their ends and, hence, the possibility of making ever
more complex tools serving more diverse and remote purposes. Ac-
cording to the hypothesis, the ability to anticipate the future, essential
for ethical behavior, is therefore closely associated with the develop-
ment of the ability to construct tools, an ability that has produced the
advanced technologies of modern societies and that is largely respon-
sible for the success of humans as a biological species.
The second condition for the existence of ethical behavior is the
ability to advance value judgments, to perceive certain objects or deeds
 What Light Does Biology Shed? 151

as more desirable than others. Only if I can see the death of my enemy
as preferable to his survival (or vice versa) can the action leading to his
demise be thought of as moral. If the consequences of alternative ac-
tions are neutral with respect to value, an action cannot be character-
ized as ethical. Values are of many sorts: not only ethical but also aes-
thetic, economic, gastronomic, political, and so on. But in all cases, the
ability to make value judgments depends on the capacity for abstrac-
tion, that is, on the capacity to perceive actions or objects as members
of general classes. This makes it possible to compare objects or actions
with one another and to perceive some as more desirable than others.
The capacity for abstraction requires an advanced intelligence such as
it exists in humans and apparently in them alone.
I will note at this point that the model that I am advancing here
does not necessarily imply the ethical theory known as utilitarianism
(or, more generally, consequentialism). According to the “act conse-
quentialism,” the rightness of an action is determined by the value of its
consequences, so that the morally best action in a particular situation
is the one the consequences of which would have the most benefit to
others. I am proposing that the morality of an action depends on our
ability (1) to anticipate the consequences of our actions and (2) to make
value judgments. But I am not asserting that the morality of actions is
exclusively measured in terms of how beneficial their consequences
will be to others.
The third condition necessary for ethical behavior is the ability to
choose between alternative courses of actions. Pulling the trigger can
be a moral action only if you have the option not to pull it. A necessary
action beyond conscious control is not a moral action: the circula-
tion of the blood or the process of food digestion are not moral ac-
tions. Whether there is free will is a question much discussed by phi-
losophers, and the arguments are long and involved (e.g., Fischer 2006;
Bok 1998; Ekstrom 2000; Kane 1996; see also Greene and Haidt 2002;
Haidt, Bjorklund, and Murphy 2000; Hauser 2006). Here, I will advance
two considerations that are commonsense evidence of the existence of
free will. One is personal experience, which indicates that the possibility
to choose between alternatives is genuine rather than only apparent. The
second consideration is that when we confront a given situation that
requires action on our part, we are able mentally to explore alternative
 152 Francisco J. Ayala

courses of action, thereby extending the field within which we can exer-
cise our free will. In any case, if there were no free will, there would be
no ethical behavior; morality would only be an illusion. A point to be
made, however, is that free will is dependent on the existence of a well-
developed intelligence, which makes it possible to explore alternative
courses of action and to choose one or another in view of the antici-
pated consequences.

M i nd to Moral it y

The capacity for ethics is an outcome of gradual evolution, but it is an at-

tribute that exists only when the underlying attributes (i.e., the intellec-
tual capacities) reach an advanced degree. The necessary conditions for
ethical behavior come about only after the crossing of an evolutionary
threshold. The approach is gradual, but the conditions appear only when
a degree of intelligence is reached such that the formation of abstract
concepts and the anticipation of the future are possible, even though we
may not be able to determine when the threshold was crossed. Thresh-
olds occur in other evolutionary developments—for example, in the ori-
gins of life, multicellularity, and sexual reproduction—as well as in the
evolution of abstract thinking and self-awareness. Thresholds occur in
the physical world as well; for example, water heats gradually, but at
100°C boiling begins and the transition from liquid to gas starts sud-
denly. Surely, human intellectual capacities came about by gradual evo-
lution. Yet, when looking at the world of life as it exists today, it would
seem that there is a radical breach between human intelligence and that
of other animals. The rudimentary cultures that exist in chimpanzees
do not imply advanced intelligence as it is required for moral behavior.
A different explanation of the evolution of the moral sense has been
advanced by proponents of the theory of gene-culture coevolution
(Richerson and Boyd 2005; Strimling, Enquist, and Eriksson 2009; Rich-
erson, Boyd, and Henrich 2010; see also Greene et al. 2001; Haidt 2007).
It is assumed that cultural variation among tribes in patriotism, fidelity,
sympathy, and other moralizing behaviors may have occurred incipi-
ently in early hominid populations, starting at least with H. habilis.
 What Light Does Biology Shed? 153

This cultural variation may have, in turn, selected for genes that en-
dowed early humans with primitive moral emotions. Primitive moral
emotions would in turn have facilitated the evolution of more advanced
cultural codes of morality. Repeated rounds of gene-cultural coevolu-
tion would have gradually increased both the moral sense itself and the
systems of moral norms. That is, the evolution of morality would have
been directly promoted by natural selection in a process where the
moral sense and the moral norms would have coevolved.
The gene-culture coevolution account of the evolution of morality
is, of course, radically different from the theory I am advancing here, in
which moral behavior evolved not because it increased fitness but as a
consequence of advanced intelligence, which allowed humans to see
the benefits that adherence to moral norms bring to society and to its
members. The extreme variation in moral codes among recent human
populations and the rapid evolution of moral norms over short time
spans would seem to favor the explanation I am proposing. Gene-
culture coevolution would rather lead to a more nearly universal sys-
tem of morality, which would have come about gradually as our ho-
minid ancestors evolved toward becoming Homo sapiens.
Empathy, or the predisposition to mentally assimilate the feelings
of other individuals, has recently been extensively discussed in the con-
text of altruistic or moral behavior. Incipient forms of empathy seem to
be present in other animals. In humans, increasing evidence indicates
that we automatically simulate the experiences of other humans (Gaz-
zaniga 2005, 158– 99; 2008). Empathy is a common human phenome-
non, surely associated with our advanced intelligence, which allows us
to understand the harms or benefits that affect other humans, as well as
their associated feelings. Empathic humans may consequently choose
to behave according to how their behavior will affect those for whom
we feel empathy. That is, human empathy occurs because of our ad-
vanced intelligence. Humans may then choose to behave altruistically,
or not, that is, morally or not, in terms of the anticipated consequences
of their actions to others.
The question remains, when did morality emerge in the human lin-
eage? Did Homo habilis or Homo erectus have morality? What about
the Neandertals, Homo neanderthalensis? When in hominid evolution
 154 Francisco J. Ayala

morality emerged is difficult to determine. It may very well be that the

advanced degree of rationality required for moral behavior may only
have been reached at the time when creative language came about, and
perhaps in dependence with the development of creative language.
(When creative language may have come about in human evolution is
discussed in Cela-Conde and Ayala 2007).

M o ral Code s

I have distinguished between moral behavior— judging some actions

as good, others as evil— and moral codes—the precepts or norms ac-
cording to which actions are judged. Moral behavior, I have proposed,
is a biological attribute of H. sapiens, because it is a necessary conse-
quence of our biological makeup, namely, our high intelligence. But
moral codes, I argue, are not products of biological evolution, but of
cultural evolution.
It must, first, be stated that moral codes, like any other cultural sys-
tems, cannot survive for long if they prevailingly run in outright con-
flict with our biology. The norms of morality must be by and large con-
sistent with human biological nature, because ethics can exist only in
human individuals and in human societies. One might therefore also
expect, and it is the case, that accepted norms of morality will often, or
at least occasionally, promote behaviors that increase the biological fit-
ness of those who behave according to them, such as child care. But the
correlation between moral norms and biological fitness is neither nec-
essary nor indeed always the case: some moral precepts common in
human societies have little or nothing to do with biological fitness, and
some moral precepts are contrary to fitness interests.
How do moral codes come about? The short answer is, as already
stated, that moral codes are products of cultural evolution, a distinctive
human mode of evolution that has surpassed the biological mode, be-
cause it is a more effective form of adaptation: it is faster than biological
evolution, and it can be directed. Cultural evolution is based on cul-
tural heredity, which is Lamarckian, rather than Mendelian, so that ac-
quired characteristics are transmitted. Most important, cultural hered-
ity does not depend on biological inheritance, from parents to children,
 What Light Does Biology Shed? 155

but is transmitted also horizontally and without biological bounds. A

cultural mutation, an invention (think of the laptop computer, the cell
phone, or rock music), can be extended to millions and millions of in-
dividuals in less than one generation.
In chapter 5 of The Descent of Man, titled “On the Development
of the Intellectual and Moral Faculties during Primeval and Civilized
Times,” Darwin writes: “There can be no doubt that a tribe including
many members who, from possessing in a high degree the spirit of pa-
triotism, fidelity, obedience, courage, and sympathy, were always ready
to give aid to each other and to sacrifice themselves for the common
good, would be victorious over most other tribes; and this would be
natural selection. At all times throughout the world tribes have sup-
planted other tribes; and as morality is one element in their success, the
standard of morality and the number of well-endowed men will thus
everywhere tend to rise and increase” (159– 60).
Darwin is making two important assertions. First, he is saying that
morality may contribute to the success of some tribes over others,
which is natural selection in the form of group selection. Second, he is
asserting a position of moral optimism, namely, that the standards of
morality will tend to improve over human history precisely on grounds
of group selection, because the higher the moral standards of a tribe, the
more likely the success of the tribe. This assertion depends on which
standards are thought to be “higher” than others. If the higher standards
are defined by their contribution to the success of the tribe, then the as-
sertion is circular. But Darwin asserts that there are some particular
standards that, in his view, would contribute to tribal success: patriot-
ism, fidelity, obedience, courage, and sympathy.

M o ral N o rm s and Nat ural Selection

Parental care is a behavior generally favored by natural selection that

may be present in virtually all codes of morality, from primitive to
more advanced societies. There are other human behaviors sanctioned
by moral norms that have biological correlates favored by natural se-
lection. One example is monogamy, which occurs in some animal spe-
cies but not in many others. It is also sanctioned in many human cultures
 156 Francisco J. Ayala

but surely not in all. Polygamy is sanctioned in some current human

cultures and was to a greater extent in the past. Food sharing outside
the mother–offspring unit rarely occurs in primates, with the excep-
tion of chimpanzees— and, apparently, in capuchin monkeys (Bros-
nan and de Waal 2003) — although even in chimpanzees food sharing
is highly selective and often associated with reciprocity. A more com-
mon form of mutual aid among primates is coalition formation; alliances
are formed in fighting other conspecifics, although these alliances are
labile, with partners readily changing partners.
Those moral codes tend to be widespread that lead to successful
societies. Since time immemorial, human societies have experimented
with moral systems. Some have succeeded and spread widely through-
out humankind, like the Ten Commandments, although other moral
systems persist in different human societies. Many moral systems of
the past have surely become extinct because they were replaced or be-
cause the societies that held them became extinct. The moral systems
that currently exist in humankind are those that have been favored by
cultural evolution. They were propagated within particular societies
for reasons that might be difficult to fathom but that surely must have
included the perception by individuals that a particular moral system
was beneficial for them, at least to the extent that it was beneficial
for their society by promoting social stability and success (Gazzaniga
2005, 2008). Cultures, of course, do not evolve as completely differen-
tiated units. Rather, cultures often incorporate elements from other
cultures. “Far from being self-preserving monoliths, cultures are porous
and constantly in flux. Language . . . is a clear example” (Pinker 2002;
see Pinker 2011).
The norms of morality, as they exist in any particular culture, are
felt to be universal within that culture. Yet, similar to other elements
of culture, they are continuously evolving, often within a single genera-
tion. As Steven Pinker has pointed out, Western societies have recently
experienced the moralization and amoralization of diverse behaviors.
Thus, “smoking has become moralized [and is ] . . . now treated as im-
moral. . . . At the same time many behaviors have become amoralized,
switched from moral failings to lifestyle choices. They include divorce,
illegitimacy, working mothers, marijuana use and homosexuality”
 What Light Does Biology Shed? 157

(Pinker 2002, 34). Acceptance by individuals or groups of particular

sets of moral norms is often reinforced by civil authority (e.g., those
who kill or commit adultery will be punished) and by religious beliefs
(God is watching and you’ll go to hell if you misbehave). But it is
worth noticing that the legal and political systems that govern human
societies, as well as the belief systems held by religion, are themselves
outcomes of cultural evolution, as it has eventuated over human his-
tory, particularly over the last few millennia.

R e f ere nce s

Barkow, J., L. Cosmides, and J. Tooby, eds. 1992. The Adapted Mind: Evolu-
tionary Psychology and the Generation of Culture. Oxford: Oxford Uni-
versity Press.
Blackmore, S. 1999. The Meme Machine. Oxford: Oxford University Press.
Bok, H. 1998. Freedom and Responsibility. Princeton: Princeton University
Press.
Brosnan, S., and F. de Waal. 2003. “Monkeys Reject Unequal Pay.” Nature 425:
297– 99.
Cela-Conde, C. J., and F. J. Ayala. 2007. Human Evolution: Trails from the
Past. Oxford: Oxford University Press.
Copp, D., ed. 2006. The Oxford Handbook of Ethical Theory. Oxford: Ox-
ford University Press.
Darwin, C. R. 1871. The Descent of Man, and Selection in Relation to Sex. Lon-
don: John Murray. Reprint New York: Appleton and Company, 1971.
Ekstrom, L. 2000. Free Will: A Philosophical Study. Boulder, CO: Westview
Press.
Fischer, J. M. 2006. “Free Will and Moral Responsibility.” In The Oxford
Handbook of Ethical Theory, ed. D. Copp, 321–24. Oxford: Oxford Uni-
versity Press.
Gazzaniga, M. S. 2005. The Ethical Brain. New York: Dana Press.
———. 2008. Human: The Science behind What Makes Us Unique. New York:
HarperCollins.
Greene, J. D., and J. Haidt. 2002. “How (and Where) Does Moral Judgment
Work?” Trends in Cognitive Science 6: 517– 23.
Greene, J. D., R. B. Sommerville, L. E. Nystrom, J. M. Darley, and J. D. Cohen.
2001. “An fRMI Investigation of Emotional Engagement in Moral Judg-
ment.” Science 293: 2105– 8.
 158 Francisco J. Ayala

Haidt, J. 2007. “The New Synthesis In Moral Psychology.” Science 316:

998–1002.
Haidt, J., F. Bjorklund, and S. Murphy. 2000. “Moral Dumbfounding: When
Intuition Finds No Reason.” Working Paper, University of Virginia.
Hauser, M. 2006. Moral Minds: How Nature Designed Our Universal Sense
of Right and Wrong. New York: HarperCollins.
Huxley, J. 1953. Evolution in Action. New York: Harper.
Huxley, T. H., and J. Huxley. 1947. Touchstone for Ethics. New York: Harper.
Kane, R. 1996. The Significance of Free Will. Oxford: Oxford University Press.
Maienschein, J., and M. Ruse, eds. 1999. Biology and the Foundations of Eth-
ics. Cambridge: Cambridge University Press.
Martineau, H. 1832– 34. Illustrations of Political Economy, 3rd ed., in 9 vols.
London: Charles Fox.
Paley, W. 1785. The Principles of Moral and Political Philosophy. Dublin Exshaw.
Pinker, S. 2002. The Blank Slate: The Modern Denial of Human Nature. New
York: Viking.
———. 2011. The Better Angels of Our Nature: Why Violence Has Declined.
New York: Viking.
Richerson, P. J., and R. Boyd. 2005. Not by Genes Alone: How Culture Trans-
formed Human Evolution. Chicago: University of Chicago Press.
Richerson, P. J., R. Boyd, and J. Henrich. 2010. “Gene-Culture Coevolution in
the Age of Genomics.” Proc. Natl. Acad. Sci. USA 107 (Suppl. 2): 8985– 92.
Ruse, M. 1995. Evolutionary Naturalism. London: Routledge.
Ruse, M., and E. O. Wilson. 1985. “The Evolution of Ethics.” New Scientist 108:
50– 52.
Sober, E., and D. S. Wilson. 1998. Unto Others: The Evolution and Psychology
of Unselfish Behavior. Cambridge, MA: Harvard University Press.
Strimling, P., M. Enquist, and K. Eriksson. 2009. “Repeated Learning Makes
Cultural Evolution Unique.” Proc. Natl. Acad. Sci. USA 106: 13870 – 74.
Waddington, C. D. 1960. The Ethical Animal. London: Allen and Unwin.
Wilson, E. O. 1975. Sociobiology, the New Synthesis. Cambridge, MA: Belknap
Press.
———. 1978. On Human Nature. Cambridge, MA: Harvard University Press.
———. 1998. Consilience: The Unity of Knowledge. New York: Knopf.
———. 2012. The Social Conquest of Earth. New York: Norton/Liveright.
 C h a p t e r S e v e n

What Is the Nature of Perception?

-
Zygmunt Pi z l o

There have been two broad definitions of perception. The conventional,

largely content-free definition says that “to perceive means to be aware
of the external world through the medium of the senses.” This defini-
tion is rooted in Fechnerian psychophysics in which the percept is the
passive result of a causal chain of events (Fechner [1860] 1966). Fech-
ner recognized that studying perception requires establishing relations
among three different types of phenomena. One is the distal stimulus
(an object “out there”), which is a physical phenomenon (event) de-
scribed in the language of physics and geometry, such as weight, iner-
tia, surface reflectance, stiffness, size, and shape. This also holds for the
proximal stimulus, which is a physical distribution of the stimulus en-
ergy on the surface of the receptors. In visual perception, the proximal
stimulus is a distribution of light on the surface of the retina, where
the light energy is transduced (translated) into the bioelectrical messages
used by the nervous system. In auditory perception, the proximal stimu-
lus is a distribution of mechanical energy on the basilar membrane in
the inner ear, and in tactual perception, the proximal stimulus is a dis-
tribution of mechanical energy on the surface of the skin. Taste and
smell are different; the proximal stimulus is a distribution of chemical

159
 160 Zygmunt Pizlo

energy. Receptors in all the sensory systems translate the energy of the
stimulus into the electrical energy used in the nervous system. It is at
this stage, at the proximal stimulus, that physiological, rather than
physical, phenomena come into play, but it is only after the stimulus is
analyzed in the brain that the percept, a mental phenomenon, arises.
When the Fechnerian causal chain is described in this way, the events
between the distal stimulus and the percept can be characterized as a
sequence of “pushes,” a deterministic process in which the knowledge
of the cause allows one to compute its effect.
A more recent, and quite different, definition states that “percep-
tion is an inference that provides the observer with accurate information
about the external world.” This definition is very similar to Thomas
Reid’s ([1764] 2000) understanding of the nature and role of perception.
Why am I calling the percept an “inference”? I am calling it an infer-
ence because sensory data is never sufficient to completely describe an
object “out there.” This is particularly clear in visual perception because
the proximal stimulus, a retinal image, is two-dimensional (2D), but the
distal stimulus “out there” in the physical world is three-dimensional
(3D). The mapping from the object (model) to its image (data) is many-
to-one. Such problems in the contemporary theory of inverse problems
are called “forward” or “direct” problems. Under this rubric, the oppo-
site problem, mapping from the data to the model is one-to-many, and
it is called an “inverse” problem. Because of the nature of the mapping,
the inverse problem, unlike the forward problem, is ill-posed and/or
ill-conditioned. In plain English, this means that the solution of the in-
verse mapping may not exist, and if it does exist, it may not be unique,
and when it is unique, it will be unstable in the presence of noise and
uncertainty in the data. It follows that solving an inverse problem of
perceptual inference is not easy. The only way to produce a unique and
accurate interpretation is to impose a priori constraints on the family
of possible interpretations (the nature and role of these constraints are
discussed later). Reid’s understanding of what he meant by perception
is relevant here because he was the first to be credited with distin-
guishing “sensations” from “perceptions.” Sensations, in his usage, do
not refer to any object, making it impossible to talk about their veridi-
cality. Perceptions, on the other hand, always refer to an object, and
 What Is the Nature of Perception? 161

they are always veridical. In the psychology of perception, we often

use the term veridicality when we refer to the fact that the percept
agrees with (is a faithful copy of) things “out there.” It is easy to see
that Reid’s sensations correspond, in contemporary language, to what
we call “sensory coding,” the activities described as a forward problem
(above). The perception of hue, brightness, and loudness is a good ex-
ample of sensory coding (forward problems). Reid’s perceptions are
different. They correspond to the kind of perceptual inference described
as an inverse problem (above). The perception of 3D shapes and 3D
scenes are good examples of such inverse problems (inferences). Reid
insisted that perceptions are accurate, but he could not explain how
this is accomplished. He did know that for this to happen the human
mind must have some innate, hardwired intuitions of space, time, and
causality. Reid anticipated Kant (1781) when he published these views
seventeen years before Kant did. For skeptics, those who eschewed a
role for in-built intuitions such as Bishop Berkeley ([1709] 1910), Reid
offered this oft-cited rebuttal: “I have never seen a skeptic who would
walk into the fire because he did not trust his senses.” Unfortunately,
Reid’s views on perception, with his emphasis on the importance of
veridicality and arguments against a role for illusions in the study of vi-
sual perception, have largely been forgotten.

Can We Trus t and Can We Study

O ur Pe rce p t ions?

Perception is the oldest branch of experimental psychology. Systematic

observations and theorizing about the nature and veridicality of per-
ception started at least a thousand years ago (Alhazen [1083] 1989). This
should come as no surprise. There was very good reason for the natural
philosophers, physicians, and alchemists, who would become physi-
cists, astronomers, and physiologists, to wonder about the veridicality
of their perceptions. For most of recorded history, scientific observa-
tions were made simply by looking at, listening to, touching, smelling,
and tasting things. If you could not be sure that you were perceiving
(or sensing) things accurately, you would not be able to take your data
 162 Zygm unt Pizlo

seriously. Fortunately, their commitment to taking their observations

seriously was buttressed by their common sense (another concept em-
phasized by Reid), so the early physicists and astronomers had no dif-
ficulty convincing themselves that our perceptions are, or are almost
always, veridical. This allowed them to lay the foundations of modern
science. Perceptionists, who jumped on this bandwagon shortly after
the Renaissance and stayed there until the Age of Enlightenment, man-
aged to fall off, and stay off, since then. It is surprising, if not simply
irrational and self-contradictory, that many contemporary perception-
ists who argue that perceptions are never veridical trust the data they
collect. How do they manage to use their senses in designing their ex-
periments and collecting and interpreting their data, then go on to re-
fusing to accept the fact that their subjects’ perceptions can be veridi-
cal? Paraphrasing Reid, what possible basis do you have to trust your
reasoning about perception but not to trust your perception itself? Both
are produced by the same wetware!
But there is even a more fundamental question that needs to be
examined. Once we are willing to assume that we can trust our per-
ception, the question remains as to whether we can study perception
“scientifically”? Why is this a problem? Well, perceptions, as mental
events, are “private” in the sense that each of us has a special, privileged
access to our thoughts and perceptions. I know that I think (at least
part of the time), but there is no objective way I can verify that the
reader of this chapter ever experiences “thinking.” Actually I, and I
suspect you, take it for granted that most if not all other people think
and perceive in the same way you do. This belief is often supported by
the following kind of folk argument (reasoning), the walking, talk-
ing duck metaphor you have heard so often. Other people look like
me, and they react to events as I do (or as I might). When I talk to
them, they respond as I would. In other words, I can easily put myself
in their shoes. This is why I believe that other people have perceptions
and thoughts like mine. But assuming that other people think and per-
ceive as we do does not tell us how to study them scientifically. Science
can only deal with “public” events, so in order to do science when we
study perception, we must obtain behavioral data (activities we can ob-
serve) about people’s perceptions. We ask an observer (now called a
 What Is the Nature of Perception? 163

“subject”) to compare two objects and respond (tell us) whether they
are the same or different. This is what Brindley (1960) called a “Class
A” experiment. The scientist will infer the subject’s “private” percep-
tions from these behavioral (“public”) responses. It follows that study-
ing perception can, in engineering terminology, be called a “black box”
problem. This problem is difficult because the subject’s responses al-
ways confound perceptions with response biases produced by the sub-
ject’s preferences, opinions, guesses, and even social pressures. Fortu-
nately, we have an experimental methodology, called signal detection
(Green and Swets 1966), that allows the scientist to estimate percep-
tions and response biases separately. So folk arguments about percep-
tion have been replaced by signal detection experiments that allow one
to study perception scientifically. But note that if a signal detection
experiment is not used to study perception, and not everyone uses it
today, the possibility remains that the result of such efforts does not
belong in science.

Nat i v i s m v e rsus Empiricism

The origin of perceptual mechanisms comes next on our list of ques-

tions about the nature of perception. The empiristic view in percep-
tion is expressed well by what is called the “peripatetic axiom,” which
states, “Nihil est in intellectu quod non prius in sensu” (There is noth-
ing in the mind that was not in the senses first). So, for those percep-
tionists who are empiricists, the mind of a newborn human baby is
like a blank slate, and the 3D percepts of an adult human being, which
are produced by a 2D retinal image, are the result of comparing the
adult’s retinal data with his memory traces of previous images. So we
are not able to recover 3D shapes and scenes when we are born. We
must learn how to do this through dozens and hundreds and even thou-
sands of interactions with objects in the external world. It was also as-
sumed that learning to perceive required motor actions and active hap-
tic exploration (manipulating objects with one’s hands). Note that if
perceptions actually had to be learned this way, people who grew up in
different parts of the world would see objects differently. People who
 164 Zygmunt Pizlo

had never had a chance to handle a quarter-inch drill as they looked at

it, or a violin, would know nothing about their shapes. These days we
know quite well that there are actually very, very few individual differ-
ences in perception. We know that the role of personal experience in
perception is negligible. This fact makes it practical, as well as profitable,
for a high-tech manufacturer to move production of his product to a re-
gion where indigenous people are still using wooden plows and spears.
Furthermore, if we actually had to learn how to perceive veridically,
we would probably have to practice seeing to maintain it and to im-
prove our ability to use it. This, obviously, is not the case. Now, con-
sider the well-known Müller-Lyer illusion shown in figure 7.1. The
line segment on the right appears to be longer than the segment on the
left by about 15 percent. The magnitude of this illusion is roughly the
same for everyone, and it does not get smaller if the subject keeps look-
ing at it and even when he is allowed to measure the lengths with a ruler.
No matter how hard you think about the fact that the length of these
two lines is identical and no matter how many times you measure them,
the magnitude of the illusion remains the same. Empiricists have tried to
remove this illusion since it came into vogue in the nineteenth century
by asking subjects to make long series of saccadic (scanning) eye move-
ments among the three corners in the display. They reported small re-
ductions in the illusion’s size that did not persist beyond the experi-
mental session, and, to my knowledge, even these small effects have
not been confirmed with signal detection methodology. This illusion,
which is robust, as well as vivid, provides convincing evidence against
a role for learning in visual perception. It is hard to understand why em-
piricism has persisted in visual perception as long as it has despite such
observations.
The commitment to perceptual learning, characteristic of the em-
piristic point of view, usually goes together with a commitment to an
important role for cognitive and social factors in perception. This idea
loomed large just after World War II when what was called the New
Look movement in perception came into vogue (Bruner 1957, 1973;
Bruner and Goodman 1947). This fad was supposedly “supported” by
the following kinds of observations: (i) poor, but not rich, children in
Boston were found to overestimate the sizes of coins, and (ii) percep-
 What Is the Nature of Perception? 165

Figure 7.1. Müller-Lyer illusion. The two horizontal line segments demarcated
by the three corners are exactly the same length, but the line on the right looks
longer than the line on the left.

tual thresholds were higher for taboo words, including such threaten-
ing words as penis and KOTEX. This fad still has adherents as evi-
denced by recent papers showing that (iii) a hill looks steeper when a
backpack is heavy and (iv) a glass of water appears farther away when
the observer is thirsty. We now know that all of the effects “discovered”
by New Lookers derive from bad experimental methodology combined
with bad statistical analyses. All of these effects go away completely
when the experiments are done correctly. Unfortunately, empiricism in
perception shows no signs of disappearing, despite the recognition of
these problems in psychology. It is actually gaining popularity thanks
to the “contributions” of our colleagues from engineering. Most con-
temporary efforts in computer vision and artificial intelligence focus
on learning. We cannot blame these engineers for their scientific igno-
rance. There is nothing in the traditional engineering curriculum, un-
like the curriculum in experimental psychology, that requires gradu-
ate students to read the history of sensation and perception. I will add
a few classical results here that speak against empiricism for the bene-
fit of engineers who do not have this background, namely, a newborn
chicken has good depth perception: it will not jump off a post one meter
above the ground plane, and the time it takes for it to decide to jump
varies directly with the height of the post (Thorndike 1899); a new-
born chick’s monocular and binocular space perception is accurate: it
is good enough to allow it to find and peck accurately at grains scat-
tered on the ground (Hess 1956); a newborn human infant perceives
both visual and auditory directions in space accurately (Wertheimer
1961); a crawling human infant has good depth perception: it prevents
it from crawling off a “visual cliff” (Gibson and Walk 1960); and new-
born infants recognize planar (flat) shapes slanted differently in 3D
 166 Zygmunt Pizlo

space (Slater and Morrison 1985). These examples should be sufficient

to encourage engineers, as well as others, who want to study percep-
tion, and who come to it with an empiristic bias, to at least look into
what is already known about the problems inherent in empiricism.
Those of us who have this knowledge are convinced that perceptual
mechanisms are much more likely to be innate than learned. We also
know that it is very difficult (if possible at all) to demonstrate appre-
ciable and significant effects of learning in perception. If perceptions
are innate and hardwired, all humans are likely to see things the same
way, and there should be few if any qualitative individual differences
in perception. I believe that both of these claims are true. This verity,
once accepted, provides a useful tool for verifying whether an experi-
mental result reflects the operation of the perceptual system that is not
confounded with response biases. If results of several subjects, both
naive and not naive, are very similar, the confounding effects of response
bias are likely to be absent. An important caveat follows from this ob-
servation, namely, the results of individual subjects should be reported
in perception experiments. Subjects’ results should not be combined
before they are averaged. Unfortunately, this kind of averaging is still
an all too common practice. It should be discontinued because it is the
absence, or at least the relative insignificance, of individual differences
in the data that lends confidence to any conclusions drawn about what
has been perceived.
There is an additional, surprising difference between empiricists
and nativists. Nativists assume that the visual system uses computa-
tional algorithms, while empiricists assume that the visual system uses
look-up tables. Those who advocate perceptual learning are simply not
interested in discovering the perceptual algorithms at work in the vi-
sual system. According to them, perceptual learning depends entirely
on getting more and more experience with the visual world. This ex-
perience translates into better and better “priors” (likely outcomes).
This is the only role for experience in the empiristic approach. Empiri-
cists believe that computational algorithms are superfluous. The Bayes
rule, which is used in the contemporary empirical approach, is “the”
mechanism. This observation brings us to considering why perception
should be viewed as an inverse problem and how it should be studied
once this is done.
 What Is the Nature of Perception? 167

Th e Nat ure o f A Pri o ri Constraints in Solving

an I nv e rs e Pro b l e m in Perception

Consider the 2D image of the 3D indoor scene shown in figure 7.2.

There are several pieces of furniture in the center of the floor, a floor
that was chosen because its texture was quite complex visually. The
first task for the visual system is to determine whether there is an ob-
ject “out there,” how many of them are there, and where are they lo-
cated. This analysis is called figure-ground organization (FGO). FGO
is not an easy task, in no small measure, because the human retina has
six million receptors (cones), the receptors used for the perception of
shape and details. Without any a priori constraints, the visual system
would have to try all partitions of the set of six million receptors. The
number of partitions of a set with N elements is called a Bell number
(BN). The number of partitions BN of N elements grows faster with N
than 10N but slower than N factorial (N!). Consider a low-resolution
camera with only 10x8 pixels. Not much can be done with so few re-
ceptors, but to appreciate the enormity of the task at hand, you must
realize that the number of all partitions of eighty receptors is already
larger than the number of atoms in the universe, which is currently esti-
mated to be 1080. The number of partitions of six million receptors is be-
tween 106,000,000 and 1038,000,000, which is many, many orders of magnitude
larger than the number of atoms in the universe. One cannot even begin
to try to imagine how big this number is. Even if you could evaluate
billions of partitions each second, there would not be enough time to
train the 10x8 camera during the 13.7 billion years that have elapsed
since the Big Bang. There is simply no way to take Lotze’s ([1852]
1886) theory of local signs seriously. By way of reminder, Lotze’s em-
piristic theory of local signs says that when the human infant is born,
its visual system already knows that there are individual receptors and
that its visual system can also distinguish among the stimulations of
the different receptors, but the visual system at birth has no spatial in-
formation about the world “out there.” Even, a straight line segment
on the retina is not “understood” by a newborn infant as a straight line
segment, nor as any other geometrical entity.
According to Lotze, and then according to Helmholtz ([1910] 2000),
and then to Hebb (1949), the visual system learns the meaning of all
 168 Zygm unt Pizlo

Figure 7.2. One of

the 3D indoor scenes
used to test our human
subjects and our
robot, Čapek.

spatial arrangements of visual stimuli. Lotze, Helmholtz, and Hebb

simply forgot to run the numbers. Permutations had been known for at
least a century when Helmholtz wrote the first edition of his Handbüch
in 1866. He apparently never thought to check the number of permuta-
tions that would be required to perform the computations required by
any empiristic theory of vision. Had he done this, he would surely have
realized that each of us must have some writing on our “slates” when
we are born. Were the tabula rasa anywhere near blank or clean, the task
of training feature detectors in our visual system would be hopeless. This
should give pause to anyone thinking about or getting into the learning
of natural images or into natural scene statistics (Fei-Fei and Perona
2005; Long and Purves 2003; Simoncelli and Olshausen 2011). Learn-
ing particular configurations or features is an impossible approach. The
only way to go is to extract abstract characteristics of the visual stimulus
such as smooth pieces of curves, colinearity and the parallelism of edges,
and 3D symmetry by a geometrical analysis of the stimulus, not by
using look-up tables. Algorithms, not look-up tables, is how the visual
system works.
Conventionally, FGO is considered to be the first step in visual pro-
cessing, whose output is used for higher-level computing, such as the re-
covery of 3D shape and space. As you will see, this is not how things ac-
tually work, so I will discuss the nature of the 3D recovery problem
before saying anything more about FGO. We humans evolved in a
three-dimensional space well before the physicists added the eight di-
mensions needed in their theories, so, if evolution taught us anything
 What Is the Nature of Perception? 169

Figure 7.3. “Shepard boxes.” The two parallelograms representing the top faces
of these boxes are geometrically identical on this page. The only difference is
that one of them has been rotated through 90°. They look very different because
they are perceived as residing in 3D, not in 2D. From Pizlo 2008.

about the world “out there,” we surely learned that it had three dimen-
sions and this important fact was surely written on our slates before we
were born. The presence, as well as the compelling nature, of this infor-
mation can be demonstrated easily because you always see objects as
three-dimensional. Look at figure 7.3. It shows two boxes whose top
faces (2D parallelograms) are identical on the 2D page of this book (ex-
cept for a 90° rotation on this page), but they do not look the same. The
fact that these two parallelograms are geometrically identical is easy to
verify by cutting a parallelogram that matches one and then checking
that it matches the other. You do not see things in the 2D space of the
page in front of you, but note that this is not an illusion. The two 3D
boxes are really different, so when you see them as different your per-
cept is actually veridical, not illusory! Your visual system recovered the
3D shapes in a single computational step without any access to the 2D
image on the page or on your retina. We cannot see our retinal images
no matter how hard we try.
Recall that the 3D recovery from a 2D image is a difficult inverse
problem. It is difficult because each retinal point could have been pro-
duced by any of the infinite number of points on the projecting ray of
light (Berkeley [1709] 1910). Now that we appreciate this, let us make
some simplifying assumptions. First, assume that the visual system
 170 Zygm unt Pizlo

can produce a 3D recovery by using ten bits of information for the di-
rection of depth. This amount of information permits a precision of
one millimeter for an object whose range in depth is one meter. Next,
if the object occupies the central 20° of the visual field, as many as
three million cones will be stimulated (one-half of all the cones on the
retina). It follows that for any given 2D image of a 3D object there are
10003,000,000 possible 3D interpretations. This amounts to 109,000,000 inter-
pretations. This number is not very different from the number estimated
earlier in the discussion of the computational complexity of FGO. Both
problems seem to be computationally impossible. This seeming curse
of computational intractability could actually be a blessing for a vision
scientist. The veridical perception of 3D shapes and scenes is so very
difficult that it seems possible that there might be only one way to solve
such a problem. If this is true and if you can develop a theory of vision
that actually solves it, it seems likely that this theory would be the theory
of vision. We all know that this “impossible” problem can be solved
because our visual system does it nearly perfectly every day. We do re-
cover 3D shapes from 2D images veridically. Until recently, no one knew
how this was done, but now that we have built a machine that can see
like us, we have moved a bit closer to having the theory of vision (see
Li at al. 2011, 2012; Pizlo et al. 2010, 2014).
So how does our machine work, and can the answer to this ques-
tion help us understand how the human visual system works, too? I
will start by pointing out that the only way to produce a unique and
veridical perceptual interpretation is to use a priori knowledge about
what we perceive. This idea was first emphasized by the Gestalt psy-
chologists (Koffka 1935; Köhler 1920; Wertheimer [1923] 1958). They
conjectured that the visual percept is the simplest possible interpreta-
tion of the retinal image. The rest of this section illustrates why and
how this claim works.
Figure 7.4 is a 2D image of a 3D cube. All human observers will see
a 3D cube within a fraction of a second rather than any of the remain-
ing 109,000,000-1 possible 3D interpretations. We know that no look-up
table is, or could be, used in such a case. There simply is not enough
time to build a useful look-up table for even a fraction of the possible
3D interpretations: and even if sufficient time could be provided, there
 What Is the Nature of Perception? 171

is no way that all humans, old and young, born in Asia, Europe, or the
Americas, would have at their disposal an identical look-up table per-
mitting them to see the same 3D object. And even if, miraculously, all
of our look-up tables were identical, what chances are there that the re-
sulting percept would have been a perfect 3D cube! There are no perfect
cubes in the physical environment, so if experience was actually driving
our perceptions, we could never perceive perfectly regular (symmetri-
cal) shapes. Plato knew this well. Plato, and those who agreed with him,
namely, Descartes, Reid, and the Gestalt psychologists, held that uni-
versals such as symmetry can exist independently from particulars like
the characteristics possessed by concrete objects. This should surprise
no one, because symmetry, as a mathematical concept, can be defined
(formulated) without ever seeing a single object. What might be sur-
prising to a reader but is not surprising to me is that our 3D percep-
tions of concrete objects are driven by abstract universals rather than
by experience with such objects (i.e., particulars). This is the only pos-
sible way for us to perceive 3D scenes and 3D objects veridically (see
Pizlo 2008 for support of this claim). In the physical world of objects
where symmetry is the rule rather than the exception, the best, per-
haps the only, way to deal with the prohibitive computational com-
plexity inherent in vision is to use symmetry to recover the 3D shapes
of objects. Symmetry, the abstract mathematical construct, permits
the visual system to use a single universal prior, instead of an empiri-
cally derived prior that would have to be defined for all possible ob-
jects, all possible depths, all possible 3D viewing orientations, and all
possible partitions of the retinal image. While working on this prob-
lem, we discovered that if the 3D object is not perfectly symmetrical,
the visual system has the ability to add a correction, but it will only do
this if it matters. This becomes obvious once one realizes how easy it
is to recognize the asymmetrical posture of a flamingo standing on
one leg and how difficult it is to detect the asymmetry present in all
human faces. So, contrary to what empiricists, beginning with Aris-
totle, claimed, not everything that is in the mind had to be in the senses
first. Concrete objects are instantiations of the abstract concept of
symmetry rather than symmetry being abstracted (derived) from par-
ticular objects.
 172 Zygm unt Pizlo

Figure 7.4. An orthographic image of a cube. When your line of sight is orthogo-
nal (at 90°) to the plane of this page and your viewing distance is at least 10 times
the size of the image, this 2D image will produce the percept of a 3D cube.

The result of implementing this approach in a robot that can see

is shown in figure 7.5 (Li et al. 2012). A scene containing many com-
mon objects is at the top. This scene was viewed and interpreted by
our robot, called Čapek. The result of Čapek’s interpretation of this
scene is shown on the bottom, with the floor plan on the left and the
regions in the image that Čapek chose to represent, where individual
objects were located, shown on the right. More examples of scenes re-
covered by Čapek’s visual system can be found at http://web.ics.purdue
.edu/~li135/ObjDetect.html. Note that Čapek, like us, solved the figure-
ground organization problem in this 3D scene and in its 2D image in a
fraction of a second, using cameras whose resolution was only 600x400
pixels. These results suggest that we are on our way to having a theory
of vision that actually works. I also suspect that this is the road that
will end with the theory of human vision, alluded to earlier. I am some-
what confident about this claim because we have obtained considerable
psychophysical evidence to support it. This evidence comes from both
3D shape and 3D scene recovery experiments, as well as from the more
traditional signal detection experiments. I have reason to believe that
the newer recovery experiments provide a better way to study percep-
tions scientifically than the older signal detection approach. The new
recovery methodology works as follows. You begin by specifying a
computationally intractable problem of perception, like perceiving 3D
 What Is the Nature of Perception? 173

Figure 7.5. Top: A 2D image of a 3D scene containing children’s furniture. Bottom

left: A top view of the 3D scene shown on top. The black rectangles represent the
actual sizes and positions of furniture. The gray rectangles represent Čapek’s per-
ception of the individual pieces of furniture. Bottom right: The white polygons
indicate the distinct 2D regions Čapek “saw” as representing individual objects.

shapes and 3D scenes. You then test human observers in the task and
verify that they can solve it perceptually, quickly, and accurately, de-
spite the fact that this task cannot be solved mathematically. With this
done, you formulate a computational algorithm (model) that transforms
this ill-posed problem into a well-posed problem by using effective a
priori constraints. You then implement the algorithm in a machine,
testing the model with the same 3D scenes and 3D shapes you used
with your observers. Then (i) if the performance of your computational
model is as good as the performance of your subjects, (ii) if the com-
putational complexity of your model is similar to the computational
complexity of the perceptual mechanisms, and finally (iii) if the model
 174 Zygm unt Pizlo

fails in the same way as the humans when faced with degenerate view-
ing conditions, you can be sure that you have a model of the observer’s
perceptual mechanisms rather than his response biases. This new ap-
proach to studying and explaining perception is very similar to what
Craik (1943) advocated seventy years ago, around the time when the
first computers were being built. Craik realized even then that com-
puter simulation might become the method of choice for explaining
cognitive functions. The time has come to concentrate on implement-
ing Craik’s suggestion.

Let me conclude by summarizing how we understand perception today.

Perception is an inference that provides the observer with a veridical
representation of the external world. Illusions almost never happen in
everyday life. Perception is based on algorithms, not look-up tables.
These algorithms are automatically applied to the incoming sensory
stimulation. This automaticity gives us the impression that perception
is easy and effortless. Computationally, it is not. Perception algorithms
are innate; we are born with them. It is possible, perhaps even likely,
that some parameters of these algorithms are estimated or tuned after
birth. This makes sense because the genes that convey the information
about these algorithms are communication channels with finite infor-
mation capacity. The parameters may have to be tuned to varying input
as the child grows up and ages. It is very unlikely, however, that the al-
gorithms themselves are either learned or modified through learning
during life. This permanence and innateness allows all human beings
to perceive the external world the same way from their first day to their
last. This is crucial for successful interaction among humans and with
objects in the world “out there.” The key to understanding how per-
ception algorithms work is to recognize the critical role a priori sim-
plicity constraints play in perception. The sensory data that come in
are always ambiguous, and they are subjected to the simplest possible
interpretation. It so happens that the simplest interpretation is almost
always the correct interpretation. This predilection for choosing the
simplest interpretation suggests that the human mind is solving an “op-
timization problem.” Unlike the numerical methods most scientists
 What Is the Nature of Perception? 175

use for solving optimization problems with conventional computers,

the human mind uses physiological mechanisms that probably imple-
ment what is called the “least action” principle of physics (Lanczos
1970). The concept of optimization, which is widely used in explaining
physical and biological systems, as well as in designing engineering sys-
tems, is becoming popular in explaining cognitive systems, where it is
likely to shed light on the relationship between the mind and the brain.
It may also demystify the perennial question about whether machines
can see and think.

Not e

The author is grateful to Robert M. Steinman for suggestions and comments

on earlier versions of this essay. But even more important, I am grateful to
Professor Steinman for directing my interest to the history of science and em-
phasizing the importance of reading firsthand sources. He also convinced me
that I should not be afraid to think about the most fundamental aspects of sci-
ence and philosophy.

R e f ere nce s

Alhazen. [1083] 1989. The Optics. Books 1– 3. Trans. A. I. Sabra. London: War-
burg Institute.
Berkeley, G. [1709] 1910. A New Theory of Vision. New York: Dutton.
Brindley, G. S. 1960. Physiology of the Retina and Visual Pathway. Baltimore:
Williams & Wilkins.
Bruner, J. S. 1957. “Going beyond the Information Given.” In Contempo-
rary Approaches to Cognition, ed. J. S. Bruner, E. Brunswik, L. Festinger,
F. Heider, K. F. Muenzinger, C. E. Osgood, and D. Rapaport, 41– 69.
Cambridge, MA: Harvard University Press.
———. 1973. Beyond the Information Given. New York: Norton.
Bruner, J. S., and C. C. Goodman. 1947. “Value and Need as Organizing Fac-
tors in Perception.” Journal of Abnormal and Social Psychology 42: 33– 44.
Craik, K. J.W. 1943. The Nature of Explanation. Cambridge: Cambridge Uni-
versity Press.
Fechner, G. [1860] 1966. Elements of Psychophysics. New York: Holt, Rinehart
& Winston.
 176 Zygm unt Pizlo

Fei-Fei, L., and P. Perona. 2005. “A Bayesian Hierarchical Model for Learning
Natural Scene Categories.” In IEEE Conference on Computer Vision and
Pattern Recognition, vol. 2: 524 – 31. San Diego, CA: IEEE Computer
Society.
Gibson, E., and R. Walk. 1960. “The ‘Visual Cliff.’” Scientific American 202:
64 – 71.
Green, D. M., and J. A. Swets. 1966. Signal Detection Theory and Psychophysics.
New York: Wiley.
Hebb, D. O. 1949. The Organization of Behavior. New York: Wiley.
Helmholtz, H. von. [1910] 2000. Treatise on Physiological Optics. Trans. J. P. C.
Southall. Bristol: Thoemmes.
Hess, E. H. 1956. “Space Perception in the Chick.” Scientific American 195:
71– 80.
Kant, I. [1781] 1990. Critique of Pure Reason. New York: Prometheus Books.
Koffka, K. 1935. Principles of Gestalt Psychology. New York: Harcourt, Brace.
Köhler, W. [1920] 1938. “Physical Gestalten.” In A Source Book of Gestalt Psy-
chology, ed. W. D. Ellis, 17– 54. New York: Routlege & Kegan.
Lanczos, C. 1970. Variational Principles of Mechanics. Toronto: University of
Toronto Press.
Li, Y., T. Sawada, L. J. Latecki, R. M. Steinman, and Z. Pizlo. 2012. “A Tutorial
Explaining a Machine Vision Model That Emulates Human Performance
When It Recovers Natural 3D Scenes from 2D Images.” Journal of Mathe-
matical Psychology 56: 217– 31.
Li, Y., T. Sawada, Y. Shi, T. Kwon, and Z. Pizlo. 2011. “A Bayesian Model of
Binocular Perception of 3D Mirror Symmetric Polyhedra.” Journal of
Vision 11 (4): 11, 1– 20.
Long, F., and D. Purves. 2003. “Natural Scene Statistics as the Universal Basis
of Color Context Effects.” Proceedings of the National Academy of Sci-
ences 100: 15190– 93.
Lotze, H. [1852] 1886. Outlines of Psychology. Boston: Ginn.
Pizlo, Z. 2008. 3D Shape: Its Unique Place in Visual Perception. Cambridge,
MA: MIT Press.
Pizlo, Z., Y. Li, T. Sawada, and R. M. Steinman. 2014. Making a Machine That
Sees Like Us. New York: Oxford University Press.
Pizlo, Z., T. Sawada, Y. Li, W. G. Kropatsch, and R. M. Steinman. 2010. “New
Approach to the Perception of 3D Shape Based on Veridicality, Com-
plexity, Symmetry and Volume.” Minireview. Vision Research 50: 1–11.
Reid, T. [1764] 2000. An Inquiry into the Human Mind and the Principles of
the Common Sense. University Park: Pennsylvania State University Press.
Simoncelli, E. P., and B. A. Olshausen. 2001. “Natural Image Statistics and Neu-
ral Representation.” Annual Review of Neuroscience 24: 1193–1216.
 What Is the Nature of Perception? 177

Slater, A., and V. Morrison. 1985. “Shape Constancy and Slant Perception at
Birth.” Perception 14: 337– 44.
Thorndike, E. 1899. “The Instinctive Reactions of Young Chicks.” Psychologi-
cal Review 6: 282– 91.
Wertheimer, M. [1923] 1958. “Principles of Perceptual Organization.” In Read-
ings in Perception, ed. D. C. Beardslee and M. Wertheimer, 115– 35. New
York: D. van Nostrand.
———. 1961. “Psychomotor Coordination of Auditory and Visual Space at
Birth.” Science 134: 1692.
 P A R T I V

-
Introspection and Understanding
in the Humanities
 C h a p t e r E i g h t

Is Introspection (a First-Person Perspective)

Indispensable in Psychology?

-
Osborne Wi ggi ns

Th e S ub j e ct M att e r of Psychology

It might seem rather odd that before I can begin to address the issue of
introspection in psychology, I must first stipulate what I take the dis-
cipline of psychology to be about. That is to say, I must first indicate
what I take to be the subject matter of psychology. Before the ques-
tion of method can be asked, the question of subject matter must be
answered. This may seem rather odd, I say, because for most intellectual
disciplines we usually do not have to begin by ferreting through com-
peting proposals in order to define their topics of study. Psychology,
however, has spent most of its history in disagreements regarding its
specific topic. The proposals have ranged from the mind to overt be-
havior to the brain. Most recently interesting arguments have been de-
veloped claiming that it is all three and showing us how we might con-
ceive of these three as something of a unity. I would like to cut though all
these arguments by first taking my stand with those people who speak of
the “subjective brain” (Gillett) or “symbolic brain” (Coakley and Shele-
may) and then, while admitting fully that such a subjective brain is the

181
 182 Osborne Wiggins

reality, confessing a preference for beginning with the “subjective” or

“symbolic” side of that reality.
Another way of framing this issue of subject matter is Thomas
Nagel’s way (Nagel, 165– 80). Just as he asks, “What is it like to be a
bat?,” we might ask, “What is it like to be a human being?” And then
in examining this question we discover that there is indeed something
it is like to be a human being for the human being himself or herself.
There is something it is like to me for me to be me. This is what I shall
stipulate is an important part of the subject matter of psychology. And,
following Nagel, I shall switch from this awkward phraseology “what
it is like” to employing the traditional terms of mind and its cognates.
Still, this does not carry us far toward defining our subject matter. But
we cannot go far without utilizing a method for defining this subject
matter. Therefore I now turn my attention to the role of “introspec-
tion” in providing cognitive access to this subject matter.

Th e M i nd/ B o dy Pro b l e m and “ Introspection”

In my presentation I shall use terms other than introspection to desig-

nate the kind of self-awareness I shall propose. The word introspection
carries traditional connotations that I wish to avoid. The assumptions
on the basis of which it arose were those of a mind/body dualism; and
it was thought that there were dual methods for studying these dual
realities: introspection provides direct access to the “inner” mind, while
sense perception furnishes direct access to the “outer” world. On the
solid foundation of these two distinct modes of self-givenness, lawlike
statements and whole theories could be devised.
Many writers in the modern period exemplify these assumptions.
Here I shall cite only the case of William James in his Principles of Psy-
chology. Regarding mind/body dualism, James writes:

Nature in her unfathomable designs has mixed us of clay and flame,

of brain and mind, [such] that the two things hang indubitably to-
gether and determine each other’s being, but how or why, no mor-
tal may ever know. (182)
 Is Introspection Indispensable in Psychology? 183

And James’s characterization of psychological method seems to fol-

low almost automatically from his commitment to dualism:

Introspective Observation is what we have to rely on first and fore-

most and always. The word introspection need hardly be defined—
it means, of course, the looking into our own minds and report-
ing what we there discover. Everyone agrees that we there discover
states of consciousness. So far as I know, the existence of such states
has never been doubted by any critic, however skeptical in other re-
spects he may have been. (185; original emphasis)

The grip of these traditional presuppositions regarding both the

subject matter and the method for studying it was so strong that James
was committed to them despite the positivism which otherwise deter-
mined his approach. William Lyons in his book The Disappearance of
Introspection wonders why, given James’s positivism, this could have
been the case. He replies to his own question as follows:

Part of the answer lies in the fact that, besides being influenced by
the European positivists, he was also influenced by the heirs to
British empiricism, such as John Stuart Mill, and particularly the
Scottish philosophical psychologists, James Mill and Alexander
Bain. These nineteenth-century empiricists took for granted, as-
suming rather than arguing, that introspection was the method for
studying anything to do with the mind. (Lyons, 7)

I mention this history and quote James to illustrate it because from

it I want to retain the notion of modes of direct access to the subject
matter of psychology and, correlatively, a notion of the direct given-
ness of it. On the other hand, I set aside the traditional understanding
of a metaphysical dualism of inner mind and external world. Conse-
quently, my avoidance of the traditional term introspection. In its place
I shall employ terms like first-person perspective or, simply and more
often, self-awareness. However, my analysis will lead to a distinction
between two different forms of first-person perspective, two kinds of
self-awareness.
 184 Osborne Wiggins

S e l f- Aware ne s s and What It Is Aware Of

Let me explain further why the term introspection troubles me. I shall
claim that the subjective realm given to us in self-awareness is not just
a sphere of mental processes. It is a sphere of mental processes which
are experiences of something. In particular, they are experiences of my
body and experiences of the world. Readers who have heard some-
thing about phenomenology will probably recognize this as an adher-
ence to Edmund Husserl’s notion of intentionality (Gurwitsch 2009,
139– 56). The world that is the object is the world as experienced by the
subject. Similarly, the body is precisely the body as experienced by
the subject. And here body and mind are not experienced as separate:
the mind given through self-awareness is an embodied mind. Moreover,
it is an embodied mind situated or embedded in the world (Merleau-
Ponty; Rowlands, 51– 84). In insisting that this is the body as experi-
enced and the world as experienced, I am not propounding an idealism:
I am not denying that there is a world and body other than the world
and body as experienced by the subject. I am rather claiming that the
subject matter for the psychologist is first of all the world and body as
experienced by the subject under study.
Perhaps the need for such a claim can be recognized more readily
if I shift to that subdiscipline within psychology called “psychopa-
thology.” When the psychopathologist studies schizophrenia, for in-
stance, she will not make much progress comprehending the experi-
ences of someone who suffers from delusions and hallucinations unless
she seeks to understand the spatiotemporal and causal world as the
schizophrenic subject experiences it. I speak of the spatiotemporal and
causal world, but this is not the space, time, and causality experienced
by us so-called normal individuals. Space, time, and causality as ex-
perienced by the person with schizophrenia are different, strikingly
different. Nonetheless the psychopathologist must try to make as much
sense as she can of them in order to make some sense even of what the
words delusion and hallucination mean, and these are essential techni-
cal concepts for understanding schizophrenia. Similar remarks could
be offered about the patient’s experienced body (Wiggins and Schwartz,
269– 81).
 Is Introspection Indispensable in Psychology? 185

So let me sum up the point I have just sought to make. When I am

aware of myself in the two different forms of self-awareness I am about
to describe, I am aware not just of some sphere of mental events. I am
rather aware of mental processes that are themselves aware of some-
thing, namely, aware of their embodiment and aware of a world. As a
result, it makes no sense to contend any longer that self-awareness gives
us only an “inner” realm of “inner” mental occurrences. Hence the mis-
leading connotations of “introspection.”

Pre s ci e nt i fic S e l f-Awareness:

Th e Taci t E x p e ri e nce of S e l f in Everyday Life

Two different levels of self-awareness must be distinguished, however.

The first I shall call “prescientific” or “everyday” self-awareness. Per-
haps the second phrase, everyday self-awareness, is more fitting be-
cause by “prescientific” I mean merely that this mode of self-awareness
goes on continuously, throughout our wide-awake lives. It requires no
special effort, no turning of our focal attention toward ourselves, as is
required for the more careful and intellectually rigorous self-awareness
that I shall term scientific self-reflection. We human beings become ini-
tially acquainted with the basic phenomena of psychology through our
own prescientific experience of ourselves as we grow up and as we con-
tinually experience ourselves in our everyday lives. This prescientific
self-awareness involves a direct, although indistinct, givenness of our-
selves to ourselves. In this everyday experience we are also aware of the
experiences of other people, and these experiences of others also shape
our understanding of ourselves and of mental life in general. This aware-
ness of the experiences of other people is indirect, but it is integral to
forming our familiarity with the human mind. Hence in our prescien-
tific lives, we directly experience our own subjective lives and indirectly
experience the subjective lives of other persons; and these intertwined
and comingled experiences of the subjective teach us, in a manner
suitable for everyday communication and understanding, the work-
ings of human subjectivity as such. For example, you, as you read this
essay, are attending to what I am saying. But during the entire time you
 186 Osborne Wiggins

are also inattentively aware of your own experiencing life, however

indefinite this ongoing self-awareness may be (Zahavi; Gallagher and
Zahavi, 45– 68).
It is crucial to note that this sort of self-awareness is not an explicit
or focused awareness. It is rather implicit and tacit. It occurs “on its
own,” we might say; I do not have to actively bring it about. Of course,
at any given moment of our experience we are usually focused on or
attending to something or other. Something is the “explicit topic” of
our awareness. As the Gestalt psychologists have pointed out, how-
ever, this explicit theme appears to us within an implicit field. We al-
ways have some awareness of this field, however marginal this aware-
ness may be in comparison with our active absorption in the theme.
And whatever the changing themes of my consciousness may be, this
flowing self-awareness persists as a part of the field within which the
various themes appear. Not focusing on my experience, my attention is
directed elsewhere: I may be actively attending to the sentence in the
essay I am writing. Nonetheless, at a far more automatic or passive level
my experiencing life is indistinctly aware of its own ongoing experi-
encing (Gurwitsch 2010, 413– 537).
Prescientific self-awareness is essential for our learning the nature
of psychological events and thus for our comprehension of psycho-
logical concepts. Moreover, this prescientific learning and understand-
ing continue to inform the inquiries of the scientific psychologist.
However far the conceptualizations of the scientific psychologist ex-
tend beyond what is experienced at the prescientific level, these more
sophisticated concepts draw some of their meaning from it.
These last statements may appear dubious, however. So let me at-
tempt to shore them up. An experimental psychologist is concerned,
let us say, to pinpoint the neuronal processes in the brain that correlate
with certain visual experiences the research subject is having. Through-
out the investigation, the psychologist assumes she knows what it is
like to visually perceive something. We are back to Nagel’s question,
“What is it like to have an experience of a certain kind?” The psycholo-
gist can know what it is like to visually perceive something because she
herself has visually perceived many things in the past and this pervasive
experience was something that she herself, at a tacit and automatic level,
 Is Introspection Indispensable in Psychology? 187

was vaguely aware of. Hence her experiment, to herself and to her read-
ers, will make sense because they all are aware of visual perception
through their own implicit self-awareness of it.
Let me cite just one other example. In our daily lives, apart from
ever being schooled in the science of psychology, we possess an under-
standing of various emotional words. We use the terms, remorse, re-
gret, and guilt. Of course, in everyday speech distinctions among these
words may remain somewhat obscure. But we do assume that we know
generally what they mean, and we may on occasion find the word re-
morse appropriate when we would reject the term guilt as not truly
designating the feeling meant. Such words represent only a small sam-
pling of the broad vocabulary we employ in speaking of mental pro-
cesses, and they usually serve well in making ourselves understood by
others. My claim is that this mentalistic vocabulary of nontechnical
language derives much of its meaningfulness from the tacit but direct
awareness we have of our own experiences, as well as from the indirect
awareness we have of the experiential lives of others.

E x p l i ci t S e l f-Re flection:
A St e p i nto Psy chol ogical Science

The continuous occurrence of ordinary, implicit self-awareness makes

possible an explicit act of reflection on precisely the experiences oc-
curring at this first level. In other words, I can explicitly thematize the
experiences that previously occurred without being explicitly noticed.
As a matter of fact the descriptions I proposed earlier of what was hap-
pening at this implicit level would have been impossible if I had not at
that time been explicitly reflecting on this implicit level.
I want to stress, in addition, that the “self” that is disclosed in such
reflection is an embodied self, just as it is at the first, implicit level. And
moreover, this embodied self now brought into explicit focus as my
theme is seen by me as itself aware of the world.
In such explicit self-reflection it is possible for the self to be directly
given to one’s reflective attention. To assert such direct givenness of
something is not, however, to say that this givenness is uninterpreted.
 188 Osborne Wiggins

I agree that, on the one hand, there are no “uninterpreted data.” There
are nevertheless “data,” that is, items that are directly observable. A
datum can be both seen as something (interpreted) and at the same
time directly present itself as precisely this something. And it can as
well directly present itself as something other than what it was inter-
preted as being.
The main advantage of self-reflection for the science of psychology
is that it is the only mode of examination in psychology in which the
topic being examined is directly given to the inquirer. Hence it enjoys
the epistemological advantage involved in direct evidence: the subject
matter is directly presented to our direct inspection. We need not en-
gage in inferences regarding our topic of study: we have it itself directly
given to our observing reflection.
Now I need to substantiate as best as I can these claims regarding
the self-givenness of intentional processes to reflection. Assertions re-
garding the possibility of self-givenness, however, can be substantiated
in only one way: by inviting one’s critic to try to directly see for him-
self. Attempting to prove direct givenness by argument is pointless be-
cause the propositions in an argument will themselves ultimately have
to find their justification in things that are directly given. In brief, di-
rect givenness is the ultimate court of appeal for rational discourse.
Now in self-reflection, I may, of course, simply examine various
particular features of my experiences. I can, however, go beyond such
inspection of particulars and their particular features and begin to focus
on features of my own several experiences as examplars of a more gen-
eral kind of experience. Of course, a question of epistemological jus-
tification arises here: How can I generalize from features of my own
individual experiences to all human experiences of a specific kind? How
can I assure myself that my own experiences do exemplify features that
can be found in all experiences of the selected class? For after all a critic
might intervene at this juncture to point out that the lack of justifica-
tion for any such generalization has always constituted the Achilles’
heel of a psychological science utilizing “introspection.”
In an attempt to reply to such a critic we might begin by examin-
ing what theorists in fact do when they make general claims on the
basis of self-reflection. This examination will not, of course, provide
 Is Introspection Indispensable in Psychology? 189

an epistemological justification, but it may help us infer what these the-

orists assume will convince us, their readers, that their general claims
are cogent.

Th e Us e s o f S el f-Awareness in
Pre s e nt - Day Psy chol ogical Theory

Let us first look at an example developed by Mark Rowlands, in his

2010 book, The New Science of the Mind: From Extended Mind to Em-
bodied Phenomenology. Rowlands seeks to persuade his readers of the
thesis advanced only recently that the human mind is “extended” in a
manner not previously recognized (Clark and Chalmers; Rowlands).
His example is using his car’s GPS to direct him to a desired address.
He begins his example in the manner that philosophers often do: “Con-
sider, for example, a relatively new acquaintance of mine: my car’s
GPS (global positioning system)” (Rowlands, 13). Rowlands invites
us to “consider” his example. What is involved in “considering” it? I
suggest that as Rowlands develops his example we are supposed to
imagine ourselves doing what he is describing. He is not saying, “Re-
ally go out and really get in your car, really start it up, and really pro-
gram your GPS and really follow it as it directs you to the destination.”
In other words, we are not required to actually follow the instructions
of our car’s real GPS. We are rather supposed to imagine ourselves
programming the GPS, listening to its instructions, and following these
instructions as we drive our car. Having imagined ourselves having such
a set of experiences, we are supposed to reflect on them and determine
whether we can detect in those experiences the features to which Row-
lands then directs us. We may or may not detect those features, and
this seems to serve for Rowlands as a “test” for the claims he seeks to
advance.
In order to indicate that this argumentative practice is not rare, I
shall cite the original 1998 article by Andy Clark and David J. Chalmers
that introduced their “extended mind” thesis. After a first paragraph
which starts by posing the question, “Where does the mind stop and
the rest of the world begin?” (Clark and Chalmers, 10), they outline
 190 Osborne Wiggins

three examples of the sorts of experiences they will analyze in order to

develop their novel position. These examples are introduced this way:
“Consider three cases of human problem solving: . . .” (10 –11). The
reader is supposed to imagine a “person” having the experiences they
describe, although, of course, none of the readers is actually having
those experiences at that time. The last of their examples is indisputably
imaginary since it begins: “Sometime in the cyberpunk future, a per-
son sits in front of a similar computer screen” (10–11).
Now there are (at least) two attitudes in which we can read and
make sense of their examples. We can read the examples as if they were
characterizing the activities of someone else. In other words, we could
understand the “person” they describe from a third-person point of
view. We would still be able to grasp that other person’s experiences
but only indirectly, “from the outside,” as it were. This sort of indirect
awareness fails to present us with the presence of the experience in its
definite and graphic detail. Only the direct givenness of the experience
would do that. Consequently, if we really wish to examine the precise
features of this experience, we must, each of us, imagine ourselves hav-
ing the experiences of the “person” they portray. I must put myself in
this person’s place and exactly imagine the experiences they depict
“him” as having. I must turn “his” experiences into my own by pre-
cisely creating in myself the experiences Clark and Chalmers describe
him as having. I need not actually sit in front of my computer and ro-
tate the figures as they say: I need rather only graphically imagine my-
self sitting in front of a computer rotating the figures in that way. But
such imaginings must produce the experiences precisely as Clark and
Chalmers state.
Once I do this, I can take the next step: I can directly reflect on these
imagined experiences of mine and see if they do in fact exhibit the prop-
erties Clark and Chalmers attribute to them. This is a first-person per-
spective. Such reenacting of the experiences depicted in the authors’
examples is required if I wish even to understand the claim they are mak-
ing. The imagined experience I reflect upon gives their claim meaning
by exemplifying it and thereby adding concreteness to their proposi-
tions. And such self-reflection is necessary, moreover, if I wish to de-
termine for myself the truth of their statements. In this self-reflection I
may find that the experience exhibits features inconsistent with the
 Is Introspection Indispensable in Psychology? 191

ones they assert. In this case I have some evidence—at least some pro-
visional evidence — for disagreeing with them. And it just so happens
that, when I do reflect carefully on my own imagined experiences, I
find that I disagree with Clark and Chalmers. I maintain, against
Clark and Chalmers, that these properties should be characterized in a
different manner in order to describe them correctly.
But the determination of truth has not proceeded far: we have at
this point only two sets of people disagreeing. Hence any claims to
truth or falsity must remain modest. Its main right to be deemed a test
of truth at all arises from its appeal to others to do the same and then
to critically examine their findings. In other words, it appeals to inter-
subjective confirmation or disconfirmation. Obviously, empirical stud-
ies in cognitive science or neuroscience must be brought to bear on
this theory of extended mind if its evidential support is to gain strength.
Indeed evidence of any kind that has been rigorously developed and
bears relevance to the hypothesis should be sought. And in fact Clark
and Chalmers and Rowlands do draw on research studies in psychol-
ogy to buttress their assertions (Clark and Chalmers; Rowlands). But
these findings could help to decide our disagreement only if we cor-
rectly interpret the psychological experiences the subjects had. And
here again it would seem to require that we imagine ourselves in the
same experimental situation as the subjects and ask ourselves what they
must have been experiencing. In slightly different words, we are seek-
ing to imagine ourselves having experiences similar in all relevant re-
spects to what the subjects experienced so that we can examine reflec-
tively and carefully the properties of those experiences.
I mentioned above one advantage that explicit self-reflection brings
to psychology when I stressed the epistemological value of the direct
givenness of psychological processes to reflection. Having now por-
trayed the role that imagination plays in providing this reflection with
its direct object, I would like to indicate the second advantage provided
by this procedure. By imagining myself engaged in an experience of a
certain sort, I bring into being the experience I shall examine. To the
extent that I bring into being an experience of exactly that sort, I can
then have directly given to my reflective perceiving the exact and
definite features that I wish to scrutinize. I have a detailed and graphic
exemplar of my object of study. This precise and definite givenness of
 192 Osborne Wiggins

the features of the experience furnishes a reliable basis for reflective

examination. Without such a graphic imagining of the experience, my
reflection on it, no matter how careful, would find in it only indefinite
and inexact features.
Now if we are still considering something analogous to the tradi-
tional concept of introspection, it is indeed an untraditional kind. For
this sort of self-reflection is reflection on a self that generates new ex-
periences through imagining itself having them. Imagination plays a
far more central role here than it did in traditional psychological meth-
ods. The psychologist must first graphically imagine himself or herself
having certain kinds of experiences and then reflect on them and seek
features they might share with others.

Th e N e e d f o r Ex p e r tise

The ability to employ this method in a reliable manner requires that

the psychologist cultivate, through repeated practice, an expertise in
doing so. Almost all sciences require an expertise that draws on but
surpasses “book learning.” Such expertise develops gradually as scien-
tists immerse themselves in the daily practice of their science. Being
able to graphically imagine oneself performing certain mental acts is
not the ability of a beginner, a novice. It requires someone who has lived
through all the frustrations, failures, and accomplishments of a seasoned
practitioner (Collins).
And the work of acquiring expertise is even more crucial for the
success of self-reflection. For purposes of simplicity I have spoken
thus far about the direct givenness of one’s own experiences through
direct reflection on them. But such direct givennness in direct reflection
can be achieved only as the fruit of much practice, effort, and study. I
cannot directly observe my own experience as easily as I can directly
perceive the color of the wall. It is more like the expert radiologist’s
direct perception of the two small cancerous masses on the X-rays of a
patient’s lungs (Polanyi, 49– 65). Direct self-reflection is a hard-earned
achievement. It is an ability that requires and reveals the attainment of
expertise.
 Is Introspection Indispensable in Psychology? 193

Co ncl usion

I realize that the novelty of some of my proposals is likely to leave

them seeming dubious. In the final resort any appeal I may make for
their cogency must take the form of an appeal to the reader to try it for
himself or herself. Especially the claim regarding the productive func-
tion that imagination can provide may seem to deviate much too far
from a reliance on reality as the source of all evidence.
For a brief defense of my position I would like to suggest that all,
or almost all, of the social sciences employ the methods of imagination
far more than we have yet recognized. History, anthropology, soci-
ology, hermeneutics, and so on, all have ample need to imagine the
experiences of their subjects. Of course, any such imaginings must ul-
timately be tested through confrontation with the evidence given in
reality. But this very evidence is a manifestation, an externalization (we
might say), of the experiences of peoples, experiences the social scien-
tist aims to know.

R e f e re nce s

Clark, Andy, and David J. Chalmers. “The Extended Mind.” Analysis 58

(1998): 10– 23.
Coakley, Sarah, and Kay Kaufman Shelemay, eds. Pain and Its Transforma-
tions. Cambridge, MA: Harvard University Press, 2007.
Collins, Harry, and Robert Evans. Rethinking Expertise. Chicago: University
of Chicago Press, 2007.
Gallagher, Shaun, and Dan Zahavi. The Phenomenological Mind: An Introduc-
tion to Philosophy of Mind and Cognitive Science. London: Routledge,
2008.
Gillett, Grant. Subjectivity and Being Somebody: Human Identity and Neu-
roethics. Charlottesville, VA: Imprint Academic, 2008.
Gurwitsch, Aron. The Collected Works of Aron Gurwitsch (1901–1973). Vol. II:
Studies in Phenomenology and Psychology. Ed. F. Kersten. Dordrecht:
Springer, 2009.
———. The Collected Works of Aron Gurwitsch (1901–1973). Vol. III: The Field
of Consciousness: Theme, Thematic Field, and Margin. Ed. Richard M.
Zaner. Dordrecht: Springer, 2010.
 194 Osborne Wiggins

James, William. The Principles of Psychology. Cambridge, MA: Harvard Uni-

versity Press, 1983.
Lyons, William. The Disappearance of Introspection. Cambridge, MA: MIT
Press, 1986.
Merleau-Ponty, Maurice. Phenomenology of Perception. London: Routledge,
2000.
Nagel, Thomas. Mortal Questions. Cambridge: Cambridge University Press,
1979.
Polanyi, Michael. Personal Knowledge: Towards a Post-Critical Philosophy.
New York: Harper Torchbooks, 1964.
Rowlands, Mark. The New Science of the Mind: From Extended Mind to Em-
bodied Phenomenology. Cambridge, MA: MIT Press, 2010.
Wiggins, Osborne P., and Michael Alan Schwartz. “‘The Delirious Illusion of
Being in the World’: Toward a Phenomenology of Schizophrenia.” In
Founding Psychoanalysis Phenomenologically: Phenomenological Theory
of Subjectivity and the Psychoanalytic Experience, ed. D. Lohmar and
J. Brudinska, 269– 81. Dordrecht: Springer, 2012.
Zahavi, Dan. Self-Awareness and Alterity: A Phenomenological Investigation.
Evanston, IL: Northwestern University Press, 1999.
 C h a p t e r N i n e

Could Normative Insights Be Sources

of Normative Knowledge?

-
Allan Gi b bard

The nineteenth-century English philosopher Henry Sidgwick, Derek

Parfit (2011, xxxiii) tells us, wrote “the best book on ethics ever writ-
ten.” This was Sidgwick’s treatise The Methods of Ethics, whose first
edition was 1874. I think I agree with Parfit’s assessment. Sidgwick, like
Jeremy Bentham (1789) and John Stuart Mill (1863), was a utilitarian:
Sidgwick’s version of utilitarianism holds that an act is right if it pro-
duces the greatest net happiness, the greatest balance of enjoyment over
suffering. Opponents of utilitarianism have often thought that moral
requirements like the duty to speak truthfully and to keep one’s prom-
ises have a strength that goes beyond what they draw from happy con-
sequences. Utilitarians like Mill prided themselves on finding an em-
pirical basis for morality, whereas their opponents—William Whewell
(1845) prominent among them— claimed a basis for morality in intu-
ition. Sidgwick, as I say, was a utilitarian, but he argued that utilitari-
anism too needs intuition.
Why would he think this? Sidgwick, and his student G. E. Moore
after him, argued that moral claims differ in nature from scientific

195
 196 Allan Gibbard

claims. Moral claims aren’t naturalistic and empirical. Findings of what

we approve of contrast with this: they are psychological, and experi-
mental psychologists can study them empirically, testing their hypothe-
ses by observing people’s responses and the like. The claim that some-
thing is morally wrong, though, isn’t that people disapprove. Neither
is it a more complex psychological claim. Against this, someone might
contend that when we call something “wrong,” we mean that we would
disapprove if we rehearsed, vividly and repeatedly, everything involved.
This, if it were correct, would make questions of right and wrong a topic
for psychologists to study: the question of what we would approve of
under those conditions is a psychological question that can be investi-
gated by designing and running experiments. But Sidgwick denied that
moral questions are empirical, scientific questions, and Moore (1903,
chap. 1) constructed elaborate arguments against all such “naturalistic
definitions” of moral terms. Moral questions, they argued, can’t be an-
swered by empirical methods of experiment and observation alone.
Moore argued in addition that no metaphysical definition will cap-
ture what we mean by “good.” Plato long ago had Socrates arguing to
Euthyphro that pious can’t mean “loved by the gods.” The gods love
pious things because they are pious, Euthyphro asserts—and this can’t
mean that they love them because they love them. Such arguments may
be controversial among theologians, but it seems clear that good can’t
mean “approved by an all-powerful being,” since it’s not a sheer mat-
ter of meaning that an all-powerful being will approve good things.
Moore concluded that good is, as he put it, “a simple, non-natural
property.” It is simple in that it can’t be analyzed into component con-
cepts. Quite what he meant by “non-natural” might be hard to char-
acterize, but according to Moore, good isn’t a physical property, and it
isn’t a psychological property such as that of being approved. An al-
ternative to Moore’s view would make good a definable property, but
still non-natural. The term good means “desirable,” we can say, and de-
sirable means something like “fittingly desired.” As I myself would
put it, that something is good means that desiring it is warranted.
When we say that stealing is wrong, we can try saying, we mean that
feelings of disapproval toward stealing are warranted. That leaves an-
other nonempirical element, warrant for such things as desires or feel-
 Normative Insights Sources of Normative Knowledge? 197

ings of disapproval. People desire fame even to continue after they are
dead, for its own sake and not just as a means to something else—but,
we can ask, is such a desire warranted? One view of moral questions is
that they boil down to questions of which acts and attitudes are war-
ranted. Does a lie to spare a person embarrassment warrant feelings of
approval or of disapproval? On this analysis, there indeed is a simple
non-natural property that characterizes ethics, but it is not the good-
ness but the property of being warranted. This qualifies as a form of
non-naturalism but a form that is different from Moore’s.1
In recent decades, various moral philosophers have come to see
ethics as part of a broader subject matter involving warrant or some-
thing of the sort. The broader category is labeled “normative,” and we
can perhaps define normative issues as questions of what warrants what.
Normative questions will include not only ethical questions, but such
questions as what beliefs are warranted in light of one’s evidence. In
science, we can ask what degrees of credence are warranted for various
hypotheses, such as the hypothesis that depression is a genetic adapta-
tion. The philosopher Wilfrid Sellars characterized normative claims
as “fraught with ought,” and Kevin Mulligan calls them “oughty.” I
have been suggesting that we explain normative concepts as concepts
involving a primitive concept of being warranted. Other philosophers,
including Derek Parfit (2011) and T. M. Scanlon (1998), propose that
the basic normative concept is that of a reason to do something or to
believe something or the like. When we ask, for instance, whether a
correlation between taking a drug and death is causal, we face the ques-
tion of whether the correlation is reason to believe that the drug tends
to kill. Normative ethics, then, the study of what makes acts morally
right or wrong, counts as a normative subject matter. So do such other
subfields of philosophy as normative epistemology. Epistemology, as a
branch of philosophy, is the theory of knowledge, and normative epis-
temology is that major part of epistemology which addresses norma-
tive questions concerning knowledge and belief, such as what degrees
of credence in a claim are warranted by what sorts of evidence.
According to non-naturalists, these normative fields of inquiry
deal with properties that are non-natural. The probability of a claim,
for example, we can define, in one sense, as the degree of credence in it
 198 Allan Gibbard

that is warranted by the evidence. This concept of probability is nor-

mative in that it concerns warrant, and so according to normative non-
naturalists, the probability of a claim, in light of a given body of evi-
dence, is a non-natural property of the claim. (This is one sense among
many that the term probability can be given, but I’ll stick to this sense
in my use of the term in my examples.)
All this brings us to a fundamental and baffling problem: if nor-
mative properties aren’t ordinary empirical properties which we can
investigate via sense experience, how could we know about them? This
has long seemed the deepest problem in ethical theory. Ethical properties
aren’t natural properties, arguments like those of G. E. Moore seem to
show. Still, we take ourselves to know things about them. It would
be wrong to extort money from a couple by torturing their child; that
seems as clear as anything could be. But to say this, according to non-
naturalists, is to ascribe a non-natural property to such an act. How
could we know that the act has this property, if it isn’t one that we can
learn about through empirical, scientific investigation? This, I’m now
saying, is a kind of problem that extends far beyond ethics. Normative
epistemology is another normative subject matter. If normative non-
naturalists are right, the theory of justified credence studies properties
that are non-natural: probabilities in the normative sense that I just de-
fined. If you flip a coin twenty times and it comes up heads every time,
you have reason to believe strongly that this is no fluke and the coin is
fixed. Perhaps it has two heads, or perhaps it is weighted to land heads
always or almost always, but surely, we can conclude, this is not a fair
coin fairly flipped which just happened to come out that way. In light
of how the coin has landed, its being fair has negligible probability.
The warranted degree of credence that it is fair is close to zero. The
coin thus has a non-natural property, normative non-naturalists will
say, a property that statisticians can investigate. But again, we must ask,
how can we learn about non-natural properties? Even if we should be
skeptical of seeming ethical properties and reject them as mysteriously
non-natural, can we be skeptical that there’s any such normative prop-
erty as that of being justified as a scientific conclusion? Or more pre-
cisely, the question is normative: should we be skeptical in this way?
And if we should, is this “should” itself something non-natural, and
so something we should banish from our thinking?
 Normative Insights Sources of Normative Knowledge? 199

I have presented this problem of normative knowledge in a way

that skips past a host of philosophical controversies. To many philoso-
phers, the claims I have been presenting about non-natural properties
that we can know about will seem implausible and even absurd. Many
outside academic philosophy may respond the same way. I’ll be ask-
ing later about alternative ways of regarding normative questions—
alternatives to non-naturalism—and I’ll be searching myself for an alter-
native. I want first, though, to spend some time with the problem of how
we can have normative knowledge as it arises within non-naturalist
thinking.

Th e N e e d f or Int uitions

Sidgwick and Moore, as I say, agreed that ethical knowledge can only
rest on intuitions. It’s as clear as clear could be that it’s wrong to tor-
ture a child to extort money to pay for a vacation, but we couldn’t know
this without intuition. We can investigate what it’s like for the child
and for the parents and how the workings of society must rest on a so-
cial trust that requires security from such things happening. Still, no
normative conclusion follows from such non-normative facts alone.
How do we establish that if all these non-normative facts obtain, then
such an act is heinous? There is strong moral reason not to impose suf-
fering on people — surely this is right. How, though, could we know
such a fact if it is non-natural? And how can we investigate such ques-
tions if the methods of the empirical sciences aren’t up to the job? The
source of our knowledge of fundamental normative facts, Sidgwick
concluded, can only be an ethical intuition.
What, then, is an intuition? The term is used in ordinary language,
and it is used both by some philosophers and by some psychologists.
In this essay I consider the term as used in the non-naturalist tradition
of Sidgwick and his successors and as it is used by some social psy-
chologists. My prime example in social psychology is the book The
Righteous Mind by the social psychologist Jonathan Haidt (2012).
When Sidgwick spoke of “intuition,” he meant a self-evident judg-
ment that, by its nature, delivers knowledge—and so delivers a truth.
Sidgwick worried about how to distinguish genuine intuitions from
 200 Allan Gibbard

seeming intuitions or states we might confuse with them. A state can

seem to deliver knowledge but fail to do so, and Sidgwick’s question
was how we can tell when this is the case. Psychologists too sometimes
speak of “intuitions” but in an importantly different sense. A mental
state is classified as an “intuition” by its psychological nature, regard-
less of whether it yields knowledge, and regardless even of whether it is
true. Intuition in the psychologist’s sense has come to be seen as one of
the two basic types of thinking; the other is reasoning. Intuition is
rapid and effortless, whereas reasoning is slow and effortful.2 Haidt, in
The Righteous Mind, gives as instances of intuition the “rapid, effort-
less moral judgments that we make every day” as, for instance, we read
the newspaper (45). These are flashes of affect directed at things one
is thinking about — newsmakers and their deeds, for example. These
flashes of directed affect constitute moral intuitions in the psycholo-
gists’ sense whether or not they correctly fit the rights and wrongs of
matters.
It is no part of the psychologist’s job as psychologist to ask whether
a flash of affect conveys a truth, or even to make sense of the question
of whether it does. Still, as human beings apart from anyone’s role as a
psychologist, we can ask this question: Can such flashes of moral affect
tell us right from wrong? Are they sometimes intuitions in Sidgwick’s
sense, sources of fundamental moral knowledge? And what would it
mean, if anything, to say that they can be? If we learn that a professor
has stolen a book from the library, our immediate response is a flash of
feeling which might lead us to exclaim “That’s wrong!” He has be-
trayed his calling, undermining an institution established to serve our
common goals. How, though, can we assess whether such an act is re-
ally wrong—whether the flash of affect gets matters right?
How we can learn moral truths and indeed whether or not we can
are of course highly contentious questions among philosophers. I have
been focusing on non-naturalism, which says that being wrong is a non-
natural property, and we can know about such properties in ways that
depend in part on moral intuition. A great deal of ethical theory, though,
has been devoted to finding alternatives. The part of ethical theory that
studies the nature of ethical claims and how they can be justified is
known as metaethics. If ethics is a subdivision of normative studies
 Normative Insights Sources of Normative Knowledge? 201

more generally, then some problems in metaethics will turn out to be

special instances of more general problems in what we could call meta-
normative theory. This will include such questions as these: Are there
normative facts? If so, are normative facts non-natural, or how else are
they to be conceived? If there indeed are such things as non-natural,
normative facts, can we know them or be justified in believing them to
obtain? If so, how?
These questions concern what’s normative fundamentally. Often,
to be sure, ethical difficulties arise from the difficulties of knowing the
non-normative facts of a situation. What economic system is ethically
best, for example, depends on how the various alternative systems
that one might support tend to work. Still, knowing the natural facts,
the facts that can be studied by scientific method, won’t by itself give the
answers to ethical questions. That a system produces misery tells against
it ethically, we can all agree, but a finding that it does goes beyond the
natural facts of what misery is like. How do we know even such obvi-
ous things as that misery is bad? It’s not just that misery is bad because
of something further about it—that it makes people ill, say, or detracts
from economic production. Those aren’t the only things that make
misery bad; misery is bad in itself. The claim that misery is bad in it-
self is fundamentally ethical, and so it is fundamentally normative.
The puzzle of normative knowledge is how we can have fundamental
normative knowledge, such as the knowledge that misery is bad in it-
self. This puzzle applies as well to questions that are normative but
not fundamentally. The lesson often drawn from the experience of the
English Jamestown colony is that private property is a good thing. The
initial, communal regime of the colony produced hunger and misery,
and so was a bad system. If it’s true, though, that without private
property there is greater misery than with it, that doesn’t by itself es-
tablish that private property is a good thing. We still need the norma-
tive finding that misery is bad. Nonfundamental knowledge must rest
in part on normative knowledge that is fundamental—and so again we
are led to the question of how we can have fundamental normative
knowledge.
That misery is bad in itself is obvious and little disputed, but other
fundamental normative questions are contentious. Are happiness and
 202 Allan Gibbard

misery all that matters in itself? Or are such things as deceit and coer-
cion bad in themselves, even apart from whatever suffering they en-
gender? Here again we face the puzzle of normative knowledge: Can
we know answers to the fundamental normative questions that are
disputed—and if so, how?

N o n- Nat ural i s m and It s Alternatives

Whether and how we can know the answers to questions will depend
on the natures of the questions. Normative non-naturalists like Sidg-
wick and Moore say that ethical questions concern fundamental nor-
mative properties that are non-natural, and our knowledge rests ulti-
mately on fundamental normative intuitions. Many philosophers, as I
say, reject this non-naturalist, intuitionist package. Indeed for a long
time, this package was regarded as philosophically naive and not to be
taken seriously. Among leading ethical theorists, it has undergone a
revival only in the past couple of decades (with such writers as Russ
Shafer-Landau, T. M. Scanlon, David Enoch, Ronald Dworkin, and
Derek Parfit).3 Still, even in the wake of this revival, many philosophers
still find intuitionistic non-naturalism unacceptable. How, though, can
we avoid it? A recent extensive scrutiny of alternatives comes in Derek
Parfit’s giant two-volume work, On What Matters (2011). Parfit argues
that every position but his own non-naturalism has defects that are
fatal, and non-naturalism is what’s left standing. I won’t attempt an ex-
haustive survey of the alternative positions that have been proposed,
but I will say a few words about why I find some of the most promi-
nent ones unsatisfactory. I’ll then turn to the best answer I can give to
the puzzle of normative knowledge.
Critics of non-naturalism maintain that we have no reason to think
that there are any such things as non-natural properties, and that even
if there were, we would have no way of knowing about them. Why
would normative intuitions yield knowledge? A frequent response looks
to mathematics. Mathematical knowledge too isn’t empirical. Obser-
vation tells us how things are but not how they must be of necessity.
My own view is that each of these kinds of knowledge must be ex-
 Normative Insights Sources of Normative Knowledge? 203

plained separately, and an explanation of how we can have kinds of

mathematical knowledge won’t carry over to ethical knowledge. Mathe-
matical knowledge is of structures that things in the natural world ap-
proximate. Genetic evolution could account for our potential abilities
to have such knowledge. It shaped our species to be fairly reliable about
some aspects of the natural world—the shapes and numbers of things,
for instance — and the easiest way to be reliable about these things is
to get matters of abstract structure right and suppose that the objects
of our experience have such structures at least roughly. A story like
this doesn’t transfer over to fundamental questions of ethics. Take, for
example, the question, Do wrongdoers deserve fundamentally to suf-
fer? Parfit claims that at base, no one deserves to suffer, whereas ret-
ributivists like Kant think us morally required to impose suffering on
wrongdoers, not because of the further evil it will avert, but for the pure
purpose of apportioning recompense to desert. How would anything
like abilities to think about structures enable us to discern the right
answer in a dispute like this?
Many philosophers today appeal to “reflective equilibrium.” This
was John Rawls’s term for a state where one has eliminated sources of
bias or bad judgment, like partiality or fatigue, and has considered
carefully all the arguments that could be brought to bear (Rawls 1971,
46– 63). Reflective equilibrium is what we try to work toward in any
philosophical inquiry, said Rawls, and this goes for ethics as much
as for any other area of philosophy. Now I agree that a state of perfect
ethical discernment would be a state of reflective equilibrium. That
leaves us, though, with the question of why a state of reflective equi-
librium would be one in which one got the layout of non-natural prop-
erties right.
An alternative to non-naturalism is moral skepticism. We can try
saying that there is no such thing as moral knowledge. This is the “error
theory” of the philosophers John Mackie (1977), Richard Joyce (2001),
and others. Moral judgments are systematically mistaken, because there
are no such properties as moral rightness or wrongness. Once we ex-
tend our puzzles to normative knowledge in general, though, the price
of such blanket skepticism becomes very high. Do we never have rea-
son to do anything or to believe anything? Or does talk of reasons to
 204 Allan Gibbard

do things or to believe things make no real sense? Can we not assess

whether one has reason not to hit one’s thumb with a hammer or reason
to believe that we are descended from apes? It seems that we have evi-
dence for many of our beliefs, and by “evidence” for a conclusion, don’t
we mean observations that give us reason to believe that conclusion—
not completely, perhaps, but at least with more confidence that we
would have reason to have without the observation? Whether or not
having reason to believe a thing is a non-natural property, the claim
that, in light of our evidence, we have reason to believe the theory of
natural selection is a normative claim, and it is hard to know how to
proceed with any inquiry if such claims make no sense. And if norma-
tive claims like this make sense, why not ethical claims? I have reason
not to stick my hand in a flame; if claims like this make sense, why not
the claim that I have moral reason not to force someone else’s hand
into a flame? All these retorts are quick and need further exploration,
but at the very least, a moral skeptic faces a challenge: if we deny all
sense to moral claims, what are we to say about other sorts of norma-
tive claims, claims we seem to need in order to proceed with our lives?
Must we deny that any normative claims whatsoever make sense?
Proponents of naturalism in ethics have far more sophisticated
theories to offer than any I have mentioned. One prominent genre is
“ideal response” theories, which I alluded to earlier. To say that an act
is wrong is to say that feelings of moral disapproval toward it are war-
ranted; our question is then what warranted means. It means, ideal re-
sponse theorists say, such as would be had in an ideal state of mind.
That disapproval is warranted means that in an ideal state of mind, one
would disapprove. (Here and from now on, I’ll use the term approve
for having feelings of approval—and correspondingly with disapprove.
Sometimes by the term disapprove we mean to believe to be wrong,
but I won’t use the term in that sense, since it would be circular to try
to define wrong to mean “such as would be believed wrong in an ideal
state of mind.”) That said, we now have to say what it is for a state of
mind to be “ideal.” That a state of mind is ideal for judgment, some
ideal response theorists say, means that the thinker has gotten the facts
of the situation correctly, considered matters vividly and repeatedly,
and is impartial— and some other conditions are added. Different ac-
counts fill out this list differently. That makes for a problem: that con-
 Normative Insights Sources of Normative Knowledge? 205

ditions on a state of mind are “ideal” for judgment seems itself to be a

normative claim, and we could dispute what conditions are ideal. What,
we can ask, is at issue in such a dispute? The same sorts of arguments
that G. E. Moore aimed at naturalistic definitions of good can be di-
rected at naturalistic definitions of ideal. Suppose, for example, that
responses are affected by upbringing, so that sparing the rod or not in
childhood makes a difference to later, adult responses. Then people who
agree on whether a person satisfies a given list of criteria that includes
having been spared the rod may disagree on whether that makes the per-
son’s responses ideal. If so, we have to ask what is at issue among them.
Some philosophers reject addressing this question and trying to
answer it. Instead, they propose, let’s change the question. Let’s change
the question to one we know how to answer, redefining it as an empiri-
cal, scientific question, amenable to the methods of science. Sidgwick
and Moore rejected trying to make normative questions empirical; it is
changing the subject, they argued. Proponents of “reform definitions”
of moral terms, though, advocate precisely that: changing the subject.
We should regard our original questions, they say, as confused and ill
formed and find a substitute that speaks to our needs (see Brandt 1979).
We should bear this alternative in mind as we proceed with this
inquiry, asking whether some naturalistic reform of our language could
serve the intelligible purposes of moral language. Before we settle on
reform, though, we should look further for what we might already be
asking with our moral questions.

M o ral E pist e mology

Whether stealing a library book is wrong, I suggested, amounts to the

question of whether condemnatory moral affect toward such an act is
warranted. How can we address such questions? How can we evaluate
whether a way of feeling toward an act is warranted? In his 1959 book,
Ethical Theory, Richard Brandt formulated what he called the “quali-
fied attitude method” (250) as our standard procedure for answer-
ing ethical questions. I’ll sketch this method as an example of a moral
epistemology, a theory of moral knowledge. We have no way to answer
moral questions, Brandt took it, that is independent of moral attitudes
 206 Allan Gibbard

such as feelings of obligation or feelings of moral condemnation. Par-

ticular moral questions we decide by exercising our affective capacities.
We also decide them by applying moral principles, but moral principles
too need testing, and we test moral principles too by our moral atti-
tudes. If a principle clashes with our attitudes, we may revise it or re-
ject it. We don’t, though, regard our feelings as infallible. Not every-
thing they seem to tell us is true. Brandt, then, offered some tests for
attitudes. They must be discounted if they fail to be impartial or in-
formed. They must be discounted if they stem from an abnormal state
of mind. They must be discounted if they can’t be made to accord
with some tractable set of general principles. Attitudes that pass such
tests we might call qualified. We judge principles by bringing to bear
our qualified attitudes. And when principles conflict — for example,
that one shouldn’t lie and shouldn’t cause pain, where only a lie will
avoid pain—we resolve the conflict by bringing to bear our qualified
attitudes. An act is morally required, we can try saying, if and only if
feeling obligated to perform it would pass all the tests of the qualified
attitude method, and thus constitute a qualified attitude.
Brandt meant this not as a definition of what the term morally re-
quired means but as an account of what it is for an act to be morally re-
quired. His endorsement of the qualified attitude method, then, doesn’t
violate G. E. Moore’s strictures against naturalistic definitions, since
the purported definitions Moore inveighed against were ones meant to
say what moral terms mean. Brandt considers a claim something like
the following:

An act is morally required if and only if the attitude of feeling ob-

ligated to perform it passes the following tests (and then the tests
of the qualified attitude method are listed).

If this is not intended as a definition that gives the meaning of the term
morally required, what does it mean? We can also attribute to Brandt
this claim about warrant:

An attitude is warranted if and only if it passes the tests of the

qualified attitude method.
 Normative Insights Sources of Normative Knowledge? 207

Again, Brandt would not put this forth as a claim about what the term
warranted means. What the claim does mean, then, will depend on what
warranted means.
I’ll now sketch an account of what warranted means, an account
that could give sense to these claims. (It isn’t Brandt’s account; rather,
it is an account that I have advocated.) Then I’ll use this account to ask
whether normative intuitions could ground normative knowledge.

E x p re ssiv ism

A kind of position now called “expressivism” was first broached in the

1930s and has been developed substantially since then. Non-naturalists
are right, expressivists say, that no naturalistic definition will capture
what a normative term means. We cannot give a correct naturalistic
definition of a basic normative term like good or warranted or ideal.
That’s not, though, expressivists say, because normative terms stand
for non-natural properties. Rather, their meaning is to be explained
by characterizing the states of mind that normative terms are used to
express. According to A. J. Ayer’s emotivism in his provocative 1936
book, Language, Truth and Logic, to say “Stealing is wrong” is to ex-
press disapproval of stealing. (Again, here and subsequently, I use dis-
approval to mean feelings of disapproval, not the belief that the thing
in question is wrong.) In saying this, we give the meaning of the term
wrong.4 Ayer’s theory of moral meanings is powerful but unsatisfac-
tory, in my view: I can feel disapproval toward a thing—a sexual prac-
tice, say—and yet think it isn’t really wrong. What is it to believe that
the practice isn’t really wrong? It is to believe that disapproving of it
isn’t really warranted—which brings us back to what the term war-
ranted means.
I propose explaining the meaning of warranted, in this sense, in an
expressivistic way. What does it mean to say that a state of mind such
as approval is “warranted”? We explain it, I proposed, by explaining
the kind of state of mind that calling a state of mind “warranted” ex-
presses. In my 1990 book, Wise Choices, Apt Feelings, I called this state
of mind “accepting a system of norms,” and I tried to explain this state
 208 Allan Gibbard

of mind by its role in a psychological theory. Accepting a system of

norms that says to feel disapproval is what stands behind a syndrome
that includes tending to be in the state of mind and tending to avow
one’s acceptance in discussion. In my later book Thinking How to Live
(2003), I described this state of mind as akin to planning how to feel.
When I believe disapproval of stealing to be warranted, I “plan,” in a
certain sense, for the feelings to have toward stealing. This “planning”
here isn’t planning for action, since one can’t feel disapproval at will,
and even if one could act to get oneself to disapprove of a practice,
planning to get oneself to disapprove wouldn’t constitute believing
the practice to be wrong. Planning a state of mind isn’t in general plan-
ning to will being in it. Even with planning for action, something like
this holds. The direct effect of my having the plan is on the will, true
enough, but not because I will to will. My having the plan tends to
bring it about directly that I will accordingly. Similarly, when I believe
that disapproval is warranted, the direct effect isn’t that I will to disap-
prove but that I disapprove.
Can normative insights, then, give us normative knowledge? Nor-
mative insights, I am taking it, consist of flashes of moral affect which
prompt us to moral conclusions like “That’s wrong!” Do these conclu-
sions constitute knowledge? They do, we can say, if they deliver truths
in a way that is nondefective—but what does it mean for a way of com-
ing to beliefs to be “defective” or not? For a belief to constitute knowl-
edge, we can try saying, it must be true, it must be warranted, and the
processes that eventuate in true belief must be reliable. A chief ques-
tion, then, is whether, in one’s epistemic circumstances, the belief is war-
ranted. I have tried to explain the belief that a state of mind is warranted,
in a thinker’s epistemic circumstances, as something like a plan to be in
that state of mind if in those circumstances. The question of whether a
belief is warranted in the circumstances thus amounts to the planning
question of whether to have that belief in those circumstances. One way
I might plan, for example, is to plan to feel toward things in ways that
meet the standards of Brandt’s qualified attitude method. To plan this
way, I am saying, is to believe that moral feelings are warranted just
when they meet the standards of Brandt’s qualified attitude method.
This is to adopt a thesis in the epistemology of normative beliefs.
 Normative Insights Sources of Normative Knowledge? 209

Is it true that moral feelings are warranted if and only if they meet
Brandt’s standards? Nothing I have said so far speaks to this. Rather,
I have been explaining the meaning of this claim—the claim that an at-
titude is warranted if and only if it meets the standards of Brandt’s
qualified attitude method. I explained it in the expressivist’s way, by
saying what believing the claim consists in. To believe the claim, I said,
is to plan (in my special sense) to have moral feelings that meet Brandt’s stan-
dards. Coming to believe the claim thus consists in a kind of planning.
If you ask me whether the claim is true, I come to my answer by plan-
ning how to feel about things.
What happens, then, when I attempt this? I don’t come to a plan
that is fully worked out, but I do come to a plan roughly. Brandt’s ac-
count of how to test moral attitudes strikes me as on the right track, and
so in my planning for what moral feelings to have, I at least roughly ad-
here to Brandt’s standards. I can now express the state of mind I am in,
saying this:

At least roughly, moral feelings are warranted if and only if they

meet the standards of Brandt’s qualified attitude method.

M o ral Knowl e dge

When, then, do moral feelings yield moral knowledge? Brandt dis-

cusses southwestern American Indians of a century ago playing a game
of burying a chicken up to its neck and riding by on horseback trying
to pull the chicken out of the ground. They accepted that this hurts the
chicken but didn’t see this as making the game wrong. I myself, though,
have feelings of moral condemnation for playing this game of torment-
ing an animal. Do these feelings give me knowledge that playing the
game is wrong?
Responding to this question requires an account of when a belief
constitutes knowledge, and a prominent kind of account is the one I
already gave roughly. First, a belief constitutes knowledge only if the
belief is true. For it to constitute knowledge, moreover, it must be war-
ranted. Finally, it must stem from processes that deliver truths reliably
 210 Allan Gibbard

and have the features, in this particular instance, in virtue of which the
process is reliable in general.
Suppose I am right, then, that moral feelings are warranted when
they satisfy the standards of Brandt’s qualified attitude method. What
should we say of the case where I consider the chicken-pull game and
come to believe it wrong by relying on my qualified attitudes? I have a
qualified attitude of moral condemnation toward playing the game, and
accordingly I come to the belief that playing the game is wrong. This
belief constitutes knowledge if it is true and warranted and it stems from
a process that reliably produces true moral beliefs, operating in the ways
in virtue of which it tends to produce true moral beliefs. If these con-
ditions obtain, then an affect-laden moral insight—my feeling of con-
demnation toward playing the game—has yielded moral knowledge.
What, though, if the moral insights of one person clash with the
moral insights of another? Brandt’s investigations into memories of
playing chicken-pull indicate that this is such a case. Cultures, it appears,
disagree fundamentally in their attitudes toward cruelty to animals, and
I have drawn the chicken-pull example from Brandt’s study of Hopi
ethics; the game he reports was from the early twentieth century, as re-
called in midcentury.5 We readers abhor games that torment animals, but
young Hopi men of a century ago had no such response. There is no in-
dication that their qualified attitudes would be different—that if their
attitudes met the standards of Brandt’s qualified attitude method, they
would disapprove.
Moral claims that contradict each other can’t all be true, and moral
insights that clash with moral truths can’t be said to deliver moral
knowledge. People whose qualified attitudes disagree, though, won’t
agree on which moral claims are true, or which psychic processes are
reliable in coming to feelings that are warranted. I do claim to know
that tormenting animals for fun is wrong, but I don’t have arguments
that will bring everyone’s qualified attitudes into accord with mine.
A Hopi of a century ago might reason with perfect coherence, in a way
that satisfies every reasonable epistemic standard, and reject my con-
tention that playing the game is wrong. (Or perhaps the right epis-
temic standards for moral beliefs include a requirement to abhor cru-
elty, but maintaining this would beg the question on which these two
cultures disagree.)
 Normative Insights Sources of Normative Knowledge? 211

Te m p e re d Moral Realism

In making moral judgments, we must rely ultimately on our moral

intuitions. Sidgwick was right to maintain this. The intuitions should
be scrutinized critically, but there is no way to jettison them and re-
spond to moral questions independently of them. There appear to be
cases, though, I am worrying, where cultures differ in their moral atti-
tudes and would continue to differ even if their attitudes satisfied all
of Brandt’s standards. Could one group be right in such a case and the
other group be wrong? It is happenstance, in a way, whether I belong
to one group or another, an accident of fate. How can a belief that I ar-
rive at by such happenstance count as knowledge?
In parallel cases, after all, we wouldn’t ascribe knowledge. Suppose
two groups of medical doctors disagree as to which of two treatments
will cure me of a disease. Ignorant of this conflict, imagine, I choose
one group by chance and trust what they tell me. Even if I luckily con-
sulted the right group, I didn’t thereby come to know which treatment
would cure me. I might have come to a true and warranted belief but
not to a state of knowledge. As for the group with the right answer,
for them to count as knowing which cure will work, they must lack
some epistemic defect that accounts for the other group’s disagreeing
with them. If, for example, the group with the wrong answer can’t even
understand some of the pertinent arguments, then the fact that they
don’t agree doesn’t impugn the claim of the group with the right answer
to know. Doesn’t the same go for my belief that tormenting an ani-
mal for fun is wrong? If we qualify as knowing that playing the game
is wrong and the Hopi of a century ago whom Brandt studied didn’t
qualify as knowing that it isn’t wrong, that can only be because they
suffered some epistemic defect.
Metaethical debates are often couched in terms of “moral realism”
versus “moral antirealism.” A crucial development in expressivism has
been the rise of what Simon Blackburn labels “quasi-realism.” This is
the doctrine that we can start out with the materials of a moral anti-
realist and earn the right to speak as a moral realist does. Moral real-
ists, for example, claim that moral statements can be true or false. This,
however, doesn’t distinguish a moral realist from a moral expressivist,
for the quasi-realistic expressivist agrees that moral statements can be
 212 Allan Gibbard

true or false. He accepts a “deflationary” account of truth: the claim

“It’s true that snow is white,” he says, is equivalent to the plain claim
that snow is white. Likewise, he says, “It’s true that stealing is wrong” is
equivalent to “Stealing is wrong.” Thus if an Ayer-like emotivist says
that “Stealing is wrong” means “Boo for stealing!” he’ll say the same for
the claim “It’s true that stealing is wrong.” Expressivists can also ac-
count for realists’ claims to moral objectivity: A moral realist might say,
“Chicken-pull would be wrong whether or not anyone disapproved,”
and an Ayer-like emotivist can agree. This means, the emotivist can say,
“Boo for chicken-pull, whether or not anyone disapproves!”
In considering claims to knowledge, though, we have found a
respect in which expressivists should side against moral realists of
an extreme kind. A strong moral realist might claim that true moral
claims state moral facts, and that moral facts are factlike in every way.
Here, though, we encounter a way in which moral “facts” and the cases
that are most clearly matters of fact differ. With facts like whether the
propensities of our species were shaped by millions of years of natural
selection, if two people disagree fundamentally and neither has a clear
epistemic defect that is independent of who happens to be right, then
even the side with the right doctrine fails to know. Moral knowledge
differs in this regard: we know that tormenting an animal for fun is
wrong even if those who disagree have no epistemic defect apart from
failing to recognize that such cruelty is wrong.
Are we warranted in this moral judgment? The judgment, I am say-
ing, amounts to a plan for how to feel about chicken-pull. In judging
that playing the game is wrong, I am coming to a plan for how to feel
about playing the game. I am planning to feel obligated not to play such
a game, even for the hypothetical case of being a Hopi of a century ago
who doesn’t feel obligated not to play. Surely, in planning how to feel, I
can only go by the feelings I am prone to have in epistemically ideal
conditions. The standards of Brandt’s qualified attitude method, I am
taking it, capture what sorts of conditions are epistemically ideal for
moral judgment. If this is right, then my judgment that playing chicken-
pull is wrong, even if no one disapproves, is warranted. I now voice my
warranted judgment:

Playing chicken-pull is wrong even if no one disapproves.

 Normative Insights Sources of Normative Knowledge? 213

Most philosophers who regard themselves as “moral realists” take

this tempered position. Even if some people with no independently
identifiable epistemic defect wouldn’t agree, they say, that doesn’t en-
tail that we fail to know that playing the game is wrong. This position
isn’t consistent, though, with the extreme moral realist position that
moral facts are factlike in every way that a nonmoral fact can be. In
this regard, we might say that moral facts are “quasi-facts,” and the
correct metanormative view isn’t a form of full realism but of quasi-
realism.
There is such a thing as moral knowledge, I am maintaining, and it
does rest on moral intuitions. It is different, though, from knowledge
of such natural facts as that natural selection accounts for the genetic
propensities of our species. This belief is warranted for anyone with
our evidence: the layout of fossils, similarities of DNA, and the like.
Not so for the belief that being cruel to animals for fun is wrong: this
belief is warranted for those with our affective sensibilities but not for
everyone who has our evidence. Communities whose affective sen-
sibilities differ from ours may differ in which moral beliefs are war-
ranted for them. We are warranted in our conviction that it was wrong
for Hopi young men to play the chicken-pull game, even though they
weren’t prone to feeling obligated not to play it. The Hopi of a cen-
tury ago would have been warranted in rejecting this conviction.
So can normative insights be sources of normative knowledge?
Normative insights are affective responses, and when they meet cer-
tain conditions, I have concluded, what they give rise to is not ex-
actly normative knowledge. It is what we might call normative “quasi-
knowledge.” Our affective responses can justify us in believing those
responses warranted.

Not e s

1. Such a view is proposed by Ewing 1939.

2. Kahneman (2011) speaks of “Type one” and “Type two” processes.
3. Dworkin 1996; Scanlon 1998; Shafer-Landau 2003; Enoch 2011;
Parfit 2011.
4. Another form of expressivism is developed by Simon Blackburn in
Spreading the Word (1984), Ruling Passions (1998), and other writings.
 214 Allan Gibbard

5. Brandt discusses whether moral disagreement over cruelty to animals

is fundamental: Hopi Ethics (1954). I discuss Brandt’s finding in my “How
Much Realism” (2011).

R e f e re nce s

Ayer, A. J. 1936. Language, Truth and Logic. London: Victor Gollancz. 2nd
ed. 1946.
Bentham, Jeremy 1789. An Introduction to the Principles of Morals and Legis-
lation. London: T. Payne.
Blackburn, Simon. 1984. Spreading the Word: Groundings in the Philosophy of
Language. Oxford: Clarendon Press.
———. 1998. Ruling Passions: A Theory of Practical Reason. Oxford: Claren-
don Press.
Brandt, Richard B. 1954. Hopi Ethics: A Theoretical Analysis. Chicago: Uni-
versity of Chicago Press.
———. 1959. Ethical Theory. Englewood Cliffs, NJ: Prentice Hall.
———. 1979. A Theory of the Good and the Right. Oxford: Clarendon Press.
Dworkin, Ronald. 1996. “Objectivity and Truth: You’d Better Believe It.”
Philosophy and Public Affairs 25 (2): 87–139.
Enoch, David. 2011. Taking Morality Seriously: A Defense of Robust Realism.
New York: Oxford University Press.
Ewing, A. C. 1939. “A Suggested Non-Naturalistic Analysis of Good.” Mind
48: 1– 22.
Gibbard, Allan. 1990. Wise Choices, Apt Feelings: A Theory of Normative
Judgment. Cambridge, MA: Harvard University Press.
———. 2003. Thinking How to Live. Cambridge, MA: Harvard University
Press.
———. 2011. “How Much Realism? Evolved Thinkers and Normative Con-
cepts.” In Oxford Studies in Metaethics 6, ed. Russ Shafer-Landau, 33– 51.
Oxford: Oxford University Press.
Haidt, Jonathan. 2012. The Righteous Mind: Why Good People Are Divided
by Politics and Religion. New York: Pantheon.
Joyce, Richard. 2001. The Myth of Morality. Cambridge: Cambridge Univer-
sity Press.
Kahneman, Daniel. 2011. Thinking Fast and Slow. New York: Farrar, Straus &
Giroux.
Mackie, John L. 1977. Ethics: Inventing Right and Wrong. Harmondsworth:
Penguin Books.
 Normative Insights Sources of Normative Knowledge? 215

Mill, John Stuart. 1863. Utilitarianism. London: Parker, Son, & Bourn.
Moore, G. E. 1903. Principia Ethica. Cambridge: Cambridge University Press.
Parfit, Derek. 2011. On What Matters. 2 vols. Oxford: Oxford University
Press.
Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University
Press.
Scanlon, T. M. 1998. What We Owe to Each Other. Cambridge, MA: Harvard
University Press.
Shafer-Landau, Russ. 2003. Moral Realism: A Defence. Oxford: Clarendon
Press.
Sidgwick, Henry. 1907. The Methods of Ethics. 7th ed. London: Macmillan.
1st ed. 1874.
Whewell, William. 1845. Elements of Morality including Polity. London: John W.
Parker.
 C h a p t e r T e n

Truth and Knowledge in Literary Interpretation

-
Carsten Dut t

Sabine MacCormack, in memoriam

As the title of my essay indicates, I am not dealing here with interpre-

tation in general—a far-reaching and highly complex matter1—but in-
stead with a specific (and notoriously controversial) academic genre of
it, which is— for better or worse — my genre: literary interpretation.
My question is, What is the role and relevance of truth and knowledge
in literary interpretation? In order to answer this question, I will argue
that the pursuit of truth and of what Bernard Williams has called its
virtues— accuracy, circumspection, intellectual integrity2— sustain
literary interpretation as a knowledge-seeking and, when felicitous,
knowledge-enhancing practice. Furthermore, caring about truth, not
only in the sense of seeking to answer veridically each single question
one may raise in interpreting a work of literature but also in the sense
of subscribing to the ideal of a comprehensive true account of that very

216
 Truth and Knowledge in Literary Interpretation 217

work, could help us to counter and perhaps halt the ongoing tendency
toward uncritical pluralism, parochialism, and mutual noncommuni-
cation between different schools— some observers have called them
“tribes”3—of literary interpretation. Although, according to my view,
truth is and should be acknowledged as the standard goal for forming
beliefs and making assertions in interpreting literature, it is undeniable
that some typical results of literary interpretation and criticism do not
fall under this description and its cognitively normative force. For
clearly some of the remarks that critics offer in lectures, books, or ar-
ticles on works of literature have a different aim and scope from that
of discovering and ascertaining what is true about these works in terms
of genesis, structure, meaning, function, impact, or reception. I’m re-
ferring to remarks with which critics seek to flesh out what is only
suggested or even left open in a literary work. For example, critical dis-
course quite often weaves together what as a story line is only loosely
connected, noncausally juxtaposed, or even randomly scattered, and
thus in certain respects undetermined. Also, critical discourse quite
often rounds out characters by imaginatively supplementing motives
for their thoughts or actions which are in fact not, either explicitly or
implicitly, given in a literary work. The great Polish phenomenologi-
cal philosopher and literary theorist Roman Ingarden has dubbed this
type of interpretive activity concretization.4 Now, obviously, within
the framework of a realist conception of truth, to which— in spite of
all antirealist and constructivist rhetoric — human beings in practice
do and should adhere,5 concretizations as complementing determina-
tions of what in a literary work are places or zones of indeterminacy
can be neither true nor false, for there is no fact of the matter that would
allow for ascribing positive or negative truth values to them. So within
certain limits of aptness different and even mutually exclusive con-
cretizations of a literary work can be interpretively legitimate and fruit-
ful, even though they are not acquisitions of truth and a fortiori not
acquisitions of knowledge. I argue that a sufficiently nuanced account
of literary interpretation has to take this into account. It has to observe
that in addition to being a cognitive process of acquiring truths and
knowledge, literary interpretation is also a creative process of adding to
the aesthetic, philosophical, moral, or religious value of experiencing a
 218 Carsten Dutt

work by pertinent and appropriate operations of concretization. In

this respect literary interpretation fulfills a function quite similar to
that fulfilled by performative interpretations of musical scores in the
concert hall or of dramatic works on stage. It would be mistaken, how-
ever, to think of the cognitive and the creative functions of literary in-
terpretation as neatly separable. In fact, interpretations that enhance
the value of experiencing a literary work by showing it in the light of
creative concretizations are cognitively dependent on that work’s in-
terpretation in terms of truths about its intrinsic and relational proper-
ties. In order appropriately and productively to complement a work’s
zones of indeterminacy, literary critics have first to identify these zones
and second to discern how they are embedded in well-determined
structures. Furthermore, quite differently from performative interpre-
tations, which by their very choices (in playing a composition or stag-
ing a play, etc.) practically exclude each other, critical interpretations
of literature can theoretically integrate different and even mutually ex-
clusive concretizations on a higher level of understanding. And this
is one of the many reasons why subscribing to the idea of a compre-
hensive and true account of a literary work is indispensable for good
interpretive practice.

Let me start by summarizing some insights about interpretation as an

endeavor of acquiring truth and knowledge.

1. Pace Nietzsche and Heidegger, not all perception or apprehension

can aptly be viewed as interpretive; some is clearly non- or (as Rich-
ard Schusterman has put it) preinterpretive perception and appre-
hension.6 And typically, preinterpretive states of perceiving and
apprehending in which we acquire, process, and store a lot of in-
formation about the world around us serve as the points of depar-
ture for interpretive activities proper.
2. Interpretive activities proper involve the formation of questions
and hypotheses about a given object, the interpretandum. They in-
volve the exploration and weighing of possible classifications in
terms of genesis, structure, meaning, purpose, or function regarding
 Truth and Knowledge in Literary Interpretation 219

the interpretandum. And therefore interpretation proper involves

conscious thought, explicit reflection, inference, and reasoning.
3. Interpretation proper requires some form of nonobviousness: “the
nonobviousness of what is being interpreted” to the interpreter.7 If
he or she securely knows that x is F, then forming and stating the
belief that x is F for her are not acts of interpreting x.8 Interpreta-
tion proper is a step or more commonly a series of steps toward
knowledge one seeks to have about x but does not yet have.
4. However, interpretation proper also requires some knowledge
pertaining to x, the object in question. Merely guessing that x is F
is not an act of interpreting x. The concept of interpretation is un-
doubtedly more demanding: it is internally linked with the con-
cepts of justification and reliability in the process of forming inter-
pretive beliefs and coming up with the corresponding interpretive
assertions about a given object.9 When we interpret an object with
regard to some question of the type, is x F? we standardly do not
flip coins; for flipping coins is not a reliable and intellectually jus-
tifiable way of belief formation; it may lead us to a true belief and
to a true assertion about x, but only accidentally, only by chance;
and in the business of interpretation we don’t want to be subjected
to chance — at least not when we take that business seriously; we
want to arrive at beliefs that are true or most likely to be true, and
these beliefs are, of course, the well-grounded ones, beliefs sup-
ported by sound reasoning and sufficient evidence. In order to ar-
rive at interpretive beliefs of such quality, we have to have relevant
propositional knowledge and to know how to use it productively
in the process of interpreting the object in question.10 There is no
interpretation from scratch. In other words: felicitous interpreta-
tion adds to the knowledge we must have and make use of in order
to get our interpretive business going.

Let me flesh out what I have said with an example—not from lit-
erature, but from the visual arts. A rather marginal case of everyday
interpretation, it will take us from the so-called lifeworld to scholarly
contexts of interpretation. So let us suppose we take a stroll through
the lovely Pinacoteca Communale di Faenza in Italy. Suppose that our
 220 Carsten Dutt

attention is caught by a strikingly beautiful painting by the seventeenth-

century artist Francesco Maffei (1605– 60), representing a young woman
with a sword in her left hand, who, with elegant support from her right
hand, is holding on her left arm a platter with the decapitated head of
a bearded man on it (fig. 10.1).11
According to the concept of interpretation I have just outlined,
we do not interpret Maffei’s painting when we perceive or describe the
depicted figures in such a way. We do not interpret; we simply see and
know according to our habits of seeing that this painting depicts a
young woman with a sword in her left hand and in her right a platter
with the decapitated head of a bearded man on it. And that is that. But
now interpretive-minded gallery visitors among us may start asking
themselves who exactly this young woman and her trophy are, and in
doing so at least some of these visitors may not be restricted to purely
guessing; instead they may embark on the business of interpretation by
using pertinent knowledge, biblical and iconographic knowledge in this
case. Some may hold that Maffei’s painting shows Salome with the head
of Saint John the Baptist on the platter, mentioned in the Gospel of Saint
Matthew;12 others may doubt this interpretation and point to the sword
in the young woman’s hand, which scarcely accords with the story of
Salome. Rather it indicates that the depicted figure might be Judith with
the head of Holofernes, whose sword the brave young lady took “and
approached to his bed, and took hold of the hair of his head, and said,
‘Strengthen me, O Lord God of Israel, this day.’ And she smote twice
upon his neck with all her might, and she took away his head from
him.” But what about the platter? As we all know, Judith “gave Holo-
fernes’ head to her maid; And she put it in her bag of meat.”13 There is
no platter in the Book of Judith, but there is a platter in the Gospels
story of Salome and the beheading of Saint John the Baptist.
Now, this is what we call a conflict of possible interpretations: it is
a paradigmatic case of divergent evidence in forming interpretive be-
lief. And note that this conflict about scriptural reference and icono-
graphic content is not a purely positivist or antiquarian one; it is of
deep import to the understanding and appreciation of the painting —
its hermeneutic and aesthetic identity. For it is obvious that under the
description of Judith with the head of Holofernes the visual salience of
Figure 10.1. Francesco Maffei (?), Judith, ca. 1640, oil on canvas, 26.7 x 35.4 in.,
Pinacoteca Comunale di Faenza, Italy
 222 Cars ten Dutt

the depicted scene and especially of the depicted woman will differ
strongly from its visual salience under the description of Salome with
the head of Saint John the Baptist. Just look at these lascivious eyes . . .
Now, some ludic minds may be fine with this situation and enjoy the
back and forth of taking Maffei’s painting now this way and then the
other: Is it a duck, or is it a rabbit? Neither, of course, for it is Judith
with the head of Holofernes, as Erwin Panofsky convincingly argued
more than fifty years ago against older interpretations that identified
the depicted lady as Salome.14 Panofsky did so by providing evidence
from what art historians call the history of types (Typengeschichte). The
pertinent materials show that it was quite common in northern Italian
art of the period to “transfer” something like Saint John’s platter to de-
pictions of Judith (evidenced as such by the presence of Judith’s maid
in some of these paintings), whereas there hasn’t been found a single
painting in which Salome would have been ennobled with a sword like
that of Judith. Also, the features of the decapitated head have more in
common with Holofernes heads than with Saint John the Baptist heads.
Of course, Panofsky’s comparative evidence, his argument from the
history of types, is not a watertight proof for the truth of his icono-
graphic claim. Positivistic art historians would surely expect more
and different external evidence, ideally perhaps a contract between the
painter and a noble commissioner.
I am convinced that Panofsky would have very much appreciated
such further evidence, for he undoubtedly shared an intentionalist
stance toward the iconographic interpretation of that painting in par-
ticular and the visual arts in general. According to an intentionalist
framework of interpretation, an interpretive statement is true if it re-
ports correctly the intentions an artist has realized by painting a certain
picture—in this case Francesco Maffei’s intention to paint a Judith
with the head of Holofernes in close-up and upgraded with a platter.
Needless to say that artists as well as writers form and realize their artis-
tic or writerly intentions within the constraints of artistic or literary
conventions—as we form and realize our semantic and communicative
intentions as speakers of a natural language within the constraints of its
grammar and certain communicative rules and conventions.15
The given example, I think, nicely illustrates how wonder and a
fortiori the awareness of a certain lack of knowledge vis-à-vis an ob-
 Truth and Knowledge in Literary Interpretation 223

ject initiate interpretation and how interpretation typically proceeds in

order to reach and secure its results: some questions, at least one, come
up; pertinent knowledge is activated and applied to form hypotheses
and to explore interpretive possibilities; the evidence at hand is assessed
but does not allow for a justified and reliable decision between inter-
pretive possibilities; additional evidence is searched for, achieved, and
weighed; the less plausible hypothesis is dropped in favor of a well-
founded and therefore very likely true interpretation; the correct an-
swer to the interpretive question (or questions) at stake. Note how a
rather tentative and probing mode of interpretation—“It could be a
depiction of Salome”; “It could also be a depiction of Judith”—develops
into a determinative mode: “It is most likely a depiction of Judith, and
not of Salome”; “It is a depiction of Judith.”16 Interpretation initiated
by the awareness of a certain lack of knowledge aims at filling that very
lack. It aims at acquiring and justifying relevant true beliefs about the
object interpreted, and in that sense interpretation standardly aims at
knowledge, for knowledge is, according to the standard analysis of the
concept, justified true belief.17

In literary interpretation, to which I turn now, questions of knowledge-

seeking and therefore truth-oriented interpretation come in various
sizes or, to put it differently, on various levels of the object interpreted.
Some of them apply to individual words, phrases, or sentences of
literary works: “Are they ambiguous?” “Are they ironic?” “Should
this verse be understood literally or metaphorically?” Other questions
apply to larger parts of a given work and some to the work as a whole:
“What is its thematic content?” “What is its overall outlook and aim?”
“What is its communicative purpose and function?” And yet other
questions apply to the work in relation to other works or to its social,
cultural, philosophical, or biographical contexts, and so on. The web
of interpretive issues in dealing with literature is chronically vast and
full of repercussions: findings on the level of microinterpretation often
affect our results on the level of macrointerpretation—and vice versa.
A single ironic word in a seemingly full-blown happy ending of a novel
can alter the work as a whole, and overlooking that single ironic word
would result not in a local but in a total misinterpretation. Needless to
 224 Carsten Dutt

underscore that answers to a certain interpretive question typically

prompt a whole bunch of follow-up questions about the work and/or
its contexts. That is why I am offering the metaphor of a web of in-
terpretive questions, issues, and tasks. For example, we simply can-
not discuss or answer questions about the communicative purpose and
function of a certain work without discussing and answering questions
about the relation between its formal and semantic features, and so on.
Now, what does the literary interpreter need to know in order to
deal accurately and circumspectly with this web of questions? Clearly
she has to master the concepts involved. She has to master genre con-
cepts like “comedy,” “sonnet,” “novel,” or “tragedy”; she has to mas-
ter linguistic and stylistic concepts from general and apparently simple
ones like “word” or “sentence” or “stanza” to more specialized ones
like “ottave rime,” “genus grande,” or “mise en abyme”; she has to
master aesthetic concepts like “the sublime” or “the beautiful” and
historical concepts like “romanticism” or “classicism.” The literary in-
terpreter needs, to a certain degree, mastery of these concepts in order
to find out whether and, if so, how they apply to the literary work she
is interpreting. On the other hand, by making use of her conceptual
mastery the interpreter will be enabled to refine it, for by accurately
and circumspectly applying literary concepts in the business of inter-
pretation, a critic will gain a more nuanced outlook on the intension
and extension of the concepts she uses. Thus successful literary inter-
pretation serves the enhancement not only of empirical but also of
conceptual knowledge.
Let us now turn to some objections against the cognitivist account
of literary interpretation for which I have argued. Sentences with which
literary interpreters answer interpretive questions—sentences like “X is
a tragicomedy”; “Y is a parable that advocates the value of forgiveness”;
“Z is not as one might think a political proposal but a political satire”—
are assertions by which critics present themselves as having knowledge,
that is, appropriately justified true beliefs, about the works interpreted.
But can they have such knowledge? Some theorists of literary inter-
pretation, in the English-speaking world most notably the philoso-
phers Joseph Margolis18 and Robert J. Matthews,19 have denied this.
According to them there can be no such thing as veridical interpretive
 Truth and Knowledge in Literary Interpretation 225

knowledge. Why not? Because, according to these theorists, there are

no truths to be found in the process of interpretation. Interpretive state-
ments in literary and art criticism, Matthews argues, are “typically nei-
ther true nor false”; they are “as a general rule epistemically weak . . .
because they are radically underdetermined by the interpreted work.”20
In which sense should such statements be underdetermined? Here is
Matthews’s answer:

Statements are true (or false), when they are, in virtue of certain
facts about the world, but “the facts” typically fail to decide the
truth or falsity of interpretive statements in art criticism. To main-
tain that these statements are nonetheless “either true or false” is
in effect to deny the special relation that true (and false) statements
bear to the world. . . . Clearly there is, as many have suggested, a
similarity between the interpretations of the critic and those of the
performing artist. . . . [I]n both cases the interpretation is under-
determined by the interpretandum. In neither case does the inter-
pretandum determine a unique interpretation; rather it constrains
them only loosely. And precisely because interpretation does in-
volve “going beyond” the given, interpretations require the added
contribution of the interpreter.21

Is Matthews right? I don’t think so, for he unduly generalizes a

special type of interpretive statement that in a sufficiently nuanced ac-
count of literary interpretation—let me call it a differentialist account—
should be distinguished from interpretive statements whose truth or fal-
sity is determined by the interpretandum they refer to.
Indeed, Hamlet’s build and walk are not determined by Shake-
speare’s play; each actor has to add these and other features as part of
his interpretation to “complete” the tragic prince of Denmark. And as
we all know, from Laurence Olivier to Derek Jacobi and Simon Rus-
sell Beale, there is a wide and undoubtedly still extendable range of in
many aspects radically different but equally outstanding performances
of Hamlet. Not only would it be closed-minded caviling; it would be
a misuse of the term true if one should allocate it exclusively to Oli-
vier’s or exclusively to Jacobi’s interpretation of Hamlet. Great creative
 226 Cars ten Dutt

interpretations by performance are not “true,” even though they are

eye-opening in the sense that they show us moving, gripping, thought-
provoking concretizations of Shakespeare’s drama we were previously
unable to imagine.
Analogously, critics quite often add discursively to a literary work
by filling out what Roman Ingarden in his theory of the (partly) sche-
matic character of the literary work of art has called “places of inde-
terminacy” (Unbestimmtheitsstellen eines literarischen Werkes).22 Think
for instance of a critic whose interpretation deals with the motives of a
fictional character— or more likely a whole set of fictional characters.
Typically, there are many places or zones of indeterminacy in a fic-
tional work’s representation of its characters. Why does Mme Chauchat
leave The Magic Mountain after her passionate one-night love affair
with Hans Castorp? Out of decadent tedium or in order to protect the
vulnerable young German from the sufferings of infatuation? Though
even this question might be suggested by the work itself, it contains no
information, neither explicit nor implicit, that could answer it. Crit-
ics who nevertheless come up with an answer in their rendition of The
Magic Mountain’s narrative or in their characterization of its first part’s
heroine (or antiheroine?), neither discover nor miss a hidden truth in
the work; instead they transform an indeterminacy built into Mann’s
novel into a more determinate but comparatively partial account of the
work. Interpretive statements of this kind— though they may more
or less fit in the overall context and be more or less consistent with the
“laid down” character traits of Mme Chauchat— are neither true nor
false. For there are no facts, that is, neither stretches of discourse nor
logical or pragmatic implications of pertinent stretches of discourse,23
in Thomas Mann’s Magic Mountain that would determine whether
Mme Chauchat acted out of this or rather out of that motive.
Admittedly, there are typically many zones of indeterminacy within
literary works.24 But this does not warrant Matthews’s and other theo-
rists’ exaggerated claim that literary interpretation is “epistemically
weak” by just hinging on indeterminacy. On the contrary, it seems
to me that carefully determining (and this tasks includes, of course,
carefully explaining) the scope and forms of indeterminacy in a given
work— syntactic indeterminacy is one form, narrative indeterminacy
 Truth and Knowledge in Literary Interpretation 227

another— is and should be one of the primary tasks of literary inter-

pretation in its truth- and knowledge-seeking mode.
Let me once again quote the suggestive but potentially distortive
metaphor used by Matthews that literary interpretation does involve
“going beyond” the given. In order to make sense of this job descrip-
tion, the ones on the job should be able to determine where they are
“going beyond” what is given. If a (not local but total) principle of
indeterminacy would rule out that possibility, going beyond would
not be going beyond. And only if we have knowledge about a literary
work’s determined features will our creative efforts of going beyond
these features and thereby adding to the value and significance of ex-
periencing that very work have a good chance of being appropriate
and productive.
Of course, in order to interpret a given literary work we do con-
stantly “go beyond”: beyond its given text. Granted that the original
wording has been preserved or soundly reconstructed, we classify and
explain a work’s features and its overall status not just in terms of its
text but also in terms of realized authorial intentions, effective literary
conventions and aesthetic rules, historical circumstances, and so on.
Let me therefore recall the utterly important distinction between a lit-
erary work and its text.25 To ask and determine what is in a text is not
the same as asking and determining what is in a work. A text has lexi-
cal, syntactic, and semantic features, identification of which determines
the text’s content. But this is linguistic, not yet literary, content, a prod-
uct of linguistic conventions, of syntactic and semantic rules that bear
on the language in question. A text that is not contextualized further is
in many aspects not subject to interpretive constraints and can mean
or rather be made to mean anything that the language for that stretch
of written discourse permits. In contrast, a literary work is contextual-
ized; not only is it written in a certain time and culture, but it is also
typically written for a certain purpose or set of purposes and therefore
typically invested with semantic, pragmatic, and aesthetic intentions
its author has realized by writing that very work. Retrieving this invest-
ment by true and aptly justified statements about the work and thereby
achieving an understanding of it as the product of the intentional ac-
tivity of its historically situated author is and will be an indispensable
 228 Carsten Dutt

step for knowledge-seeking literary interpretation—in spite of all the

arguments the different schools of rigid anti-intentionalism have
brought up.26 Ignoring authorial intent is clearly unwise and in many
cases a ticket to arbitrarily misusing a work and missing the attainable
truth about it. Saying this, I do not at all deny that the meaning of a lit-
erary work and the meaning that its author intended in the process of
writing it can and do often diverge. I do not at all deny that a literary
work is often richer and more complex than its author himself would
have thought. Furthermore, the idea that knowledge-seeking inter-
pretation acquires and justifies true beliefs about a literary work is not
restricted to the task of establishing which semantic, pragmatic, or
aesthetic intentions have been successfully realized in a literary work.
Not only bygone states of mind that have shaped a work’s features and
thereby left their imprint on them are “truth makers” for interpretive
claims; also public semantic facts and in virtue of them a wide range of
semantic properties can objectively be part of a literary work’s meaning
and therefore “truth makers” for interpretive claims about it. As Tom
Stoppard wittily and insightfully suggested, the relation between an
author and a critic resembles that between a traveler and a customs in-
spector: The inspector opening the traveler’s suitcase finds a lot of
things in it which the traveler has to acknowledge are in it, although he
can truthfully protest that he has not consciously packed them.27

Let me conclude by getting back to the two main levels on which I see
knowledge as a central goal and the pursuit of it as a central obligation
for literary interpretation. As I said at the outset, the pursuit of knowl-
edge applies not only to each single question one may have to deal with
in interpreting a literary work. It also applies to the more demanding
task of achieving a comprehensive true account of a literary work that
weighs on and integrates the partial achievements of either comple-
mentary or conflicting interpretations of that very work. As a general
rule, interpretations, even excellent ones, are partial and incomplete
compared to the inexhaustible richness of great literary works of art.
So the old-fashioned new critic does not capture everything, nor does
the now-fashionable discourse analyst. But the scrappiness of first-
 Truth and Knowledge in Literary Interpretation 229

order interpretations can be overcome on a higher level of interpretive

integration, which, of course, cannot just be a container or some sort of
dustbin into which everything goes. Integrative second-order interpre-
tation should be a critical comparative assessment of extant readings. In
order to arrive at such an assessment, one has, of course, first and fore-
most to care about truth. Provided that interpreter A and interpreter B
haven’t fallen into the pitfalls of conceptual vagueness when they talk
about, say, pantheism, then A’s claim that a certain poem, Goethe’s
“Ganymed” for instance, is pantheistic and B’s claim that it is not can-
not both be true. But of course, it could be that the poem in question
shows signs both of embracing a pantheistic view of the world and of
rejecting or subverting such an outlook. Then neither of the two inter-
pretive claims would be comprehensively true, and truth-oriented in-
terpretation would be obliged to acknowledge ambiguity in the poem
and perhaps— if this follow-up question could be decided on the evi-
dence available—indecision on the poet’s side. There are undoubtedly
cases in which no first-order interpretation can be established over its
first-order rival. And that is exactly why a critically integrating second-
order interpretation is the better interpretation—a step in the ascent to
comprehensive truth and knowledge.

Not e s

Many thanks are owed to Robert Audi and Mark Roche for helpful com-
ments and excellent suggestions.
1. For thoughtful accounts, see Stuart Hampshire, “Types of Inter-
pretation,” in Art and Philosophy, ed. Sydney Hook (New York: New York
University Press, 1966), 101– 6; Göran Hermerén, “Interpretation. Types and
Criteria,” Grazer Philosophische Studien 18 (1982): 131– 61; Ronald Dworkin,
“Interpretation in General,” in Justice for Hedgehogs (Cambridge, MA: Har-
vard University Press, 2011), 123– 56.
2. Bernard Williams, On Truth and Truthfulness: An Essay in Genealogy
(Princeton: Princeton University Press, 2002), passim.
3. Dworkin, Justice for Hedgehogs, 126. For a straightforward relativist
account of tribal relativism in literary criticism, see Stanley Fish, Is There a
Text in This Class? The Authority of Interpretive Communities (Cambridge,
MA: Harvard University Press, 1982).
 230 Carsten Dutt

4. Roman Ingarden, The Cognition of the Literary Work of Art, trans.

Ruth Ann Crowly and Kenneth R. Olson (Evanston, IL: Northwestern Uni-
versity Press, 1973).
5. For a vigorous defense of “alethic realism,” see William P. Alston, A
Realist Conception of Truth (Ithaca, NY: Cornell University Press, 1996).
6. Richard Schustermann, “Beneath Interpretation: Against Hermeneu-
tic Holism,” The Monist 73 (1990): 181– 204.
7. See Jerrold Levinson, “Two Notions of Interpretation,” in Con-
templating Art: Essays in Aesthetics (Oxford: Oxford University Press, 2006),
275– 87. For an illuminating typology of epistemic situations that call for in-
terpretive efforts, see Annette Barnes, On Interpretation: A Critical Analysis
(Oxford: Blackwell, 1988).
8. Though, of course, she might interpret x as F for others who are not
yet in the know. Classroom interpretation is a striking example of this inward-
outward asymmetry in interpretation.
9. Note that this does not only apply to scholarly contexts but also to
everyday interpretation. Just think of interpreting a Japanese menu to a group
of fellow travelers. In order to be able to interpret that menu, you have to
know Japanese, or at least you have to know how to use a bidirectional English-
Japanese dictionary, the use of which would in this case be a somewhat reliable
way of forming interpretive beliefs and uttering corresponding assertions about
what the menu features.
10. As a general rule, the requisite knowledge for interpreting a given
object is of different kinds, overlapping the distinction between propositional
and practical knowledge.
11. More recently, the painting has been attributed to Bernardo Strozzi
(1581–1644) (see http://pinacotecafaenza.racine.ra.it/ita/opere/op_704.htm).
However, in the present context we can ignore the controversy over the paint-
ing’s authorship.
12. Matthew 14:8–12.
13. Judith 13:6–10.
14. See Erwin Panofsky, “Iconography and Iconology: An Introduction
to the Study of Renaissance Art,” in Meaning in the Visual Arts: Papers in and
on Art History (Garden City, NY: Doubleday, 1955), 26– 54.
15. For an enlightening discussion, see Robert Stecker, “The Role of In-
tention and Convention in Interpreting Artworks,” Southern Journal of Phi-
losophy 31 (1993): 471– 89.
16. For a discussion of these two modes of interpretation, see Levinson,
“Two Notions of Interpretation.”
17. The problems that the standard analysis of knowledge has to face since
Edmund Gettier came up with troublesome counterexamples are complex, but
 Truth and Knowledge in Literary Interpretation 231

as far as I can see, it has not been replaced by a more convincing account. See
Jonathan Jenkins Ichikawa and Mathias Steup, The Analysis of Knowledge
(http://plato.stanford.edu/archives/win2012/entries/knowledge-analysis/). Of
course, a circumspect approach to the theory of knowledge has to go beyond
justificationism. See William P. Alston, Beyond “Justification”: Dimensions of
Epistemic Evaluation (Ithaca, NY: Cornell University Press, 2005).
18. Joseph Margolis, The Language of Art and Art Criticism: Analytic
Questions in Aesthetics (Detroit: Wayne State University Press, 1965). For a
critique of Margolis’s position, see Monroe C. Beardsley, “The Testability of
an Interpretation,” in The Possibility of Criticism (Detroit: Wayne State Uni-
versity Press, 1970), 38– 61.
19. Robert Matthews, “Describing and Interpreting a Work of Art,”
Journal of Aesthetics and Art Criticism 36 (1977): 5–14.
20. Ibid., 5; original emphasis.
21. Ibid., 11–13.
22. Ingarden, Cognition of the Literary Work, 52.
23. For a general discussion, see Peter Lamarque, “Reasoning to What Is
True in Fiction,” Argumentation 4 (1990): 333– 46.
24. Also, there are different kinds or types of zones of indeterminacy. It
is clearly not sufficient to discuss them only in terms of fictional content. An
important type is indeterminacy of illocutionary role or force. To give just
one example: what kind of iussive subjunctive is uttered in the opening line of
Goethe’s poem “Das Göttliche” (The Divine)? “Edel sei der Mensch, / Hil-
freich und gut” (Let man be noble / generous and good). Is it an iussive of ad-
monition, or is it an iussive of encouragement? In literary interpretation, illo-
cutionary details of this kind quite often have to be determined by sensible
concretization. And it is not by chance that critics in order to substantiate and
justify their concretizations of what is undetermined in a work refer to what
is determined by it.
25. See Gregory Currie’s seminal paper “Work and Text,” Mind 100
(1991): S325– 40.
26. See the contributions in Gary Iseminger, ed., Interpretation and In-
tention (Philadelphia: Temple University Press, 1992), esp. Jerrold Levinson,
“Interpretation and Intention: A Last Look,” 221– 56.
27. Quoted after Barnes, On Interpretation, 121.
 C h a p t e r E l e v e n

Historical Truth

-
Aviezer Tucke r

This essay explores the meaning of historical truth. It argues that much
of historiography and the historical sciences are true in the sense of
employing reliable methods for generating sufficiently probable rep-
resentations of the past. How highly probable historiography must
be to be considered historical truth is context dependent. I defend this
probabilistic and epistemically contextual sense of historical truth as
better than its alternatives and argue against the plausibility of argu-
ments against the possibility of historical truth.
The historical sciences include textual criticism, comparative his-
torical philology, historiography (of the human past), phylogeny, evo-
lutionary biology, archaeology, natural history, and cosmology. I have
argued that over the past couple of centuries the historical sciences have
developed reliable methodologies for the inference of highly probable
representations of the past that in most contexts are considered histori-
cal truth. The philosophy of historiography should aim to discover
and describe these reliable, truth-conducive, methodologies. I have ar-
gued also that the methodologies of the sciences that generate histori-
cal truth, truth about the past, are similar. They use different informa-
tion theories about the transmission of information in time via different

232
 Historical Truth 233

media similarly to infer representations of the common sources, ori-

gins, of contemporary evidence that preserves information from that
origin (Tucker 2004).
How can we know that historians possess the reliable, truth-
conducive, methodologies that I claim they do? If they do, how can
philosophers know what these reliable processes of inference may be?
Historians do professionally many different things, and they some-
times disagree with each other on some of their ideas of best practice.
Bevir (1999) argued that the philosophy of historiography cannot fol-
low the philosophy of science in analyzing obvious “success stories”
like those of the paradigms associated with Galileo, Newton, Einstein,
or Darwin to discover what is science because such a methodology must
presuppose criteria for choosing what philosophers consider successful
science and there is no basis for justifying such criteria in the philoso-
phy of historiography. Yet, even without stating explicitly their crite-
ria and their tacitly assumed cognitive values, few historians would
doubt the success of Burckhardt’s discovery of the Renaissance in the
late nineteenth century. This discovery in the history of ideas has been
so successful that most people do not even realize that the Renaissance
is a hypothesis of the historiography of ideas that was introduced by a
Swiss student of Ranke a little more than a hundred years ago. Like-
wise, the discoveries of the Scientific and Industrial Revolutions (the
concepts did not exist during the historical periods they refer to) and
the documentary components of the Homeric sagas and the Old and
New Testaments are recognized by practically all historians to have
been successful. Historians do not spend much time contemplating
these successes because their professional practice consists of debates
with other historians rather than the philosophical analysis of what is
assumed within the professional consensus and therefore is not con-
troversial. Consequently, these success stories remain tacit in the sense
of being accepted and recognized by those who know them but are not
discussed explicitly exactly because they compose the received histori-
ographic tradition. Similar success stories can be found in historical
philology, the Indo-European hypothesis; phylogeny, its inferences of
common ancestry from genomic analysis; natural history, the tectonic
plates theory; and cosmology, in the inference of the Big Bang and the
 234 Aviezer Tucker

subsequent history of the universe. All these successes are in discover-

ing truths that were entirely hidden, unknown, for almost all the his-
tory of the human race. The task of philosophy then is to articulate
and explain, make sense of, the reliable institutional truth-conducive
practices of these historiographic successes.
The self-consciousness of historians about their own practices is
not useful for understanding historical truth and how historians infer
it. Historians disagree among each other about what they think they
are doing even when they clearly follow very similar practices. The re-
liable methodologies that historians use to infer historical truth are
often tacit. Historiographic institutional practices display the hall-
marks of what Collins (2010) called “collective tacit knowledge.” Train-
ing transmits tacit knowledge through an apprenticeship that cannot
be formalized and written down in a textbook. Tacit knowledge must
be acquired directly, by hanging around people who possess it, such as
teachers and peers. Collective tacit knowledge is acquired through so-
cial embedding. Explicit knowledge, by contrast, may be transmitted
via intermediaries.
The tacit nature of historiographic veridical methodologies allows
rationalizations. Rationalization of tacit knowledge means its explica-
tion in terms of contemporary fashionable epistemologies or ideologies
that are simple, easy to understand, and offer an epistemic legitimizing
facade. Rationalization explains the consecutive adoptions by histori-
ans of different generations of passing epistemic fads— nineteenth-
century empirical inductivism, early-twentieth-century positivism,
structuralism, and finally postmodernism—while practicing the same
archival methodologies within the same paradigmatic framework that
Ranke introduced to historiography almost two centuries ago.
Many of the classic textbooks on historiography present rationali-
zations as a professional ideology, a defense mechanism against exter-
nal criticisms. The result is often conceptually messy: the conflations
of explanation with causation, causation with conditionality, and all of
them with necessity, necessity with determinism, and necessity and
determinism with teleology and on and on.1 Symbolically, on the cover
of my Penguin paperback edition of E. H. Carr’s What Is History? there
is a reproduction of Representatives of Foreign Powers Arriving to Hail
 Historical Truth 235

the Republic as a Sign of Peace by the nineteenth-century master of

the naive style, Henri Rousseau. The relation between Carr’s rationali-
zation of historiography and historiography is not unlike the relation
between Rousseau’s painting and actual international diplomacy.
I have argued that the process of inference of representations of
past events in all the historical sciences amounts to the inference of
representations of information-generating origins from their present
information-preserving effects. The inference often is of models of in-
formation transmission “trees” or “bushes” that connect past events
to present evidence. The means of transmission of the information are
different in human history (testimonies and texts), historical linguis-
tics (languages), phylogeny (morphology and later DNA), and textual
criticism (documents). The information transmission and mutation the-
ories that these historical sciences use are different. But the stages of
inference are identical: They first attempt to prove that the evidence is
more likely given some common source of its shared preserved infor-
mation than given separate sources by finding dysfunctional homolo-
gies that are unlikely given separate sources such as biological rudiments
(such as the appendix or the wings of birds that cannot fly), infelicities
in texts, and testimonies that are not in the interest of the witnesses.
If they are successful, they try to infer the history of the transmis-
sion of information between the origin and its present effects. If they
have sufficient evidence and theoretical background they can achieve a
probable modeling of the transmission of information in time. Finally,
they infer the properties of the origins that transmitted the common
information, the origins of species and languages, historical events and
processes, and the original forms of texts. The process of inference is
Bayesian in the sense of producing a series of probabilistic inferences:
first that the similarities in the units of evidence are more likely given
preservation of information from a common origin than given sepa-
rate sources; then that the modeling of the transmission of the infor-
mation is probabilistic (explicitly quantitative in phylogeny and some
branches of historical linguistics, and tacitly so in the other historical
sciences). The ultimate inference of the state of the origin of the infor-
mation transmission is of course probabilistic in nature as well (Tucker
2004, 92–140; 2011).
 236 Aviezer Tucker

I want to argue here that the result of this typically historiographic

inference is historical truth (I refrained from discussing historical truth
in my book). When the result of this inference is probable enough in a
context, it becomes historical truth. As in jurisprudence, contexts af-
fect how high the probability must be to be considered the truth. I ex-
amine critically first alternative concepts of historical truth. Then, I
reject claims that there is no historical truth. Finally, I return to say
more about my probabilistic and epistemically contextual concept of
historical truth.

The most primitive form of rationalization is naive “empirical” real-

ism that considers historiographic knowledge empirical knowledge.
This is a strategy for suppression of the epistemology of our knowl-
edge of the past, against the examination of the tacit methods histori-
ans use to infer representations of the past. Suppression is not an argu-
ment. Historians do not and cannot possibly “see the past” and then
report what they saw as the authors of travel books report on what
they experienced in foreign lands, preferably by telling amusing anec-
dotes. Everything we know and can possibly know about the past is
always inferred, mediated through the evidence. Epistemology is built
into the historiographic enterprise, unavoidable and inevitable. Even
in the case of cosmology where we can see the past but cannot see the
present because of the time it takes light to reach us, the interpretation
of the light as traveling from distant stars and the correction of its color
for red or blue shifts and the use of telescopes are all information the-
ory laden and inferential rather than immediate.
Suppressive “empiricism” results in skepticism. Since the past is
inaccessible directly to any of our senses, a correspondence theory of
truth leads to the conclusion that there is no historical truth. With-
out direct access to the past, there is nothing for the truth to corre-
spond with.2
A more sophisticated strategy for resisting a philosophical exami-
nation of historical truth, while assuming it unreflectively, may be to
 Historical Truth 237

suggest that epistemic questions are not raised in the context of ordi-
nary historiographic inquiry because its results are probable enough
for that context. As much as in ordinary everyday contexts we are
not skeptical of the general reliability of our senses and the informa-
tion they transmit to us, arguably, in ordinary contexts, knowledge of
the past is sufficiently implicitly reliable and true. Contextualists like
DeRose (2009) and Lewis (1996) argued that skeptical doubts about
the possibility of knowledge do not arise in ordinary everyday con-
texts but only in specific philosophical contexts, and so can be ignored
in most contexts. In ordinary contexts we can ignore those distant few
possible worlds where we are brains in vats. A parallel argument may
be invented for knowledge of the past. However, there are many more
close and similar worlds where our knowledge of the past is different.
Historically, what most people believe about the history of humanity,
life, this planet, and the cosmos, has changed quite radically over the
past couple of centuries. We know about the Big Bang for less than a
hundred years. The theory of evolution revolutionized our concept of
the history of life 150 years ago. The Rankean archival methods are just
a little older. Textual criticism has changed the views of some people
about the scriptures for only the past quarter millennia or so. Some
parts of historiography are still contested or must work with difficulty
to distinguish historical truth from popular myth. By contrast, with
the exception of mystics, idealists, and some philosophers, people have
always believed in the existence of the world and the general reliability
of our senses, and even the idealists have behaved pretty much as if they
have been realists. The reliable methods that historians use to infer rep-
resentations of the past are almost always used in contexts where they
need to be examined critically rather than just contextually assumed.3

II

Peter Kosso (2001) advocated the most empiricist version of historical

realism that is not epistemically repressed and repressive. He followed
Dretske (1969) in expanding the meaning of “observation” to include
any transmission of information through the senses. If so, it may be
 238 Aviezer Tucker

possible to “observe” history through the evidence. Dretske distin-

guished in 1969 between primary and secondary ways of epistemic
“seeing.” Secondary ways of “seeing” are of objects through their ef-
fects on other objects that we perceive. Kosso loosened and broadened
the ordinary meaning of “observation” to encompass what Dretske
called a secondary epistemic way of seeing. Kosso suggested that an
information-bearing signal is conveyed through a series of interactions
beginning with historical events and concluding with the evidence.
Similar information-bearing signals connect evidence with events in
the natural sciences. Historiographic information-bearing signals are
considerably slower and are composed often of words, rather than of
photons, but arguably these properties do not distinguish epistemically
between historiography and science. Scientists understand the trans-
mission of light; historians analyze the fidelity of textual information.
Data in history play an evidential role that is similar to data in science,
images in microscopes, tracks in particle detectors, and so on. Both
bear information from less accessible objects of interest.
Evidence need not be observed to be perceived. But if we consider
historical events “observable” through the evidence, as Kosso pro-
posed, it becomes difficult to distinguish historical truth from histori-
cal evidence. Kosso is right that all the “observations” in science are of
the past. Trivially, photons that enter the retina begin their journey in
the immediate past. The difference lies elsewhere: All scientific obser-
vations in the theoretical sciences can usually be replicated when sci-
entists reproduce types of experiments or at least observations. All the
historical sciences, by contrast, depend on particular information causal
chains that cannot be replicated. If a particular information signal is
lost, it cannot be regenerated. Another way of putting this difference is
between sciences that are interested in information about a particular
token of an event and those that are interested in a type of event. The
latter sciences attempt to replicate or discover other tokens of the event
type. In the sciences that examine particular token events, they are not
replicable. For example, when scientists infer universal constants, they
are not interested in any particular token of them but in the type. By
contrast, historians are interested in particular token events like the
French Revolution rather than revolution as a type (Tucker 2012; cf.
 Historical Truth 239

Cleland 2002). Kosso acknowledges as much but retorts that the natu-
ral sciences as well deal with unique events such as the Big Bang. The
Big Bang is a token event that for all that we know is the only token of
its type. I do not interpret the uniqueness of the Big Bang as a reductio
ad absurdum of the attempt to distinguish the historical from the the-
oretical sciences but bite the bullet to argue that cosmology is a histori-
cal science.

III

Collingwood (1956), Murphey (1973, 1994, 2009), and Goldstein (1976,

1996) noted the obvious: Historiography makes no observations of his-
torical events but presents descriptions or representations or construc-
tions of such events in the presence of evidence. There are no given
objectively true historical facts: Historiography “is a science whose
business is to study events not accessible to our observation, and to
study these events inferentially, arguing to them from something else
which is accessible to our observation, and which the historian calls
‘evidence’ for the events in which he is interested” (Collingwood 1956,
251– 52). The immediate, primary, subject matter of historiography is
evidence and not events.
Murphey (1973, 16) likened George Washington to the electron:
“an entity postulated for the purpose of giving coherence to our pres-
ent experience. . . . [E]ach is unobservable by us.” This theoretical
concept of historiography considers historiographic truth a type of
theoretical truth: “Historical Knowledge is a theoretical construct to
account for presently observable data” (Murphey 2009, 6). If theories
model aspects of nature, for example, the model of the atom in theoreti-
cal physics or the double helix model of DNA in biology, historio-
graphic theories may model the past, for example, the model of George
Washington’s presidency. History is as invisible as scientific theoretical
entities. Historiography can be as true as scientific theories. Murphey
(2009, 11) proposed that the veracity of historiography can be evaluated
using the criteria for choosing between theories: whether the theory
is best at explaining all the relevant data about its subject, consistent
 240 Aviez er Tucker

with related theories, and explains new data as they are added, and, I
may add, directs historians to discover new evidence. Murphey praises
this kind of realism, which he calls constructivism, for articulating a
concept of historical truth without having to resort to correspondence
theories of truth.
However, there is a basic difference between the electron and feu-
dal society on the one hand and George Washington and the formation
of our solar system on the other. The electron and feudal society, as
theoretical entities, are types; they have no space and time, and they do
not require any particular set of token pieces of evidence to confirm
them, as different laboratories can conjure different tokens of the same
types of experiments to confirm them. By contrast, George Washing-
ton and the formation of the solar system are token events. They oc-
curred once at particular spatiotemporal locations, and they had never
happened before and will never happen again. George Washington and
the formation of the solar system are not theoretical entities. This is the
foundation for the difference between what I called the theoretical and
historical sciences (Tucker 2011, 2012) and Cleland (2002, 2009, 2011)
and Turner (2007) called the experimental and historical sciences.4 Irre-
spective of labels, some sciences attempt to study types and others
study tokens, and those two disparate goals dictate entirely different
methodologies for discovering different types of truths.
Historiography is theoretical in the sense of using theories to infer
representations of token historical events like George Washington’s
life and times from evidence in the present. These theories are typi-
cally information theories about the transmission of information in
time via various media like documents, languages, DNA, artifacts, light
from distant galaxies, and so on. These theories assist in tracing back
the information signals we receive in the present from the past. His-
toriographic representations are theory laden in the sense that they are
founded on information transmission theories. But the representations
are expressed using tokens. Even when the representations of the birth
of the solar system are influenced by astronomical theory and the rep-
resentations of historical societies are influenced by social theories, they
are still representations of tokens and not types. Their relation to the
evidence is different from that of theories in the theoretical sciences
 Historical Truth 241

which are not interested in tokens, in this or that particular experi-

ment that can go wrong, but in types and their relations that compose
theories.
Murphey (2009, 7) defended his application of “theory” to histo-
riography, relying on a quotation from Quine who described theories
as “the set of sentences about that thing one holds to be true.”5 Histo-
riography shares with scientific theories the postulating of objects and
events that are not observable and must be confirmed or inferred.
Highly confirmed historical events for which there is overdetermining
evidence are at least as well confirmed as the best scientific theories.
The weakness I find in Murphey’s account is similar to the one I
found in Kosso’s use of “observation” above. He broadened the sense
of the concept so much that it lost its potency (conceptual deflation by
hyperinflation). A robust and philosophically useful sense of scientific
theories must grant them broad scope and consilience: they must be
applicable to many different kinds of cases. They must be confirmed
by replication, if not of experiments, then at least of observations of dif-
ferent tokens of interesting types. To do that, theories must be com-
posed of types. By contrast, historiography and the historical sciences
are about token events that happened once, cannot be replicated, and
can never be observed. Historians utilize scientific theories about the
transmission of information in time. But what they infer from the token
evidence in the present and these information theories are typically rep-
resentations of events that are described using tokens and not types.
The French Revolution is a token of a revolution, but general theories
of revolutions do not infer it, nor can it refute them (Tucker 2012). It is
inferred from countless documents and physical remains through the
use of information theories about where to look for the evidence and
how to analyze it. Likewise, phylogeny infers a token common ances-
tor of species, who lived at a particular place at a particular time.
Murphey (2009, 151) claimed that replication is as possible in his-
toriography as in science. Different historians can reinspect the same
document or material remains. Murphey explained that historians do
not replicate such observations because journals would not publish
accounts of visits to archives that confirm the previous evidential ob-
servations of other historians. However, Murphey failed to notice that
 242 Aviez er Tucker

replication is boring in the historical sciences because it is of the same

tokens, while it is interesting in the theoretical sciences because it is of
different tokens of the same types, necessary for confirming theories
about types. In the historical sciences replication of observation of
token evidence is interesting only if it challenges previous such obser-
vations, not if it agrees with them. By contrast, in the theoretical sci-
ences, replications that agree are interesting in what they tell us about
their types, while replications that fail can be discarded, explained away
as “something went wrong with the experiment,” unless replicated sev-
eral times.

IV

David Carr (1986) considered historiography to have a narrative form.

Yet Carr also thought that it is a true story because historiographic
narratives are isomorphic to an independently existing historical nar-
rative. He claimed that time and collective action, as they are experi-
enced, have a narrative structure. We have an experience in common
when we grasp a sequence of events as a temporal configuration. Its
present phase derives its significance from its relation to a common past
and future (Carr 1986, 127). Each community has a narrative that con-
stitutes it. This narrative is the unity of story, storyteller, audience, and
protagonist. Storytellers tell stories whose protagonist is the com-
munal we of their audience, the community. A second-order, historio-
graphic, narrative is told by historians. It might have, for cognitive or
aesthetic reasons, a different content from a first-order narrative. But
if both narratives share an identical subject, the narrative forms remain
isomorphic, according to Carr.
Instead of correspondence, some kind of isomorphic communal
self-consciousness becomes the essence of historical truth in Carr’s nar-
rative realism. But historians do not wish always to talk about com-
munal subjects that were present in the consciousness of a historical
community, nor do they always participate in a kind of collective en-
tity where they can benefit from historical self-consciousness. Social
historians may write legitimately about social categories whether or not
 Historical Truth 243

the historical subjects of these categories were conscious of them. Carr’s

model of historiography does not fit most of historiography and all
the other historical sciences that do not deal with the phenomenology
of consciousness of historical communities through time. There can
be and there are mutually inconsistent competing isomorphic second-
order historiographic narratives that tell different stories that share their
protagonists and plot structure. They cannot all be true.

A combination of realist empiricist elements and antirealist elements

can be found in what I call “strudel and apples” philosophies of histo-
riography. It is obvious that in order to bake a strudel pie we need ap-
ples. Some workers have to go to the apple orchard to pick the apples
off the trees. They must choose which apples are ripe enough and
which are rotten and should be discarded. The apple pickers are neces-
sary, and their work is important. But they are not experts or geniuses.
Their work does not require much expertise or experience. However,
once the apples are brought to the kitchen, the genius of the chef is
displayed in how she works with those apples, how she slices or cooks
them, which ingredients are added (cinnamon, raisins, etc.), and how
she makes the dough and bakes the strudel. If observers wish to un-
derstand why a particular strudel is good or bad or excellent, they
usually study what happens in the kitchen and not what goes on in the
orchard. Good apples are all the same, but each good strudel is differ-
ent. Likewise, arguably, when historians walk into the orchard of the
archive, they pick and choose the ripe “facts” and discard the forged
or otherwise unreliable evidence. The facts are collected in a basket and
brought to the kitchen of the historian, where the master historian con-
ceptualizes, slices, and combines them to form explanations and pose
causal links as well as structure them in the form of a narrative. Ar-
guably, if philosophers of historiography are interested in understand-
ing historiography they should not pay too much attention to the nec-
essary but boring collection of facts but study the construction of
concepts, explanations, and narratives. Historical facts then are true
 244 Aviez er Tucker

but boring. The historiographic narrative is interesting, and contains

true ingredients, but on the whole is not true or false.
Hayden White (1987, 1– 25) distinguished three levels of histori-
ography, annals, chronicles, and narratives, that roughly correspond
with the questions of what happened, why, and what it means. Each
level, claimed White, requires extra conceptualization. According to
White, the conceptualization comes from the narrative forms that are
innate rather than historical, in the mind or in language rather than in
the past. White suggested that the evidence-based chronicle can be
true or false but not the narrative. Ankersmit (2001) as well considered
historiography a combination of the scientific with the aesthetic. His-
torical truth makes historiographic narratives merely adequate, and
can be shared by multiple narratives. The aesthetic dimension distin-
guishes historiographic narratives from one another and is the more
interesting subject matter for the philosophy of historiography.
However, there are no ripe and ready “facts” in historiography
that the historian can pick as an apple picker can pick ripe fruits off
a tree. The archive does not resemble an orchard. It is not made of
distinct atomic units that need to be selected and then put together,
sliced, and cooked in the historian’s narrative workshop. Historio-
graphic “facts” are postulates that explain present evidence (Murphey
2009, 6). Since it is trivially true that the present is the effect of the
past, the historian requires information theories to identify which
present effects are likely to preserve which types of information about
the past.
Narrative historiography should be compared with popular science
or scientific textbooks, not with scientific research. As Kuhn noted, if
we want to understand science, we need to look at what scientists have
been doing, at the history and sociology of science, rather than at the
textbooks that scientists write about their practices and the histories
of their disciplines, which may resemble fairy tales. It is hardly sur-
prising or even very interesting that historians with different interests,
values, and group identities write different historiographic narratives.
It is much more interesting and surprising that they are able to agree
on so much about history and not just about “atomic facts” despite all
their differences.
 Historical Truth 245

VI

The most famous antirealist philosopher of historiography is probably

Michael Dummett, who argued (Dummett 1978) that sentences about
the past are not assertoric, they do not assert anything, they are nei-
ther true nor false, because there are no clear truth conditions such
as observations that would allow or disallow us to assert them. He
claimed that since the past is inaccessible directly, all we have is what
historians tell us; further ontological assumptions about the past are
not warranted.
Dummett exaggerated the veridical difference between direct ob-
servation and what we infer from evidence that can be observed in the
present. What we see with our own eyes can be less reliable than what
we infer from multiple independent testimonies. The states of affairs
that make historiographic propositions true do not have to correspond
with them. They can be theory-laden descriptions of evidence, most
notably multiple units of evidence such as testimonies, genomes, or
languages, that are independent of each other and preserve the same in-
formation from their common sources or origins. Dummett seems to
have reached his conclusion in a state of innocence of any of the his-
torical sciences, their methodologies or practices.
Years later Dummett (2004) declared his earlier antirealist conclu-
sions about historical truth “repugnant” and made the opposite argu-
ment, for the truth of our knowledge of the past. However, through
this reversal, Dummett has maintained his innocence of the historical
sciences and their philosophy. He seems to have had no idea how his-
torians reach consensus on their conclusions and why so many people
consider them true. Dummett’s philosophy was “radically naive and
inappropriate . . . for statements about the past, and especially ones at
any significant historical remove. Goldstein’s [1976] discussion has then
the virtue of highlighting what sorts of inferential practices actually
come into play in constituting the past. Better actual practices than
philosophical fictions to the same effect” (Roth 2012, 322).6
Instead of examining historiographic inferences, for example, from
multiple independent information-preserving testimonies and other
units of evidence, Dummett maintained what he called a justificationist
 246 Aviez er Tucker

view that bases truth on a subjunctive: Historical truth is what some-

one present in the past would have observed. This justificationism is
too narrow and too broad. Too narrow, because nobody would have
observed the Industrial Revolution or the Renaissance. Contempo-
raries observed machines and paintings. Ideational changes within the
minds of historical people such as humanism during the Renaissance
could not have been observed while they happened. Events and pro-
cesses are not necessarily concrete enough to be observed by a single
observer in a limited time. As Roth (2012, 322) noted, most of what
historians write about, including what Danto (1985) called narrative
sentences that refer to two distinct times (e.g., “the First World War
began with the assassination of Archduke Ferdinand in 1914”), was
not and would not have been observable by contemporaries. Justifica-
tionism is too broad, because some subjunctives are difficult to sub-
stantiate and so are epistemically useless. Dummett’s account misses a
discussion of historical subjunctives, what observers would have seen
in history had so-and-so been true, and, most significantly, how his-
torians can know whether these observers would have seen what is
imputed to them. Can a sentence about the conditions in the early
universe that is not evidence based be true if somebody could have ob-
served it under conditions where no human could have survived? What
should be assumed about the information-extracting technologies of
the hypothetical observer? If we allow any technology however science-
fictional, then the whole procedure is redundant because we can apply
the counterfactual to the present and talk about an ideal observer in the
present with superior technology to extract much more information
from present traces than is possible today. If not, then what level of
technology is relevant? It seems that the later Dummett, just like his
earlier self, did not understand what goes into justifying assertions
about the past.
Dummett did not distinguish the probabilities of what multiple in-
dependent witnesses testify to from those of single observations. Ob-
servation in the past was for him like an observation in the present. It
could be transmitted by memory, eyewitness reports, or through a
chain of testimonies. Dummett (2004, 68) writes, “We are in as good a
position to know by testimony that the past event was observed as if
 Historical Truth 247

we were told by a living observer. Dying does not deprive anyone of

the status either of an observer or an informant: the dead remain mem-
bers of the community— the community I have been referring to as
“we”—with whose collective and imperfectly shared knowledge I have
been concerned.” Murphey (2009, 16–17) chides Dummett for ignor-
ing that observations by others are testimonies and their evaluation re-
quires reference to reliability and competence. Murphey also noted cor-
rectly that the long dead are not members of our community and their
testimonies are not the mythical observation sentences of the posi-
tivists; they require interpretation (17–18). This requires historical pro-
fessional expertise. Historians, as Murphey noted, do not consider the
testimonies they use to infer from historical truth observations. Testi-
monies are often mutually contradictory and false. Historians look for
the best explanations of testimonies irrespective of the truth of their
propositional content.
Dummett (2004, 44) argued in favor of his subjunctive justifica-
tionism: “If the account of meaning demanded that we allow as true
only those statements about the past supported by present memories
and present evidence, then large tracts of the past would continually
vanish as all traces of them dissipate.” True, information decays. In
earlier stages in the history of the universe more of the universe would
have been visible and information decayed about vast stretches in the
history of the universe, the planet, life, and humanity. But those large
stretches are indeed beyond the scope of historical knowledge. The op-
posite tendency of recovering nested information about large stretches
of the past is the result of advances in historical science. True, memo-
ries fade, but little of our knowledge of the past is based on them. The
other evidential sources of information have actually multiplied over
the past couple of centuries with new methods for extracting infor-
mation from existing evidence. We know many more truths about the
past today than we did two centuries ago. So, while the tendency in cos-
mic time is for information to decay, the tendency in the much shorter
modern era of historical science has been to discover more of the past
by developing new methods for extracting existing nested information
about the history of biological evolution from fossils and genomes, of
human prehistory from genomes and languages, and of human history
 248 Aviezer Tucker

from documents. The significant result is that what we take to be true

about the past is the balance between these two tendencies for infor-
mation decay and scientific extraction (cf. Cleland 2002, 2011; Turner
2007; Tucker 2011).
Dummett concluded, “We should be committed to a metaphysical
conception according to which nothing exists but the present: the past
would be a mere construct out of whatever in the present we treat as
being traces of it. We could not so much as think of a statement about
the past as having once been true, though now devoid of truth-value,
save in terms of present evidence that evidence for its truth once ex-
isted. This conception, though not incoherent, is repugnant: we can-
not lightly shake off the conviction that what makes a statement about
the past true, if it is true, is independent of whether there is now any
ground that we have or could discover for asserting it” (2004, 44). Con-
structing the past “out of whatever in the present we treat as being
traces of it” is what the historical sciences have been doing to great
effect. The historical sciences have been called many things over the
history of philosophy; “repugnant” is new, even if Dummett did not
realize that he was addressing them rather than a metaphysics he dis-
approved of. Sentences about the past that have no evidence for or
against can still be meaningful, but then Dummett would have had to
have a different theory of meaning with a weaker or no connection be-
tween meaning and truth, and that is probably what he found so repug-
nant. The very claim of the historical sciences for truth, for generating
true propositions about the past, is evidence based. Without evidence
there can be no truth in the historical sciences.
In his discussion of the reality of the past and the future, Dum-
mett noted the basic metaphysical asymmetry between the past and
the future: “We assign to the past those events capable of having a causal
influence upon events near us, so that we can receive information from
them and of them, but have no means of affecting them; and we assign
to the future those events that we can affect, but from which we can
receive no information” (2004, 86). However, he did not draw the ob-
vious conclusion from this basic metaphysical insight, namely, that his-
torical truth is founded on the information we receive from the past,
not on observations of the past, real, or subjunctive.
 Historical Truth 249

McCullagh (1998) also attempted to explicate a concept of “truth”

appropriate for historiography. He elaborated a “correlation” theory
of truth to replace the untenable correspondence relation between his-
toriography and history. Historiographic descriptions can be true in
his opinion in the sense that a description of the world is true if it is
part of a coherent account of the world, and if the observation state-
ments implied by that account would have been confirmed by people of
the appropriate culture and with the appropriate interests. This shows
how the truth of a description depends upon the way the world is while
allowing that it also depends upon the rules of the language of the de-
scription, and upon the concepts of the world which those rules refer
to. McCullagh acknowledged that it is impossible to prove that histo-
riographic descriptions are true in this sense, but historians want to
get as near to the truth as they can. So, like Dummett, he explicated a
sense of truth that is applicable to history but is not useful epistemi-
cally to answer whether a proposition about the past is true or not.
The subjunctive about what contemporary observers would have ob-
served and conceptualized brings McCullagh’s correlation theory of
truth close to Dummett’s account. It is both too narrow, because in-
formation from the past can be preserved without contemporary ob-
servations, and too broad, because in many cases there is insufficient
evidence to confirm the subjunctive.

VII

Murphey (2009, 12–13) suggested that historical skepticism is motivated

by naive realism. If philosophers of historiography consider correspon-
dence to be the only theory of truth, they must conclude with the early
Dummett and the postmodernists that there is no historiographic truth;
they must retreat to what Murphey called “some form of linguistic
idealism according to which all that we know we can know is our own
language” (12). The main challenge to historiographic antirealism is to
explain a series of “miracles.” First, the miracle of consensus: If histo-
riography is not true in some sense, what can explain the broad agree-
ment among historians about historiography? There are two candidate
explanations, politics and methodology.
 250 Aviezer Tucker

The political explanation claims that the consensus among histori-

ans can be explained by factors external to historiography such as poli-
tics, economic interests, social background, and so on. This externalist
explanation is less convincing in the history and philosophy of histori-
ography than in the philosophy of science because of the greater het-
erogeneity of historians in comparison to scientists. The economic and
social barriers to entrance to the historical profession are lower than
those for entering science. Being a historian does not require long and
expensive education and access to laboratories and specialized instru-
ments. Consequently, the backgrounds of historians are more diverse
and heterogeneous than those of scientists. Indeed, many historians
belong to opposing political, social, and religious camps. It is difficult
to imagine how these diverse and even contentious backgrounds can
cause so much agreement among historians, unless it reflects common
knowledge of historical truth!
Alternatively, internal factors can account for the consensus, yet
not imply truth. Leon Goldstein (1976, 1996) suggested that historiog-
raphy has established a consistent family of methods for the interpre-
tation of evidence. Consistent application of these methods yields
determined interpretation of evidence according to strict professional
norms. A methodological consensus, however, does not imply the truth
of their results. Goldstein denies that determined historiography is a
true representation of the past or anything more than the most plausible
interpretation of the evidence. This antirealism is ontologically more
parsimonious than historical realism; it makes fewer assumptions. Yet
it pushes the miraculous a step further. It eliminates the antirealist mys-
tery of consensus on historiography at the cost of a new miraculous
mystery, a concensus on methodology that is not necessarily truth con-
ducive: Why has a uniquely heterogeneous and uncoerced large group
of historians come to agree on the information theories and methods
that define the historiographic community since the emergence of the
Rankean paradigm at the beginning of the nineteenth century? Gold-
stein failed to offer a convincing explanation of the historiographic
consensus on theories and methods.
If historiography is the best explanation of the historical evidence,
it does not imply that it is a representation of the past; it could be
 Historical Truth 251

wrong, it is fallible. Yet the theories and methods of historiography are

shared by a large heterogeneous and uncoerced community of histori-
ans, as well as by biblical and classical scholars, philologists, detectives,
and judges. When historians reach an uncoerced, uniquely heteroge-
neous and large consensus on historiography, the best explanation is
that they possess knowledge of the past (Tucker 2004, 25– 45).
The case for the truth of historiography and the historical sciences
is stronger than that for the theoretical sciences like physics. One of
the strongest arguments against the truth of scientific theories is the
pessimistic induction: The refutation of many of the scientific theo-
ries that dominated science up to a century or so ago may imply that all
scientific theories that have not been proven false yet will be proven
false in the future (Laudan 1981). Even apparently highly well con-
firmed theories like Newtonian physics were proven wrong. Scientific
paradigms succeed each other, and whether or not this succession is
progressive in some sense, it may not indicate increasing approxima-
tion to the truth. Realist attempts to salvage some truth-value in scien-
tific theories that were proven unsuccessful have concentrated on find-
ing elements that are preserved after the collapse of a paradigm and the
refutation of its theories and are salvaged and carried over for the next
paradigm and its theories (Park 2011). In the historical sciences, by
contrast, there is every reason to believe in an optimistic induction,
that much of our present knowledge and theories about the past will
continue to withstand the test of time. All the historical sciences have
had one and only one founding revolution that transformed them into
sciences, usually through the introduction of new theories about the
transmission of information in time. The documentary hypothesis in
biblical studies, the Indo-European hypothesis in historical compara-
tive linguistics, Rankean archival-based scientific historiography, Lyell’s
paradigm in geology, Darwinian biology and the hypotheses about the
origins of species, genetic phylogeny, and so on, have not been proven
false, nor have we had to revise any of their main conclusions or as-
sumptions. To be sure, they have expanded. We know today much more
about the history of life than Darwin did or about social and cultural
history than Ranke did. But the founding texts, assumptions, and in-
formation theory–based methodologies have preserved their truth for
 252 Aviezer Tucker

two hundred years. It seems hard to believe that we will come to reject
one day the hypothesis that men and apes have a common origin or
the theory of evolution upon which it is based, or that the author of the
Iliad could not be the author of the Odyssey or the philological theory
upon which it is based. A positive induction may at the very least argue
for a probabilistic gap between the historical and theoretical sciences
that makes the first more likely to be true.

V I II

The common scientific theoretical background that allows historians

to reach uncoerced consensus concerns the transmission of informa-
tion in time, not the evolution of society. This point has often been
confused because in textual criticism, comparative linguistics, and evo-
lutionary biology the evolution of the studied system, texts, languages,
and species is the transmission of information in time and its selection,
whereas in historiography and archaeology the evolution of society
and the transmission of information from and about past events are in-
dependent of each other. The theoretical background allows scientists
who examine possible common origins to prove first that the similarity
between effects is more likely given a common information source
than given separate sources and then to prove which origin hypothesis
increases the likelihood of the evidence more than others. These hy-
potheses put together compose our probable knowledge of the past,
historical truth. Such hypotheses may have all levels of generality or
concreteness; they may be descriptive or explanatory.
Our knowledge of history is limited by the information-preserving
evidence that survived the ravages of time in the historical process that
connects history with historiography. Much of historical truth is over-
determined: Much of the evidence we possess at the moment for World
War II or for the genetic history of mammals or for the common ori-
gins of Czech and Polish is redundant. Historians could discover the
same historical truths even if much of the evidence disappeared. How-
ever, in other cases, the evidence provides weaker support for historio-
graphic hypotheses or even underdetermines it. For example, Murphey
 Historical Truth 253

(2009) examined the hypothesis that the Gospel of John was written in
response to the gnostic Gospel of Thomas. This hypothesis is coherent
with everything else that is accepted as historically true about the his-
tory of Christianity. The linguistic and thematic-theological similarities
between the two texts are explained by the writing of John as a response
to the earlier contemporary appearance and circulation of Thomas.
However, the probability of this hypothesis is clearly lower than that
of the Anti-Federalist Papers being written as a response to the Fed-
eralist Papers, or of Marx’s theses on Feuerbach being written as a re-
sponse to Feuerbach and the young Hegelians.
Murphey analyzed this lower probability in terms of the absence
of direct evidence. Nowhere in John is there any direct reference to
Thomas, nor is there a third contemporary source that attests to the
influence of Thomas on John. However, there is no such direct evidence
for the relation between Sanskrit and Greek or for the importation of
the potato to Polynesia from South America prior to European ex-
ploration. Yet these hypotheses have considerably higher probability;
they are historically true. The reason is that there is no probable sepa-
rate sources hypothesis that can possibly explain the similarities be-
tween Sanskrit and Greek and the genetic identity between the South
American and Polynesian potatoes. Sanskrit and Greek must have had
a common origin, and the originally Andean potato must have some-
how been transported to Polynesia. By contrast, there is some small
probability that the themes and ideas of John could have been devel-
oped independently of Thomas, given the various theological and doc-
trinal debates of early Christianity.
The question is whether the probability of the hypothesis about
the relation of John to Thomas is sufficiently high to be considered his-
torical truth. The answer in my opinion depends on the context of in-
quiry. Epistemic contextualism suggests that what we consider truth, the
determination of truth-value, depends on context. For example (DeRose
2009), it depends on context whether we consider true a proposition
that a certain bank branch is open on Saturday because somebody
remembers so. If the stakes are low, we consider it true. If the stakes
are high, if going to the bank on Saturday rather than Friday to deposit
a check may result in missing a payment on a house and initiating
 254 Aviez er Tucker

foreclosure, memory is an insufficient foundation for truth. If much is

at stake, we apply a more exacting standard of truth, and we would not
say we know the bank is open on Saturday until we call there or check
the web page of the bank. In common law, there are different standards
of evidence for criminal law (guilt beyond reasonable doubt), civil law
(the preponderance of evidence), and licensing cases (presumption of
guilt). The standards of evidence correlate with the severity of the con-
sequence to the accused. In criminal cases that may result in incarcera-
tion or worse, it is very high. In civil cases, where the worst outcome
for the accused is substantial financial loss, it is lower. In cases of licens-
ing, where the worse outcome is a moderate fine, the standards of proof
are low (Ho 2008). For example, in the recent “celebrity” murder cases
of O. J. Simpson and Robert Blake, the defendants were acquitted of
killing their wives in a criminal court but were held responsible for the
deaths in civil courts. Without epistemic contextualism, it would seem
that they were at once guilty and not guilty of killing their wives. How-
ever, the juries found the probability of their committing these crimes
lower than beyond reasonable doubt but sufficiently high to be justified
by the preponderance of the evidence. My impression is that in most
historiographic cases, the standard of proof is comparable to that in civil
cases, following the preponderance of evidence. But when the effect of
the historiographic account may be sufficiently significant in the pres-
ent, it may require a more stringent criterion of truth, comparable to
“beyond reasonable doubt.” When living people are affected by what is
accepted as historical truth, the stakes and standard are higher. It is one
thing to infer from a document that somebody was a spy for the Habs-
burg monarchy in the nineteenth century; it is quite another to prove
that a living person was an informer for the Communist regime a few
decades ago on a comparable evidential basis. In the case of the influ-
ence of Thomas on John above, the textual evidence may suffice for a
person who considers John a text from the Roman era. But a person
for whom the correct interpretation of John may be of utmost theologi-
cal significance may find the probability of the influence of Thomas on
John insufficient for considering it truthful.7
In some cases there is insufficient evidence for preferring one his-
toriographic hypothesis as more probable than all others. To achieve
 Historical Truth 255

a preponderance of evidence, one hypothesis has to be clearly more

probable than any of its alternatives. But when there is a limited body
of evidence, there may be multiple possible hypotheses. For example,
the Neolithic figurines of buxom women that are found all over Eu-
rope may be images of the political rulers of matriarchal societies, or
fertility goddesses, or Stone Age pornography. Hitler surely had some
kind of mental illness or personality disorder, but was he a psycho-
path, a hysteric, a bipolar, or a combination of the above? Such hy-
potheses are underdetermined by the evidence. The social manifesta-
tion of underdetermined historiography is the absence of the kind of
uncoerced uniquely heterogeneous consensus we find in determined
parts of historiography (Tucker 2004, 140 – 84). Underdetermined his-
toriography does not justify a skeptical interpretation that “anything
goes” in historiography and the belief that historiography has the same
veridical status as fiction. The evidence that underdetermines some
hypotheses in some parts of historiography may still suffice to re-
fute most hypotheses, leaving only a handful of alternatives. Still, mul-
tiple similarly probable, yet mutually inconsistent historiographic hy-
potheses cannot receive a realist interpretation, cannot claim to be true
at the same time.

Not e s

1. Even worse than rationalization is resistance, fighting off any discus-

sion of historiographic epistemology and methodology by using academic po-
litical and economic powers. Methodological angst has led to the elimination
of classes in historiographic methodology, theory, and philosophy and the
exclusion of theoreticians or philosophers and other metalevel thinkers about
history from academic employment if they are young, and their marginaliza-
tion if they are already part of the profession, while pressuring them to avoid
raising foundational questions in public. History departments rarely offer post-
graduate seminars in theory. They teach methodology tacitly through practice.
This may explain a self-reinforcing mechanism by which trained historians are
not conscious of the reliable processes of reasoning and belief formation about
the past that they actually wield as they learn unreflectively tacit methodology
through practice. Consequently, they are anxious about its reliability and resist
 256 Aviezer Tucker

discussions of the epistemology of our knowledge of the past. They then edu-
cate another generation tacitly, and so on.
2. Unless we assume some sort of idealist ontology and believe we can
somehow commune with the ideas of dead historical figures and then see if
they correspond with their historiographic representations (see Murphey
2009, 5).
3. The archival contexts of documents are significant, but they are no-
where as complex and heterogeneous as those of human action. The possible
variations and forms of explanations of documents are far less numerous than
those of human action. Consequently, making explicit the tacit assumptions
and processes historians use to explain their evidence is much easier than expli-
cating human action (cf. Collins and Kusch 1999). For example, the best expla-
nation for the independent diaries of soldiers that state that on a certain date
their unit came under heavy bombardment and therefore panicked and re-
treated is that indeed they came under heavy bombardment and therefore pan-
icked and retreated. There is no need for further knowledge of the psychology
of human action under fire. Had this explanation of the retreat depended on
psychological theories, it would have been indeterminate, since under fire sol-
diers are known to retreat out of fear, become paralyzed with fear and stay put,
or become emboldened with rage and charge on, and these are the exhaustive
and mutually exclusive three possible responses. The mere description of the
stimulus together with background conditions and contemporary psychology
would be insufficient for explaining the actual retreat. But psychology is redun-
dant here. The preservation of information on panic and retreat is the best ex-
planation of the independent evidence for it. The relevant background theory is
not psychological but informational: the reliability of independent witnesses
who witnessed and participated in the events and wrote down their experiences
immediately after the events took place for no external purpose, in diaries.
4. For my reasons for preferring the terminology of theoretical to ex-
perimental sciences, see Tucker 2009.
5. Murphey (2009, 75–102) distinguished four types of historiographic
theories: narratives, hypotheses for which there is indirect evidence, social sci-
ence theories about the past like Turner’s Frontier Theory, and philosophies
of history (which some call substantial or speculative).
Narratives and theories can, but do not have to, coincide. Hypotheses
about the past are indeed part of historiography, but they use tokens rather
than types. Social science theories are theories but are parts of the social sci-
ences and not historiography (Tucker 2012). Philosophies of history are nei-
ther science not historiography (Tucker 2004, 14 –17).
6. Unfortunately, Roth (2012) did not practice what he preached in his
own “irrealist” account, an excessively mechanical application of Ian Hack-
 Historical Truth 257

ing’s theory of historical ontology to the philosophy of historiography. In-

deed, historiographic conceptualizations of the past are fluid. But to under-
stand the conceptual choices historians and their communities have made, it is
necessary to study them, e.g., how and why concepts like the Renaissance or
the Industrial Revolution became accepted. To do that, Roth would have had
to conduct research in the history of historiography. He would have found
then the extent to which historiographic conceptual choices depended on in-
formation theories that connected a broad scope of evidence to new fruitful
concepts.
7. Obviously, the discovery of more evidence would be the elegant way
to decide whether the influence of Thomas on John is true or not, even ac-
cording to an exacting “beyond reasonable doubt” criterion. Unfortunately,
in many cases history has not preserved information about the past, and histo-
rians must make do with a fixed and limited scope of evidence. In such cases
standards of truth that vary with context can explain what historians accept
and reject as true and why. In the context of debates about epistemic con-
textualism, this explains why epistemic contextualism cannot be reduced to
certain epistemic attitudes to evidence. Unlike Nagel’s (2012) claim, some con-
texts where the stakes are high cannot be transformed into attitudes that re-
quire action for pursuing more evidence. Sometimes this is impossible and the
truth of propositions depends on their contexts.

R e f ere nce s

Ankersmit, Frank. 2001. Historical Representation. Palo Alto, CA: Stanford

University Press.
Bevir, Mark. 1999. The Logic of the History of Ideas. Cambridge: Cambridge
University Press.
Carr, David. 1986. Time, Narrative, and History. Bloomington: Indiana Uni-
versity Press.
Cleland, Carol E. 2002. “Methodological and Epistemic Differences between
Historical Science and Experimental Science.” Philosophy of Science 69:
474– 96.
———. 2009. “Philosophical Issues in Natural History and Its Historiogra-
phy.” In A Companion to the Philosophy of History and Historiography,
ed. A. Tucker, 44– 62. Malden, MA: Wiley-Blackwell,
———. 2011. “Prediction and Explanation in Historical Natural Science.”
British Journal for the Philosophy of Science 62: 551– 82.
Collingwood, R. G. 1956. The Idea of History. Oxford: Oxford University
Press.
 258 Aviezer Tucker

Collins, H. M. 2010. Tacit and Explicit Knowledge. Chicago: University of

Chicago Press.
Collins, Harry, and Martin Kusch. 1999. The Shape of Actions: What Humans
and Machines Can Do. Cambridge, MA: MIT Press.
Danto, Arthur. 1985. Narration and Knowledge (including the integral text of
Analytical Philosophy of History). New York: Columbia University Press.
DeRose, Keith. 2009. The Case for Contextualism: Knowledge, Skepticism,
and Context. Vol. 1. Oxford: Oxford University Press.
Dretske, Fred I. 1969. Seeing and Knowing. Chicago: University of Chicago
Press.
———. 1981. Knowledge and the Flow of Information. Cambridge, MA: MIT
Press.
Dummett, Michael. 1978. Truth and Other Enigmas. Cambridge, MA: Har-
vard University Press.
———. 2004. Truth and the Past. New York: Columbia University Press.
Goldstein, Leon J. 1976. Historical Knowing. Austin: University of Texas Press.
———. 1996. The What and the Why of History: Philosophical Essays. Leiden:
Brill.
Ho, H. L. 2008. The Philosophy of Evidence Law: Justice in the Search for
Truth. Oxford: Oxford University Press.
Kosso, Peter. 2001. Knowing the Past: Philosophical Issues of History and Ar-
chaeology. Amherst, NY: Humanity Books.
Laudan, Larry. 1981. “A Confutation of Convergent Realism.” Philosophy of
Science 49: 19– 49.
Lewis, David. 1996. “Elusive Knowledge.” Australasian Journal of Philosophy
74 (4): 549– 67.
McCullagh, C. Behan. 1998. The Truth of History. London: Routledge.
Murphey, Murray G. 1973. Our Knowledge of the Historical Past. Indianapo-
lis: Bobbs-Merrill.
———. 1994. Philosophical Foundations of Historical Knowledge. Albany:
State University of New York Press.
———. 2009. Truth and History. Albany: State University of New York Press.
Nagel, Jennifer. 2012. “The Attitude of Knowledge.” Philosophy and Phe-
nomenological Research 84: 678– 85.
Park, Seungbae. 2011. “A Confutation of the Pessimistic Induction.” Journal
of General Philosophy of Science 42: 75– 84.
Roth, Paul A. 2012. “The Pasts.” History and Theory 51: 313– 39.
Tucker, Aviezer. 2004. Our Knowledge of the Past: A Philosophy of Historiog-
raphy. Cambridge: Cambridge University Press.
———. 2009. “The Philosophy of Natural History and Historiography, re-
view of Derek Turner, Making Prehistory: Historical Science and the Sci-
 Historical Truth 259

entific Realism Debate (Cambridge: Cambridge University Press, 2007),

and Rob Inkpen, Science, Philosophy and Physical Geography (London:
Routledge, 2005).” Journal of the Philosophy of History 3: 385– 94.
———. 2011. “Historical Science, Over- and Under-determined: A Study of
Darwin’s Inference of Origins.” British Journal for the Philosophy of Sci-
ence 62: 825– 49.
———. 2012. “Sciences of Tokens and Types: The Difference between His-
tory and the Social Sciences.” In The Oxford Handbook of Philosophy of
the Social Sciences, ed. Harold Kincaid, 274 – 97. Oxford: Oxford Univer-
sity Press.
Turner, Derek. 2007. Making Prehistory: Historical Science and the Scientific
Realism Debate. Cambridge: Cambridge University Press.
White, Hayden. 1987. The Content of the Form. Baltimore, MD: Johns Hop-
kins University Press.
 C h a p t e r T w e l v e

Truth and Unity in Chinese

Traditional Historiography

-
Nicola Di Co s m o

Let’s imagine Chinese historiography as an immensely long, apparently

seamless piece of cloth woven in broadly similar patterns but with shift-
ing shades and subtle differences, extending over twenty-six centuries.
If we take that bolt of cloth and unfurl it to its starting point, it would
take us to a long, frayed, multistrand beginning with weak and loosely
connected threads. No matter how tattered and frail, those strands
produced what is arguably the foundation of China as a unified cul-
tural entity. During the Han dynasty (221 bc–ad 220), scattered annal-
istic traditions were pulled together and woven into a much stronger
historical fabric by the founders of imperial historiography, Sima Qian
(ca. 145 bc– 86 bc) and Ban Gu (ad 32– 92), respectively, the authors of
the first comprehensive history of China, the Shiji (Memoirs of the
Grand Historian), and of the first dynastic history, the Hanshu (Nien-
hauser 2011; Durrant 2005).
The warp and weft of human events were placed on the historical
loom by Sima Qian and, ostensibly, his father, Sima Tan, with a clear
sense that what was conceived of as history resulted (almost alchemi-

260
 Truth and Unity in Chinese Traditional Historiography 261

cally) from the fusion of two types of knowledge. On one plane were
events that unfolded in diachronic sequence, and therefore could be ar-
ranged chronologically according to the annalistic model. On another
plane were thematic narratives, such as biographical accounts, treatises
on matters of concern to government and society (astronomy and eco-
nomics, for instance), stories about certain categories of people, essays
on foreign peoples and geography, or, in other words, valuable knowl-
edge that would guide, instruct, and admonish.
Underlying both categories— events and accounts—were philo-
sophical questions on the function of history. The inherited Confu-
cian tradition assumed that historical knowledge was to be selected on
the basis of moral standards and deployed in the form of exemplars
and parables for the edification of present and future generations. On
the other hand, annals and records were also a storehouse of accu-
mulated experience for the instruction of rulers and officials in the
management of public affairs (Balazs 1961). Sima Qian’s revolution-
ary construction of a unified frame of historical knowledge posed the
foundations for dynastic as well as comprehensive histories and estab-
lished the classic model of official and private historiography, repre-
senting a watershed between an ensemble of scattered and loose col-
lections of records that conveyed (in the general interpretation) mostly
moral judgments in the form of annalistic notations and historical
knowledge as an organized practice, intellectual endeavor, and cultural
patrimony. We could say, without fear of exaggeration, that the Shiji
is responsible for the creation of the largest repository of historical
knowledge in human civilization, to state a fact that was once defined,
with good reason, as a “truism” (Pulleyblank 1961, 135).
Historical documents of an annalistic nature had been produced
for centuries before the Han dynasty, but in order to preserve and cor-
rectly transmit both the information about the actual events they
contained and the lessons to be gleaned from it, the “facts” had to be
collected, organized, evaluated, and understood with regard to their
multiple implications for the explication of the causes and effects of
human actions and how these actions and events fit into a larger philo-
sophical matrix. Moreover, one of the most important innovations we
find in the Shiji, the monographic treatises (shu), as well as some single
 262 Nicola Di Co smo

or collective biographies (liezhuan), transcended the single event and

allowed history to become comprehensive of various types of knowl-
edge, from economic policy to astronomy, state ritual, geography, and
foreign peoples (Hulsewé 1961).
Shortly after its composition, the structure established by Sima
Qian was modified to fit the model of the dynastic history by Ban Gu,
and ever since scholars debated the nature of the correct matrix to
which historical knowledge should conform. These debates generated
not just a vast amount of historical literature but also, by the eighth cen-
tury, penetrating studies of historical criticism. At the same time, the
high political and intellectual stakes of historical production led to the
imposition of ideological orthodoxy and state censorship. But there was
always an internal dialectical relationship between historical ideals and
the methods of historical writing. Subordination of history to philoso-
phy can be seen in its most unalloyed form with the triumph of Neo-
confucian thought in the Song dynasty, while the philological turn of
the Qing period provided new avenues to expand the range and tools of
historical research. The vein of historical criticism that runs through the
full spectrum of Chinese traditional historiography indicates a preco-
cious and heightened awareness of what it meant to control the past.
The creation of the History Office under the Tang dynasty led to a
growing bureaucratization of historical research, and from the seventh
century ad on the state never relinquished its prerogative (or shirked its
duty) to produce the history of the previous dynasty. This became an
unchallengeable cultural and political responsibility fully intertwined
with the politics of legitimate succession and proper rule.
The dynastic history tradition continues to cast a powerful shadow
even on contemporary historiography, as can be seen in the government-
sponsored project to produce an official history of the Qing dynasty.
The historian Dai Yi, the doyen of Qing history, who was appointed
editor in chief of this massive project about ten years ago, recently said,
“It seems that the traditional quadruple structure [of the dynastic his-
tories], containing basic annals, biographies, tables, and treatises, should
not be followed to the letter, nor should it be disposed of in its en-
tirety. Our job is to decide what to discard and what to keep” (Dai, 11).
Surprisingly, the new dynastic history commissioned by the Chinese
 Truth and Unity in Chinese Traditional Historiography 263

state harkens back to a bygone model that seemed to have been com-
pletely condemned and buried after decades of Marxist historiography
(Weigelin-Schwiedrizik 2005). What appears to be an act of extreme
anachronism is witness of the continuity of a discussion about the
shape and content of history that has been an integral part of the Chi-
nese cultural tradition from the very beginning.
Seen from the point of view of “truth” and “unity,” historical
knowledge can arguably be structured as a series of dichotomies, or
binary combinations, that, through their internal dialectical dynamism,
contributed to the growth of historical ideals as well as to the “craft”
of the historical profession. The dichotomies that can be singled out,
without any ambition to be exhaustive, are as follows. First, we have
the tension between private and public historiography, which includes
a subset of other questions, in particular, the responsibility of the his-
torian to “tell truth to power” and the equally weighty responsibility
of each dynasty to compile a dynastic history of the previous dynasty.
A subset of this question, which has proven extremely important for
the development of historical criticism, is the opposition between the
slavishness of official historiography and the creativity of individual
works. A second binary set, which at times intersects the first, is about
the fundamental purpose of history, and in particular the tension be-
tween the moral and philosophical foundations of its enterprise and
the need to produce an objective record of the past. A third source
of tension, which emerges almost from the beginning of a reflexive and
self-conscious attitude toward historical writings, rests on the division
between the “dynastic” and the “comprehensive” models, which are
rooted in nearly opposite views of what historical knowledge should
accomplish. An additional contrast can be drawn in reference to the
internal organization of historical knowledge between the annalistic
(the barebones records) and the narrative elements. Finally, a long-
standing issue is the relationship between history as a self-standing
category of knowledge and the canonical Confucian Classics, of which
the two most salient aspects are the separation of history from the clas-
sics and the “historicist” turn in which attempts were made to reinter-
pret the Classics as products of a specific historical context, as pro-
pounded by the eminent Qing historian Zhang Xuecheng (see below).
 264 Nicola Di Co smo

Among the many complex aspects that characterize the evolu-

tion of historical writing in China, truth and unity can be taken as con-
tinuous and central concerns. However, they both need to be qualified.
If a notion of “truth” can be applied to philosophical discussions in
early China by looking for a specialized lexicon that may indicate se-
mantic analogues, we face difficulties discussing historical knowledge
that ought to be addressed in a less straightforward manner (Schmidt-
Glintzer 2005, 116). On the one hand, historians surely wondered about
the trustworthiness of their sources, and one way to address “truth” is
by asking how historians strove to provide a truthful representation of
facts and to assess the reliability of documents. At another level, his-
torians questioned their sources about the true meaning of an event,
often by seeking to establish correlations of a metaphysical nature or
by assigning moral values. The question of unity, on the other hand, is
somewhat less difficult to address, as it turns on how best to represent
historical change once China defined itself and was recognized as a
culturally (if not always politically and territorially) coherent entity.
Whether compressed in the dynastic model or stretched across eras,
the notion of change (represented most typically in a cyclical fashion)
is inextricably linked to China’s avowed historical unity.

Trut h

In history writing “truth” primarily concerns the reliability of sources,

that is, whether a certain piece of information is something that can be
taken as having really occurred in the manner described. It also refers
to the “true value” of a given piece of information, that is, the inter-
pretation of its meaning once inscribed in a ritual or moral context, in-
cluding its religious and political aspects. Truth and truth claims can
only be dated to a time in the development of Chinese historiography
when the activity of recording events and royal utterances was accom-
panied by a narrative exposition in which the author, whoever that
may have been (authorship is a contested question in early Chinese
texts, and the Shiji itself, while attributed to Sima Qian, was initiated
by his father, Sima Tan), specifically doubted the reliability of a specific
 Truth and Unity in Chinese Traditional Historiography 265

fact or judgment. Sima Qian famously dismissed several claims that he

found in ancient texts, such as the Shanhaijing, a literary compilation
that contains plenty of supernatural elements or fantastic stories, as
simply fanciful— as unreliable historical sources.
In preimperial times, the earliest instance of an inquisitive attitude
emerges in the wake of the commentarial tradition, in response to the
Springs and Autumns. In the Zuozhuan (Commentary of Master Zuo),
for instance, the annalistic structure incorporates an expanded narra-
tive description of events, cause-and-effect relations are established,
and moral lessons are made explicit with the inclusion of dialogues
and speeches. Most important, the historian’s own voice, in phrases
such as “the master says . . . ,” makes its appearance (Schaberg 2005).
The explicit addition of the author’s own opinion in works such as the
Shiji and the Hanshu thus functions as a rhetorical device charged with
the task of consciously separating “subjective” statements from the gen-
eral narrative, which then acquires, by contrast, the semblance of “ob-
jective” and unadulterated history. The notion that historical accounts
contain truthful statements in early Chinese historiography, meaning
that it is possible to identify passages that show greater or lesser fac-
tual adherence to actual events, has been challenged by interpretations
emphasizing how literary or stylistic aspects shape and inflect what-
ever sense of historicity a given account may carry (Li 2011, 431).
The question of “truth,” however, should not be addressed by
searching for a correspondence between a text and an actual historical
event—a volatile concept under any circumstances—but rather to see
whether “truth” or a notion akin to it, such as the reliability or accu-
racy of reports and documents, can be observed in the construction
of the narrative as part of the author’s declared or assumed intent.
Did ancient historians care about transmitting true or false records?
Scholars have assumed for a long time that in the early dawn of Chi-
nese historiography the term later translated as “historian” (shi) indi-
cated a class of court scribes required to record all the actions and ut-
terances of the kings.
Yet not a single annalistic compilation for a given state has been
preserved for either the Spring and Autumn or the Warring States pe-
riods, with the exception of the Chunqiu and its commentaries, none
 266 Nicola Di Co smo

of which were the product of court scribes, according to the tradition.

Given the number of states that came into existence over the course of
half a millennium, one would assume that records were produced in
large numbers, but to my knowledge no such compilations have been
recovered even though over several decades archaeological research
has unearthed a large variety of texts. It is indeed possible that each
state had its own genealogies and some form of recording, and an ar-
gument has been made that these compilations were “dictated” by the
political situation (Lewis 2011, 450). Without actual evidence, how-
ever, how they were produced and what they looked like remain open
questions, since it is equally unclear what the shi (the “scribe” as a
court official) was actually in charge of: annals, chronicles, genealo-
gies, daily records, or all of the above. The absence from the extant
record of annalistic compilations and chronicles—with the exception
of the Chunqiu (Spring and Autumn Annals of Lu, attributed to Con-
fucius) and its commentaries— that can be attributed to court scribes
suggests caution when we are called to define the manner in which
annals were redacted and the historical values that inspired them. It
is therefore a matter of conjecture if, and in what manner, a faithful,
“true to the facts,” if slavish chronicle of all events was ever compiled
by specialized, court-employed personnel. Recently some scholars have
pointed out that the primary concern of early historiographers was
not the factual reality of an event but rather its “ritual reality” (Gentz
2005, 235). What this notion entails is that adherence to reality was
subordinated to adherence to certain principles such as devising the
intention of the historical actor, placing him or her in a social hierar-
chy, examining the event in relation to established codes of conduct,
and other elements that effectively discriminated between what ought
to be recorded and the “nonrecordable,” therefore presenting a pic-
ture of the unfolding of events entirely controlled by a ritual matrix.
If the chief example of this type of historiography is the Chunqiu,
its commentaries somewhat paradoxically include more historical ma-
terial by aiming to instruct the reader on the reasons why a given event
was included or expunged from the record. In particular, the need to
provide an interpretation and an explanation of why a certain record
was missing in Confucius’s work implicitly points to the fact that such
 Truth and Unity in Chinese Traditional Historiography 267

an event nonetheless took place and was purged by Confucius because

of specific concerns. Known in Chinese historiography as “appropri-
ate concealment” (Ng and Wang, 38; Li 2011), this technique was un-
derstood as a form of moral chastisement.
The Zuozhuan (the Chunqiu’s chief commentary) often “reveals”
information that does not appear in the Chunqiu as facts that had been
concealed. In this act of revelation we can arguably see the implicit im-
perative, on the side of the historian, to be “true” to the event no mat-
ter how unpalatable and condemnable. The same work shows also
concerns about causality, since the occurrence of an event is often ex-
plained in a given political and social context or chain of events. The
circumstances of an event may also include elements that one may
regard as irrational or fantastic, such as dreams and the appearance of
ghosts. The reporting of “paranormal” events was part and parcel with
the explicatory structure, and it would certainly be erroneous to ex-
pect a fully rationalistic attitude in parsing plausible from fanciful.
Could one lend credence to an event that defied normal sensorial ex-
perience but was all the same related by otherwise trustworthy (i.e.,
neither deranged nor self-serving) people? The historian often chose
to report without making a judgment. Truth, if we understand it as
fidelity of the text to the representation of events, in preimperial histo-
riography resides also in reporting accurately what is known about an
event, no matter how fanciful. The “context” in which an event was
immersed is supposed to provide accurate information and moral rele-
vance to the narrative of a single event but not necessarily a rationalis-
tic explanation or internal coherence. Whether such a goal was even-
tually achieved in the Zuozhuan is at any rate immaterial because the
text that has reached us was certainly a composite product, in which
layers of accretions can be recognized (probably composed in the late
fourth century; Li 2011), and to seek a definite authorial intention
would be misdirected. What matters, rather, is the introduction of a
context that elaborates on the event by explaining its meaning as well
as its genesis and by doing so establishes a higher degree of persuasive-
ness and plausibility even in the presence of different, and sometimes
outlandish, explanations. This may not be a truth claim as we under-
stand it, but it is nonetheless a step that takes us closer to Sima Qian’s
 268 Nicola Di Co smo

implicit faithfulness to contradictory sources by reporting multiple

versions of the same event, which is a consistent feature of his histori-
cal method.
The transition from the Warring States to a unified empire in 221 bc
in historiographical terms means above all the transition, fully con-
scious in Sima Qian’s own programmatic statements, from the Zuo-
zhuan (as the most important product of the Chunqiu tradition) to
the Shiji. These two texts are on different evolutionary planes, each of
which defined a different historical conception, sensibility, and meth-
odology. Stephen Durrant, in discussing the question of truth in Shiji,
takes a pragmatic approach. He looks at truth in history in terms of
accuracy but also in terms of the greater truth that is supposed to be
embedded in historical knowledge. The authority of the Classics, the
Shangshu (Book of Documents), the Shijing (Book of Odes), and the
Chunqiu (Spring and Autumn Annals), was not questioned, at least
not directly, and the information they provided was accepted as foun-
dational to China’s intellectual tradition. The intent of sorting out
what was reliable and authoritative on one side and what was fanciful
and deceptive on the other carries with it a determination of truthful
representation of the material that the historian had at his disposal,
and therefore of the shape and substance of “history” as it was handed
down, and as we have it today. The challenge for Sima Qian was to
bring together in a coherent fashion all the different “traditions” while
at the same time determining which documents and traditions were
“true”(or simply more reliable) whenever he had to make a choice. In
this he chose to side with the texts that carried the greatest authority.
At the same time, since much of the Shiji covers the period when
Sima Qian and his father were alive (the first half of the Former Han
dynasty), for those contemporary or near-contemporary events the
standard for truth or accuracy involved directly the work of the histo-
rian as a witness and collector of facts, not just as interpreter and trans-
mitter of ancient texts. This double epistemology indicates the open-
ing of an important gap in Chinese historiography, which will be in
evidence from Sima Qian on and especially in relation to the compila-
tion of dynastic histories, between the usage and interpretation of in-
herited documents and the production of current records. Both opera-
 Truth and Unity in Chinese Traditional Historiography 269

tions were open to multiple theories and vulnerable to a variety of fal-

lacious practices, not to mention political pressures, personal inter-
polations, and intellectual agendas.
By joining these two planes of historical production into a single
overarching framework, Sima Qian’s Records created a holistic view
of Chinese history and laid the foundations on which later historians
could build. The model of the dynastic history that established itself
with the Han dynasty was therefore conceived from the beginning as
a modular (or segmental) addition to Sima Qian’s magnum opus but
should not be seen as a simple imitation. The Hanshu, while concep-
tually indebted to the structure of the Shiji, introduced changes in
the overall organization of the source materials, of which there were a
greater number than had been available to Sima Qian due to the ex-
pansion of the state bureaucracy and which were also more easily ac-
cessible because of the official status accorded to historians.
The styles of narrative history, moreover, varied considerably from
historian to historian, and the dynastic histories, while outwardly simi-
lar in structure, did not approach the question of “truthfulness” or re-
liability in the same manner. Fan Ye (398– 446), author of the History
of the Later Han Dynasty (Hou Hanshu), introduced significant sty-
listic innovations, and while his work has been praised for objectivity,
it has also been charged with reporting speeches that are either highly
dubious or fully invented by the historian (Bielenstein 1954, 52– 61).
The quality of the history that was produced was often judged by later
historians on the basis of the documents that were marshaled and of the
narrative style. As a “transmitter” of knowledge, the historian could
choose simply to report what he had learned and then let the reader de-
cide, and questions about the plausibility or reliability of a certain piece
of information, speech, or account were not always posed.
One category of historical records that has attracted particular in-
terest is that of portents and omens, which were often charged with
political significance. The History of the Later Han has a dedicated trea-
tise on portents and abounds with mentions of solar eclipses and natu-
ral disasters (Mansvelt Beck 1990, 156 ff.), but Fan Ye was certainly not
the first to pay special attention to portents. Sima Qian was ambiva-
lent about omens (Durrant 2011). As a Han astronomer, one of the
 270 Nicola Di Co smo

critical aspects of his intellectual formation concerned a belief in the

interconnectedness of human and heavenly realms, and therefore of
special linkages between, for instance, atmospheric and planetary phe-
nomena and human events. In this order of ideas, anomalies were seen
as carrying some meaning on the human level. If inexplicable things
could occur in the heavens as well as in the earthly world, the bound-
ary between the expected and the unexpected in human events could
not be defined in any rigorous manner either. The expectation of natu-
ral portents and anomalies, which could be charged with an array of
different meanings from political prognostications to after-the-fact ra-
tionalizations, required that even strange events be duly noted. Truth
was preserved by reporting all that seemed to be relevant, no matter
how improbable. Ultimately, it was the sensitivity of the historian that
remained the final measure, and as far as I can tell there is no argument
in early Chinese historiography that did not take the historian to be
the gatekeeper of historical truth.
A fundamental turn took place with the bureaucratization of of-
ficial historiography under the Tang dynasty (618– 907). The replace-
ment of an individual’s labors with the complex and cumbersome ac-
tivity of clerks and officials working in various departments under the
purview of the History Office (a branch of the state administration)
made a much greater volume of documents available to various com-
mittees of historians. The goal of these various offices was the compi-
lation of the Veritable Annals (shilu), which constituted the essential
chronicle of all events of an emperor’s reign period and would be used
by the succeeding dynasty to prepare the full dynastic history of the
previous one. The annals were a distillation of various types of rec-
ords, which included the Diaries of Activity and Repose and the Rec-
ords of Current Government, compiled respectively by the court di-
arists and the Great Minister, which were submitted to the History
Office. In turn, this office produced the Daily Records (Twitchett 1992).
All these materials, as well as reports and documents generated by other
branches of the government, flowed through a process of selection and
compression into the compilation of the Veritable Annals, to which bi-
ographies of eminent people who had died during the period treated by
each section of the Annals were then appended (Twitchett 1961). The
different phases of production and redaction were supposed to ensure
 Truth and Unity in Chinese Traditional Historiography 271

not just a complete set of records but also a type of history that reduced
the historical fact to a straightforward historical notation and thus lim-
ited as much as possible moralistic judgment and political manipula-
tion. The word shi in shilu, translated as “veritable” in the sense of true
and honest, indicates that there was an explicit concern for an account
that would not betray the truthfulness of the historical record. This
entailed whittling the record down to its basic essence, compressing it
into what one could call the “naked truth” of the event (Vogelsang
2005, 161– 62). The compilation of the shilu will remain a central fea-
ture of official historiography down to the Qing dynasty.
While making record-keeping activities grow to industrial propor-
tions, the Tang dynasty also produced the first systematic critique of
historical writing, of which truthfulness and adherence to a historical
ideal that valued independent thought were central aspects. In his Shi
Tong (General Principles of History), Liu Zhiji (661– 721) attacked
official historiography on several grounds (Ng and Wang 2005, 121– 28;
Pulleyblank 1961). First, he critiqued historiography as a collective en-
terprise. The main target of Liu Zhiji’s criticism was the “committee
work,” as every phrase had to be approved at various levels of scrutiny
before it could be inserted in the final version. Second, Liu protested
the fact that private historians had limited access to official documents.
Third, he denounced the intervention of powerful people who tried to
influence the historiographical work. Another point of criticism was
that official historians were not free to develop their own methods or
apply their own standards and were instead forced to follow the guide-
lines set by their supervisors, who were mere civil servants, and thus
incompetent as historians. He argued that while the various categories
in which historical knowledge was organized (in particular, annals,
treatises, and biographies) were acceptable as a general structure, each
era required additions that could accommodate new knowledge. Ac-
cording to Liu, historical writing should be concise, dry, and synthetic
and refrain from an overly ornate literary quality, as style might inter-
fere with a transparent exposition and obscure the facts.
If we try to extrapolate a sense of “truth” from Liu’s “principles,”
we would find it in his efforts to safeguard historiography from being
polluted by the bureaucratic apparatus and his focus on the essential.
While Liu called into question the reliability of received texts, including
 272 Nicola Di Co smo

some Confucian Classics, he emphasized the rectitude and passion of

historians as individuals devoted to their craft and how this would pro-
tect the truthfulness of the historical narrative from mistakes and inter-
polations. The sophisticated methods of analysis that the philological
school of the Qing period developed to investigate textual authenticity
were still to come (see below), but the fact that historical criticism de-
veloped at this early stage testifies to the advances of Tang historiogra-
phy, not just in terms of the institutions and “bureaucratic technology”
involved in the production of official histories, but also in the lucid
awareness that at least some historians showed about the dangers to
which this gargantuan organization exposed historical knowledge.
Later historians often voiced Liu’s complaints, such as the inability
of collective official enterprises to produce high-quality historical works,
the lack of due attention to the intrinsic value of each document and
source, the inadequate, careless treatment of these sources, and the ab-
sence of a critical opinion and judgment of the historical event, to which
a moral content should be assigned, so that it would not just be a mere
record of the past. The Ming scholar Wang Shizhen (1526– 90) about
eight years later, in his “Critical Treatise on Historical Errors” (Shi-
sheng kaowu), made a vigorous critique of Ming official historiography
that was reminiscent of Liu’s positions (Ng and Wang 2005, 213–14). He
appreciated the fact that the History Office was able to gather official
documents, which a single historian could not do, but he found that
the highest standards of critical interpretation and philological accuracy
could only be found among private historians working independently.
At the same time he acknowledged that several privately produced his-
tory works contained subjective personal views, often the product of
disgruntled literati venting their anger and frustration.
The aforementioned dichotomy between private and official histo-
riography therefore presents dangers to the rendition of “truthful” ac-
counts on either side of the divide. If officially produced histories were
open to general incompetence, political pressures, and the absence of a
critical mind that could organize and give meaning to the various rec-
ords, private works suffered from limited access to documents, an often
biased interpretation of events, and the historian’s own limitations as a
creative scholar.
 Truth and Unity in Chinese Traditional Historiography 273

One of the most acclaimed historians of all ages, Ouyang Xiu, who
directed for a period of time the compilation of the New Tang History
and authored private works of historical scholarship, such as the New
History of the Five Dynasties (Xin Wudai shi or Wudai shiji), was the
most prominent Confucian intellectual of the eleventh century (Lee
2002). The intellectual agenda introduced in historical writing led to
both a greater attention to the language of historical writing and an im-
poverishment of the overall historical record. Ouyang Xiu tried to imi-
tate the original Confucian method of implicit judgment and privileged
terse and essential usage of an allusive language as a means to summon
historical lessons for philosophical purposes. However, Ouyang deleted
from the record a large portion of documents preserved in the Old Tang
History because they did not meet his high standards for philosophical
and historical worthiness. The elimination of a large part of the available
historical documentation makes his history a less valuable source than
the Old Tang History, but the standards he adopted set an example that
would be praised and imitated by later scholars (Ng and Wang 2005,
136– 41). Ouyang’s commitment to accuracy and reliability in the se-
lection of documents does not mean that he saw history simply as a
catalog of events to be transmitted in an objective and detached manner.
The author’s voice is preserved, and his sense of moral outrage and
philosophical leanings are openly expressed in the sections in which he
exposes his own opinion, which begin with the phrase “I lament . . .”
(Hartman and DeBlasi 2012). If until then historical production had
been a bureaucratic enterprise that served the government as a whole
and preserved a cultural record that was shared by scholars, officials,
and politicians as a common heritage, with Ouyang Xiu it was appro-
priated by the Confucian literati to impose their sense of what history
ought to be, teach, and transmit. In regard to the question of truthful
representation, Ouyang deeply believed in an objective recording of
events and did not support the theory of legitimate succession (zheng-
tong) to determine the proper sequence of dynasties. The debates about
the legitimate transmission of imperial rule that flourished under the
Song and also during the Jin dynasty (Chan 1984) necessarily hinged
on historical questions, since the very legitimacy of a given dynasty de-
pended on its moral foundations, and these moral foundations rested
 274 Nicola Di Co smo

on the interpretation of historical events. Given that the authority of

the emperor was theoretically universal and indivisible and that there
could only be a single legitimate “Son of Heaven,” the political impor-
tance of a historiography that upheld the correctness of the succession
was self-evident. It is not an accident, however, that issues of zheng-
tong were downplayed in Song historiography. The territorial unity
that China had achieved during the Tang had not been reconstituted,
and the Song partial reunification occurred after about sixty years of
division during which various short-lived dynasties succeeded each
other and fought for legitimacy. The Song lived in a period of multiple
rulerships, and while they never doubted that their dynasty was the
legitimate successor to the Tang, some practical accommodation was
required (Wang 1983).
Both Sima Guang (see below) and Ouyang Xiu (1007– 72) champi-
oned what was undoubtedly seen as History, with a capital H: a type of
History that was as reliable as possible and at the same time animated
by a moral purpose. Historical analogy and the “exemplary lives” in-
cluded in the biographical sections were vehicles of moral teaching used
not only for the edification of the emperor and state officials but also,
and more importantly, for the recruitment of civil servants through the
examinations system. Officials were expected to be knowledgeable in
history and used historical allusions, quotations, and references, in a
variety of contexts: memorials to the throne, debates, policy decisions,
and even military operations.
The triumph of Neoconfucian historiography is nowhere more
evident than in the writings of the preeminent philosopher of his age,
Zhu Xi (1130–1200). His work, the Tongjian gangmu (Outline and De-
tails of the Comprehensive Mirror), based on Sima Guang’s encyclope-
dic opus, became part of state orthodoxy toward the end of the South-
ern Song and continued under the Mongols when recruitment of official
bureaucrats by means of state exams was reinstated (1313). This work
is the most celebrated expression of the notion that history should
serve a greater “truth,” namely, a philosophical ideal to guide society
and government according to Confucian codes and norms. Historical
precedents drawn from Sima Guang’s Zizhi Tongjian (Comprehensive
Mirror for Guide in Government) were extracted and reworked using
the “praise and blame” coded language of the Chunqiu (Hartman 2012,
 Truth and Unity in Chinese Traditional Historiography 275

50; Ng and Wang 2005, 159– 64). Truth was to be found by penetrating
the perennial, unrelativized moral essence of universal principles re-
vealed in the seamless continuum of recorded events, sieved through
the sensitivity of the historian and their relevance to one’s own age.
Qing scholarship by and large rejected the Neoconfucian ortho-
doxy that set in during the Song dynasty. For Qing historians, espe-
cially of the eighteenth century, the issue of truthfulness in history
rested primarily with the philological precision and source criticism
that became the hallmark of the school of “evidential research.” Docu-
ments were checked carefully for errors and contradictions and sub-
jected to a search for authenticity that in turn led to the development
of various ancillary disciplines, such as archaeology and historical ge-
ography. Qing emperors themselves took a keen and direct interest in
historical production. The philological turn made the didactic aspects
of history far less relevant and raised the standards of professionalism
in the historian’s craft. The analytical tools developed by Qing schol-
ars allowed them to correct errors present in the standard histories
through a meticulous comparative examination of extant documents.
The treatises, biographies, and annals that were utilized in the compi-
lation of standard histories and other works, because of the convo-
luted, hasty, or simply incompetent handling of the documents, had
produced internal inconsistencies and contradictions that detracted
from the value of the work. Evidential scholarship went about restor-
ing “truth to history” in an especially rigorous manner (Ng and Wang
2005, 244), and several products of official historiography that were
regarded as especially poor, such as the Song and Yuan dynastic histo-
ries, were targeted for emendations and corrections. The view of schol-
ars such as Qian Daxin (1728–1804) was that the historical accounts had
to be checked against all extant documents and relevant works. The
expansion of the range of historical sources led to what one might call
a “dense history,” in which research in various areas was brought to-
gether in order to elucidate the actual historical facts.
Together with high-quality textual exegesis, the Qing period is
notable for the emergence of influential works of historical criticism.
Zhang Xuecheng (1738–1801), arguably the most important historian
of the Qing dynasty short of the reformists of the end of the nineteenth
century, in his work “General Principles of Literature and History”
 276 Nicola Di Co smo

(Wenshi tongyi) combines a notion of historical cyclicality, whereby

permanent universal principles manifest themselves over and over
again, expressing a notion of change and evolution. Institutions, gov-
ernments, laws, and political forms evolve but are still subject to fail-
ure or success depending on how they may fit those universal principles
(dao) that harmonize nature, cosmos, and humankind. Without the
knowledge of how society, institutions, laws, and the human condi-
tion in general have evolved, it would be impossible to perceive what
action, solutions, and measures a certain era requires. In this he remained
a fairly orthodox Confucian while rejecting a notion of history fully
subordinated to the philosophical tenets and didactic purposes of the
Neoconfucian school (daoxue). Besides this fundamental work of his-
torical criticism, Zhang’s most important contribution as a historian
was in local history, focusing on the collection of local documents, in-
scriptions, and manuscripts as an integral part of the historical patri-
mony. It has been noted that Zhang’s historiographical ideal was one
in which the historian was able to convey a sense of empathy toward
the past as a moral mission and as a way of making historical knowl-
edge relevant to the present (Mittag 2005, 392). One scholar has com-
pared Zhang to Giambattista Vico for the attention paid by both (who
lived one generation apart) to the philosophical foundation of history
and to its moral purpose. Demiéville saw in Zhang Xuecheng’s work
something analogous to Vico’s aim “to break the bulkhead between
philology and philosophy, to fecundate them mutually, to convert the
‘certainty’ of historical facts, as established by philology, into truths of
reason which were the object of philosophy (verum ipsum factum)”
(Demiéville 1961, 185; original emphasis).

Unit y

There are various ways to approach the question of “unity,” but a pos-
sible starting point is the distinction between dynastic and compre-
hensive history, which is one of the great dichotomies that traverse the
full spectrum of Chinese historiography. The unity of historical knowl-
edge was predicated for the first time by Sima Qian. In the celebrated
letter to his friend Ren An, the historian candidly exposed the reasons
 Truth and Unity in Chinese Traditional Historiography 277

that he had embarked on an enterprise that was so uniquely consum-

ing, and which proved so dangerous.

I too have ventured not to be modest but have entrusted myself to

my useless writings. I have gathered up and brought together all
the old traditions of the world that were scattered and lost. I have
examined the events of the past and investigated the principles be-
hind their success and failure, their rise and decay, in 130 chapters.
I wished to examine all that concerns heaven and humankind, to
penetrate the changes of the past and present, putting forth my
views as one school of interpretation. (Watson 1993, 236)

The quest for patterns and laws in human events belongs to the
same intellectual milieu as the work of the astrologer who explores the
heavens in search of regularities that might reveal a transcendent truth.
To bear fruit, the examination has to be constant and the records accu-
rate, detailed, and comprehensive over a long period. Therefore the
accuracy of the event has to be matched by a unity of the whole record
that allows the historian (and astronomer) to spot regularities and es-
tablish cyclicalities. The “cycle” is a notoriously complicated concept
in Chinese history and philosophy. On the one hand, it may simply
refer to a dynastic cycle, but in fact it stems from metaphysical notions,
such as the dao or universal principle, that go back to the very roots of
Chinese historical thought (Huang 1995). However, regardless of how
these “cycles” were understood, they implied the existence of tran-
scendent laws that determined the serial occurrence of phenomena in
the natural world and in human affairs.
In particular, change was inscribed by Sima Qian in an imperial
vision, and the motives for his historical revolution were closely linked
to the unification of the empire achieved by the Qin dynasty and con-
solidated by the long-lasting Han dynasty. The empire, now politically
and culturally coherent, found in the Shiji its perfect historical expres-
sion. The historian deliberately gathered the scattered records from
the various preimperial kingdoms in a single composite architecture
that transmitted a holistic vision of history both in chronological scope
(from the remotest origins to the present) and in political and territorial
terms. The concept of tianxia (All under Heaven) expressed both the
 278 Nicola Di Co smo

universalistic claims of the emperor and the actual spatial and cultural
extension of the empire, whose frontiers were marked by foreign,
often hostile independent peoples. Sima Qian had indeed a clear sense
of the limits of the empire, marked as they were by the inclusion of
foreign peoples and of far-flung regions which were also essential to
the definition of China as a bounded community. No area or period of
recorded human activity was regarded as laying outside the scope of
historical investigation. The model of this universal, comprehensive
view is reflected also in the compilation of histories of foreign peoples,
such as the attempt to establish a genealogy of the Xiongnu (a nomadic
people who established a powerful empire hostile to the Han dynasty)
from a mythical beginning that was traced back to a time more or less
consistent with the beginning of China’s antiquity. Sima Qian’s views
on the history of his own times and representation of events drew criti-
cism from his immediate successors mainly for political reasons, given
that the historian’s unorthodox and independent views were often un-
derstood as a direct challenge to the emperor’s authority. His model of
comprehensive history was then abandoned in favor of the dynastic
form, but the unsurpassed beauty of the grand civilizational vision that
inspired it remained culturally influential.
This model was to be revived, over a thousand years later, by the
celebrated work by Sima Guang (1019– 86), the Zizhi Tongjian (Com-
prehensive Mirror for the Practice of Government) (completed in 1084),
which covers the period from the Warring States to the year before the
founding of the Song dynasty (403 bc to ad 959). As a universal his-
tory, rather than a dynastic history, it aimed to present a detailed year-
by-year annalistic record from the Golden Age of Confucius to the
present. As a “mirror,” the Zizhi Tongjian maintained the vital func-
tion of history as an aid to the present in all political and administra-
tive matters, or, in other words, as essential knowledge to assist any-
one involved with government functions (Hartman 2012). The unity
of the narrative and the inherent patterns and cycles that one may draw
by comparing various periods could instruct statesmen and emperors
as to the best course of action, or at any rate to be less blind to the
potential consequences of a certain course of action.
What is especially remarkable in Sima Guang’s method is the fu-
sion of a highly moralistic approach with equally high standards of
 Truth and Unity in Chinese Traditional Historiography 279

accuracy in vetting, organizing, and cross-checking documents in a

process aimed to eliminate contradictions (by favoring the most reliable
sources) or filling out lacunae (by looking for additional secondary
sources). Those sources that were suspected of political or personal bias
were simply eliminated. The thirty fascicles comprising the “Examina-
tion of Discrepancies” show a remarkable effort to wrestle with prob-
lems of source analysis and critical interpretation. The discussion of
what version of an event “seems to be closest to the truth” (Pulley-
blank 1961, 157; cited in Ng and Wang 2005, 149) indicates the ongoing
concern with the reliability of the source and the production of a rec-
ord that was as consistent as possible with all available information,
in order to transmit what could be regarded as the “real event.” At the
heart, Sima Guang aimed to report faithfully those human achievements
that needed to be taken into account, and he did so in an annalistic
form that included every year of Chinese history from the age of Con-
fucius on without overtly grafting overarching philosophical or moral
principles onto the flow of events. According to Thomas Lee (2002,
70), the greatest achievement of Sima Guang’s historiography is to be
found in textual criticism and the rejection of supernatural forces or
other fanciful explanations.
A long-running question in Chinese historiography consists of
the evaluation of change. “Change,” understood often, as mentioned
above, in terms of cyclical recurrence, tended to oscillate between pure
philosophical speculation, in which historical eras were identified with
dominant “agents” and “forces,” each of which carried certain qualifi-
cations, and the search for “laws” and patterns of change inherent to
human institutions and principles of governance. The circularity inher-
ent in these schemes was supposed to both explain the past and present
and predict the future. But already in the Tang and during the Northern
Song the issue of change was approached historically by searching for
discontinuities and transformations within the flow of human activities
and the development of institutions. Zheng Qiao (1104– 62), the South-
ern Song author of the Tongzhi (Comprehensive Treatises) assembled
pieces of former histories to create a long narrative not constrained by
the dynastic model that would reveal causal nexuses as well as fractures and
continuities in a number of areas relevant to society, politics, and gov-
ernment: rituals, music, laws, civil service, and economy (Ng and Wang
 280 Nicola Di Co smo

2005, 156– 58). Ma Duanlin (1254–1324/5), a Yuan historian, in his Wen-

xian tongbao (Comprehensive Investigation of the Literary Traditions),
documented the evolution of government institutions from the dawn
of history to the end of the Song. He propounded a rational evolu-
tionary view of history by uncovering the reasons for changes in the
historical process through a periodization based on distinctive devel-
opments of government institutions (Chan 1982, 69– 72). The focus on
institutions, examined systematically in twenty-four sections and rang-
ing from land taxation and population to court rituals, music, and the
military, allowed him to show history not just as a series of events, but
as organic development within which continuity and change had to be
evaluated on a grander scale than was allowed by the dynastic history
(Ng and Wang 2005, 185– 86). Following Zheng Qiao, he questioned
the applicability of the Five Agents theory and the moralistic principles
of the Chunqiu to history. In terms of Confucian values, he espoused
the promotion of welfare, the suppression of private interest, and a
commitment to a higher moral order.
An encyclopedic approach to history was one of the defining fea-
tures of Song and post-Song historiography and is responsible for the
en masse rearrangement of the historical patrimony into different con-
figurations, introducing a comprehensive vision that gave a different
sense and direction to historical knowledge. Although such a decon-
structive operation had didactic and moralistic purposes, it also led to
major advances in historical criticism and in a general dismissal of the
supernatural or metaphysical in favor of an anthropocentric view of
human events. It has been noted that the widespread use of the word
tong in Song historiography links semantically the notion of com-
prehensive (in the sense of complete or encyclopedic) and the idea of
comprehension (in the sense of penetrating and understanding) already
present in Liu Zhiji’s Shitong, that is, “comprehending history” (Lee
2002, 75).
Finally, a few words should be said on the relationship between the
Classics and History as this is one of the major themes in Chinese histo-
riography, and is directly relevant to a conception of historical knowl-
edge as an independent branch of intellectual inquiry. In a nutshell, one
could say that History was part of Classical learning from the very be-
ginning, given the status accorded to the Shangshu and the Chunqiu,
 Truth and Unity in Chinese Traditional Historiography 281

but it also broke free of that tradition by being accorded a separate sec-
tion in the early classifications of books that we find in the dynastic his-
tory of the Sui dynasty, compiled under the supervision of Wei Zheng
(580– 643). If during the Han and the Tang dynasties historical practice
and ideals had moved away from Classical learning and had become
independent forms of intellectual production, the literati of the daoxue
school of the Southern Song and Ming strongly reaffirmed the subor-
dination of human affairs to an ethical hermeneutic that used the Con-
fucian Classics as its control system. Only during the Qing dynasty,
with evidential research and “exegetical criticism” (to borrow the term
coined by D.W. Robinson), did historical research become an intellec-
tual activity that was able not just to gain meaning from the reconstruc-
tion of a much more accurate and error-free historical record, but to
generate its own theories of knowledge. The most important exponent
of Qing historical criticism, the aforementioned Zhang Xuecheng, fa-
mously said that even the Six Classics are historical writing. This dictum
has been understood in different ways, but in general we can say that
with it Zhang Xuecheng placed the Classics in their own historical con-
text, and in doing so relativized their universal meaning, dislocating the
usual perception according to which the Classics had eternal significance
while the works of history dealt with an ever-changing reality whose
underlying patterns and laws, when they could be devised, were only
valid for the period in question. For Zhang, the Classics were produced
by the sages of antiquity because here was a contingent necessity to
manage the state and to nurture the people (Mittag 2005, 390), and there-
fore the creation of the Classics should be seen as the product of specific
historical circumstances (Elman 2002, 135). During the Qing, too, how-
ever, the Confucian canon remained central to Qing scholars’ learning
and continued to embody the ideals, with respect to social order, moral
conduct, and proper governance, that had inspired literati for centuries.
What changed was the intensity with which the past was scrutinized,
not just to gain moral lessons, but to establish the best version of it.

We can say that truth and unity, in the manner defined above, have
been critical ingredients in the development of Chinese historiography.
The foundations of the organization of historical knowledge were laid
 282 Nicola Di Co smo

at the very beginning of imperial China, in the particular intellectual

milieu that followed the imperial unification due to the efforts of indi-
viduals who understood history as a fundamental branch of knowledge,
and indeed, in Sima Qian’s view, the type of knowledge that could unify
all knowledge. The key principles that guided historical writing from
the beginning, and especially the standard histories, show a large de-
gree of continuity, yet were never static. The institutionalization of his-
tory writing led to the preservation of a large mass of documents and
elevated history to a government enterprise, a societal value, and an ob-
ject of study, but from a very early stage also encouraged experimenta-
tion, innovation, criticism, and much rewriting. A factor of change was
the tension between the production of historical works as records of
the past, their political implications, and the moral teachings that the
historical narratives were supposed to convey. Each of these realms had
its own standards of “truth.” The philosophical underpinnings of his-
tory writing, with their emphasis on “praise and blame,” the didactic
value of history, and its subordination to a philosophically and politi-
cally correct interpretation of events, continued to clash with the need
to organize, gather, select, and evaluate historical documents with ob-
jectivity and trustworthiness. Finally, the bureaucratization of histori-
cal writing to a certain degree stymied historical production but did
not prevent private historians from voicing criticisms and striving for
a better type of history. The contrasts and dialogue between the two
stimulated the growth of historical ideals and alternative modes of his-
tory writing, and continue to do so to this very day.

R e f e re nce s

Balazs, E. 1961. “L’Histoire comme guide de la pratique bureaucratique (les

monographies, les encyclopédies, les recueils des statuts).” In Historians
of China and Japan, ed. W. G. Beasley and E. G. Pulleyblank, 78– 94.
London: Oxford University Press.
Beasley, W. G., and E. G. Pulleyblank, eds. 1961. Historians of China and
Japan. London: Oxford University Press.
Bielenstein, Hans. 1954. “The Restoration of the Han Entry, with Prolegom-
ena on teh Historiography of the Hou Han Shu.” Bulletin of the Mu-
seum of Far Eastern Antiquities 26:1–81.
 Truth and Unity in Chinese Traditional Historiography 283

Chan, Hok-lam. 1982. “‘Comprehensiveness’ (t’ung) and ‘Change’ (pien) in

Ma Tuan-lin’s Historical Thought.” In Yuan Thought: Chinese Thought
and Religion under the Mongols, ed. Hok-lam Chan and W. Theodore de
Bary, 27– 87. New York: Columbia University Press.
———. 1984. Legitimation in Imperial China: Discussions under the Jurchen-
Chin Dynasty (1115–1234). Seattle: University of Washington Press.
Dai, Yi. 2009–10. “The Origin of the Qingshi (Qing History) and Its Initial
Planning.” Chinese Studies in History 43 (2): 6–14.
Davis, Richard L. 2004. Introduction to Historical Records of the Five Dynas-
ties, by Ouyang Xiu, lxiii–lxxix. Trans. Richard L. Davis. New York: Co-
lumbia University Press.
Demiéville, P. 1961. “Chang Hsüeh-ch’eng and His Historiography.” In His-
torians of China and Japan, ed. W. G. Beasley and E. G. Pulleyblank,
167– 85. London: Oxford University Press.
Durrant, Stephen. 2005. “Truth Claims in the Shiji.” In Historical Truth, His-
torical Criticism and Ideology: Chinese Historiography and Historical
Culture from a Comparative Perspective, ed. Helwig Schmidt-Glintzer,
Achim Mittag, and Jörn Rüsen, 93–114. Leiden: Brill.
———. 2011. “The Han Histories.” In The Oxford History of Historical Writ-
ing, vol. 1: Beginnings to ad 600, ed. Andrew Feldherr and Grant Hardy,
485– 508. Oxford: Oxford University Press.
Elman, Benjamin A. 2002. “The Historicization of Classical Learning in Ming-
Ch’ing China.” In Turning Points in Historiography: A Cross-Cultural
Perspective, ed. Q. Edward Wang and Georg G. Iggers, 101– 44. Rochester,
NY: University of Rochester Press.
Feldherr, Andrew, and Grant Hardy, eds. 2011. The Oxford History of Histori-
cal Writing, vol. 1: Beginnings to ad 600. Oxford: Oxford University Press.
Foot, Sarah, and Andrew Robinson, eds. 2012. The Oxford History of Histori-
cal Writing, vol. 2: 400–1400. Oxford: Oxford University Press.
Gentz, Joachim. 2005. “The Past as a Messianic Vision: Historical Thought and
Strategies of Sacralization in the Early Gongyang Tradition.” In Histori-
cal Truth, Historical Criticism and Ideology: Chinese Historiography and
Historical Culture from a Comparative Perspective, ed. Helwig Schmidt-
Glintzer, Achim Mittag, and Jörn Rüsen, 227– 54. Leiden: Brill.
Hartman, Charles. 2012. “Chinese Historiography in the Age of Maturity,
960–1368.” In The Oxford History of Historical Writing, vol. 2: 400–1400,
ed. Sarah Foot and Andrew Robinson, 37– 57. Oxford: Oxford University
Press.
Hartman, Charles, and Anthony DeBlasi. 2012. “The Growth of Historical
Method in Tang China.” In The Oxford History of Historical Writing,
vol. 2: 400 –1400, ed. Sarah Foot and Andrew Robinson, 17– 36. Oxford:
Oxford University Press.
 284 Nicola Di Co smo

Huang, Chun-chieh. 1995. “Historical Thinking in Classical Confucianism—

Historical Argumentation from the Three Dynasties.” In Time and Space
in Chinese Culture, ed. Chun-chieh Huang and Eric Zürcher, 72– 88. Lei-
den: Brill.
———. 2004. “The Philosophical Argumentation by Historical Narration in
Sung China: The Case of Chi Hsi.” In The New and the Multiple: Sung
Senses of the Past, ed. Thomas H. C. Lee, 107– 20. Hong Kong: Chinese
University Press.
Hulsewé, A. F. P. 1961. “Notes on the Historiography of the Han Period.” In
Historians of China and Japan, ed. W. G. Beasley and E. G. Pulleyblank,
31– 43. London: Oxford University Press.
Lee, Thomas H. C. 2002. “New Directions in Northern Sung Historical Writ-
ing.” In Turning Points in Historiography: A Cross-Cultural Perspective,
ed. Q. Edward Wang and Georg G. Iggers, 59– 88. Rochester, NY: Uni-
versity of Rochester Press.
Lewis, Mark. 2011. “Historiography and Empire.” In The Oxford History of
Historical Writing, vol. 1: Beginnings to ad 600, ed. Andrew Feldherr and
Grant Hardy, 440– 62. Oxford: Oxford University Press.
Li, Wai-yee. 2011. “Pre-Qin Annals.” In The Oxford History of Historical
Writing, vol. 1: Beginnings to ad 600, ed. Andrew Feldherr and Grant
Hardy, 415– 39. Oxford: Oxford University Press.
Mansvelt Beck, B. J. 1990. The Treatises of the Later Han: Their Author, Sources,
Content, and Place in Chinese Historiography. Leiden: Brill.
Mittag, Achim. 2005. “What Makes a Good Historian: Zhang Xuecheng’s Pos-
tulate of ‘Moral Identity’ (shi de) Revisited.” In Historical Truth, Histori-
cal Criticism and Ideology: Chinese Historiography and Historical Cul-
ture from a Comparative Perspective, ed. Helwig Schmidt-Glintzer, Achim
Mittag, and Jörn Rüsen, 365–405. Leiden: Brill.
Ng, On Cho, and Q. Edward Wang. 2005. Mirroring the Past: The Writing and
Use of History in Imperial China. Honolulu: University of Hawai’i Press.
Nienhauser, William H., Jr. 2011. “Sima Qian and the Shiji.” In The Oxford
History of Historical Writing, vol. 1: Beginnings to ad 600, ed. Andrew
Feldherr and Grant Hardy, 463– 84. Oxford: Oxford University Press.
Pulleyblank, E. G. 1961. “Chinese Historical Criticism: Liu Chih-chi and
Ssu-ma K’uang.” In Historians of China and Japan, ed. W. G. Beasley
and E. G. Pulleyblank, 135– 66. London: Oxford University Press.
Schaberg, David. 2005. “Platitude and Persona: Junzi Comments in Zuozhuan
and Beyond.” In Historical Truth, Historical Criticism and Ideology: Chi-
nese Historiography and Historical Culture from a Comparative Perspec-
tive, ed. Helwig Schmidt-Glintzer, Achim Mittag, and Jörn Rüsen, 177– 96.
Leiden: Brill.
 Truth and Unity in Chinese Traditional Historiography 285

Schmidt-Glintzer, Helwig. 2005. “Why Has the Question of Truth Remained

an Open Question Throughout Chinese History?” In Historical Truth,
Historical Criticism and Ideology: Chinese Historiography and Historical
Culture from a Comparative Perspective, ed. Helwig Schmidt-Glintzer,
Achim Mittag, and Jörn Rüsen, 115– 30. Leiden: Brill.
Schmidt-Glintzer, Helwig, Achim Mittag, and Jörn Rüsen, eds. 2005. Histori-
cal Truth, Historical Criticism and Ideology: Chinese Historiography and
Historical Culture from a Comparative Perspective. Leiden: Brill.
Twitchett, D. C. 1961. “Chinese Biographical Writing.” In Historians of China
and Japan, ed. W. G. Beasley and E. G. Pulleyblank, 95–114. London:
Oxford University Press.
———. 1992. The Writing of Official History under the T’ang. Cambridge:
Cambridge University Press.
Vogelsang, Kai. 2005. “Some Notions of Historical Judgment in China and the
West.” In Historical Truth, Historical Criticism and Ideology: Chinese His-
toriography and Historical Culture from a Comparative Perspective, ed.
Helwig Schmidt-Glintzer, Achim Mittag, and Jörn Rüsen, 143– 75. Leiden:
Brill.
Wang, Gungwu. 1983. “The Rhetoric of a Lesser Empire: Early Sung Relations
with Its Neighbors.” In China among Equals: The Middle Kingdom and
Its Neighbors, 10th–14th Centuries, ed. Morris Rossabi, 46– 65. Berkeley:
University of California Press.
Wang, Q. Edward, and Georg G. Iggers, eds. 2002. Turning Points in Histo-
riography: A Cross-Cultural Perspective. Rochester, NY: University of
Rochester Press.
Watson, Burton. 1993. The Records of the Grand Historian: Qin Dynasty. New
York: Columbia University Press.
Weigelin-Schwiedrizik, Susanne. 2005. “History and Truth in Chinese Marxist
Historiography.” In Historical Truth, Historical Criticism and Ideology:
Chinese Historiography and Historical Culture from a Comparative Per-
spective, ed. Helwig Schmidt-Glintzer, Achim Mittag, and Jörn Rüsen,
421– 64. Leiden: Brill.
 P A R T V

-
Art and Religion
 C h a p t e r T h i r t e e n

The Architecture School within the University

-
Michael Ly ko udi s

Traditionally, the city is where we have gathered to find shelter, secu-

rity, and economy. The city serves our nature, our individual needs,
and our common purposes. It is where we come to find truth. The idea
of the city historically has embodied and brought together all of the as-
pects of culture from a multitude of professions and vocations: politics,
law, commerce, agriculture, and worship. For traditional and classical
architects and urbanists of today, the city and its architecture represent
the physical manifestation of the unity of knowledge.
In this essay I first outline how architects engage at varying levels
virtually all the disciplines found at the modern university. To be archi-
tects, we do not need to be, nor can we be, scholars in many of these
disciplines, but within the confines of the needs of the art of building,
we must master many of them and apply them to the creative process
and execution of our craft. Second, I outline how the traditional city
and its architecture represent the unity of knowledge as it mediates be-
tween nature and the man-made and the universal and the specific
through the monumental and the vernacular.
With respect to the sciences, the architect needs to have an under-
standing of mathematics. On the most pedestrian level, area and volume

289
 290 Michael Lyko udis

calculations, proportions, and geometry are essential to the design pro-

cess as well as to meeting the need for increasingly complex computa-
tions with respect to building codes, zoning ordinances, and environ-
mental sustainability requirements. For understanding the complex
structural, mechanical, and electrical loads on a building and commu-
nity, the architect is trained to use algebra, trigonometry, and calculus
and apply these to his or her work. Truth be told, we generally like to
let our friends the engineers actually be the ones on the record for the
calculations, but that is a liability issue more than a professional neces-
sity. In fact, some architects, such as Santiago Calatrava, insist on de-
signing their own structural systems. A foundation in engineering itself
is also essential. An understanding of and ability to perform the me-
chanical, structural, electrical, and civil engineering on a project is re-
quired of the architect if he or she is to speak knowledgeably with the
engineer. Physics is another area in which architects must have a solid
grounding. An understanding of the materials and properties that keep
buildings standing, liquids moving, sounds isolated, and air managed
requires a fairly substantial background in physics. Biology is probably
an unexpected area in which architects must be knowledgeable. From
biology, architects have learned analytical methodologies such as the
transect as they try to better understand the city as an ecosystem. As
we try to also better appreciate the city and the countryside and the
distinction between the two for reasons of environmental sustainability,
we need to be educated about a region’s fauna and flora as well as the
effect of microorganisms on buildings and their interiors. Chemistry is
an essential ingredient in the education of the architect. For example,
galvanic action and other chemical reactions can play a critical role in
the performance and durability of a building.
In the traditional architect’s mind, architecture and the city are
more about a sense of place than they are about a sense of time. Thus
anthropology links some of the concerns of the biologist, examining
the natural resources in a given locale and how the local culture uses
and adapts the local geology and patterns of circulation through local
conventions. In addition, psychology and organizational behavior are
needed to better understand how communities interact, the dynamics
within one’s own office, and, of course, the mind-set of the all-important
client.
 The Architecture School within the University 291

The architect must also know the law: from zoning and building
codes and limits to professional liability and contracts. Architecture is
ultimately a business. Successful architects must become entrepreneurs
and adept at marketing, finance, accounting, and business ethics.
Most important, architecture is an art. In antiquity, the architect
managed the sculpture, painting, and iconographic program of the
building. The building itself is an artistic object, subject to principles
of composition, such as solid-void relationships, as are the other arts.
The architect places all these disciplines in the service of resolv-
ing competing needs and forces. Dissimilar things can coexist within a
given structure as long as there is mediation among them. Architects,
and particularly classical architects, mediate among these to achieve
balance, harmony, and order.
The relationship of solid to void is a good place to begin to exam-
ine how we are able to describe our physical world and, by extension,
the solar system, galaxy, and universe. Principally, in the visible uni-
verse, we have solids (matter) and voids (space). In a sculpture or a
building, the voids are equally as important as the solids. The pattern
of the openings or voids in a wall creates rhythms that make up the
composition. The changing size of these openings adds to the grammar
and syntax of architectural compositions, much like notes on a piece of
music, which was once considered the allied art to architecture. Each
bar has a measure that has a beat: quarter notes, half notes, and the like,
tone and melody. The rhythm of the building can be seen in the regular
cadence of openings and the melody in the punctuating changes in size
and proportion of the openings. We can think of a layered facade as the
many instruments of an ensemble playing their own part in the compo-
sition; each would be lessened without the company of the others.
What the solid-void relationships in architecture actually represent
is the very basic nature of the visible universe. This leads us to the prin-
ciple that dissimilar things can coexist within the same structure, and
they are defined by what they are as much as by what they are not. A
colonnade, for instance, is as much about the space between the col-
umns and the columns themselves. The character of a colonnade changes
significantly as one adjusts the distance between the columns.
Take what is often referred to as human nature. Our nature is
not uniquely human; we are part of the cosmos and are hardwired to
 292 Michael Lyko udis

comprehend, analyze, appreciate, and create within its structure. We are

also made to stand upright and look forward. We privilege what we see
as a front, face, or facade. We organize our built environment around a
Cartesian plane or mathematical model that not only describes how we
perceive the physical world in three dimensions, but how we can also
rationally organize our cities, towns, and villages into streets, squares,
and blocks, place one stone on top of another to build our shelters, and
create what is often described as “the city as a second nature.”
The Roman castrum, or military encampments, were expediently
organized in two dimensions with a north-south axis called the Cardo
and the east-west axis called the Decumanus. The Roman city followed
that principle, and hundreds of cities, towns, and villages in Europe were
arranged along those lines. We see this around the world as well: Teoti-
huacán in Mexico, the diagram for the ideal Chinese city, or Marseille.
The other human natures that the city facilitates and expresses are
those of the public and the private. We are social creatures and find
knowledge, economy, inspiration, comfort, and joy in the company of
others. The opposite of our social nature is the need for privacy. We
gain security, inspiration, and comfort in the privacy of our homes. We
cannot be isolated for long periods, and we need to return to our public
realm after being recharged by our isolation. Another part of the pri-
vate realm has traditionally been commerce, which until the twentieth
century was largely the domain of individuals. The need for identity,
to compete and vie for excellence, is as much a part of the human con-
dition as the need for collaboration and cooperation. The city with its
public and private spaces facilitates this dual and dissimilar nature.
The private realm is represented by two types of buildings in the idea
of the city: the residential fabric that offers privacy and shelter and
the commercial entities that offer the competitive and individual enter-
prises making up the vitality of a culture. On the opposite side, the pub-
lic realm is made up of capital buildings, town halls, libraries, schools,
and other public institutions that both celebrate the accomplishments
of a culture and facilitate its governance, education, and broader cul-
tural aspirations.
With respect to buildings, walls give us privacy and identity and
openings modulate the relationship of a space to the outside world.
 The Architecture School within the University 293

Thus, on one end of the spectrum, bathrooms and kitchens have small
windows, and on the other, living rooms and libraries can have large
ones. Colonnades and arcades, which can be seen as walls—with as
much of the material removed as possible—make the most porous con-
nection between the public domain, such as streets, and the private or
semipublic spaces behind them. Together, at the scales of both urban-
ism and architecture, the city, with its streets, squares and blocks, walls,
openings, and roofs, facilitates the delicate balance between our public
and private natures— the need for individual aspirations and the com-
munity that fulfills our common purposes.
The principles represented by the traditional city and its architec-
ture are present through time and place in the history of the world. In ar-
chitecture we call those elements types. In the traditional city, types are
categorized according to the scales of urbanism, architecture, and con-
struction. At the urban scale the types are streets, squares, and blocks; at
the architectural scale the types are sacred buildings, public buildings,
and private buildings; and at the scale of construction the types consist
of walls, openings, and roofs. It should be noted that there are cultures
that are outside these typological categories and that there are cultures
that do not use all of the types.
Consider the building types illustrated and discussed by Carroll
William Westfall in the book coauthored with Robert Jan van Pelt, Ar-
chitectural Principles in the Age of Historicism.1 The types are organized
into the three categories: sacred, identified by the Temple and the Tho-
los; public, identified by the Theater and the Reggia or Palace; and pri-
vate, represented by the Domus and Taberna. Westfall bases these types
not on specific architectural forms or functions but, more broadly,
around the religious, civic, and individual purposes found in most cul-
tures around the world.
We are accustomed to think of history as a series of episodic rup-
tures, mostly disconnected, except for explanations of how one period
led to another due to a series of specific and particular events. Thus we
conceive of the Classical, the Gothic, the Renaissance, the Baroque,
and the Neoclassical as unique and separate. That there are cultural and
historical exigencies that separate these times may be true, but what is
more important is the continuities that thread them together.
 294 Michael Lyko udis

The general Cartesian organization of a city or town is there,

whether or not the city is a grid at the local scale. Each intersection of
streets is a visible Cartesian condition that comes together in a T or X
connection. Since early antiquity, the urban tissue has been made up of
streets, squares, and blocks. There is a clear distinction between pub-
lic, private, and sacred buildings irrespective of time, and each is rec-
ognizable as such.
Similarly we consider African, European, Asian, and American
cultures as separate and distinct. This may be the case, but it is also pos-
sible to see the continuities between them as well. I mentioned before
the relationship between the Chinese and Roman cities. Both are ar-
ranged as idealized grids with major and minor axes. The fabric of both
is made up of streets and blocks. How these are composed changes ac-
cording to local custom and the conventions of the times, but that is a
secondary issue when considering the formal continuities.
With respect to architecture we see the presence of certain build-
ing types in both the Chinese and Greco-Roman traditions. We also
can see that the use of building elements such as walls, openings, and
roofs are common to most cultures. Notions of sacred, public, and pri-
vate realms identify buildings in most cultures with monumental ar-
chitecture. We can also look at a parallelism between cultures and rec-
ognize several common building types. While we do not want to make
false attributions of one culture’s influence on another, the formal
continuities suggest that there is something universal that is present
and visible in all cultures.
“Character” has to do with the quality of urban, architectural, and
structural norms that are specific to time and place. Thus the architec-
tural character of the Gothic differs from the Renaissance; the Greco-
Roman differs from the Chinese; and so on. In an age of globalization,
when we are trying to fit all cultures to one industrial model, we can
see how the world is unified on the one hand by typology and, on the
other, how each culture and time differs according to character.
The political philosopher Peter Murphy defined beauty as follows:
“Beauty is the sense that no force can ultimately overwhelm another
force. Beauty is the line that keeps the contending forces apart and it is
the shape that keeps those forces together. The ugly lacks both shape
and line. It is ‘formless.’”2 I would add to that statement that, in gen-
 The Architecture School within the University 295

eral, balance is associated with beauty, while imbalance is not. Classi-

cal architecture relies on the idea that the natural materials used in its
making (usually stone or wood) and the resulting architectural ele-
ments are in static equilibrium. Stone tends to be best in compression
and not in tension, making it ideal for durable bearings. Wood has
some tensile properties, so it is used in making the rafters or trusses of
the roof. Together the stone and wooden assembly of the classical com-
position represents a body at rest. Thus balance and durability are har-
monized; as a building weathers and erodes it stays in structural bal-
ance for a long time. In structures that use materials in tension, once
their structural members have corroded, they fail quickly.
So while we can appreciate and see many works of art or architec-
ture as beautiful that do not fall into the classical or traditional camps,
the issue of imbalance related to sustainability requires a critique of
our current attitudes toward the city and its architecture.
The Industrial Revolution brought humanity a renewed ability to
feed, clothe, and shelter itself. In no small way this was due to the
relatively inexpensive sources of energy, especially oil and other fossil
fuels. According to the U.S. Department of Energy, nearly 70 percent
of our energy goes to the built environment and the transportation
network required to serve suburban sprawl and our cities.3 We have
consumed approximately half of the earth’s proven oil reserves, and
with the fast-growing economies of China and other populous coun-
tries, we are increasing our thirst for fossil fuels. This represents an
imbalance, and it is not sustainable. With the passage of time, this prob-
lem will become more acute. Compounded with the effects of climate
change, the impending confluence of these two crises threatens our
civilizations.
It has been declared by the distinguished architect Demetri Por-
phyrios that “classicism is not a style. . . . [I]t is the philosophy of free
will nurtured by tradition.”4 More specifically, classicism, in addition
to being the idealization and representation of nature, is a philosophy
of cultural conservation and sustainability. In the classical mind, the fu-
ture and the past are part of a continuum. The ever-changing character
of classical and traditional architecture reflects the changes in conven-
tions over time. With each new generation, the classical architectural
and urban elements of continuity remind us of our debt to the past as
 296 Michael Lyko udis

a conscious act of conservation rather than consumption. We could

think of tradition as the projection of society’s highest aspirations into
the future, thus ensuring that the best, and perhaps the most sustainable,
aspects of a culture endure. Tradition is therefore not duplication but
rather a process that is always inventing itself. It is the inventive quality
of tradition that allows each generation to shape the future in its own
manner, and it is tradition’s projection of the past that provides the
sense of stewardship that is required for sustainability.
Tradition is largely based on the artistic principle of imitation. Imi-
tation in this case is not the servile copy but rather a continuous reinter-
pretation of a thing or an idea. Consider two monasteries: Mount Saint
Michael in England and Mont Saint Michel in Normandy. The English
version is an imitation of the French one. Imitation is how we learn and
how we invent. It is a process based on our nature and in nature.
The same is true of seemingly dissimilar objects in nature. Changes
in nature are continuous but are always based on some precedent.
Something is always becoming something else. Nature imitates itself
as we as part of nature develop our cultures and civilizations through
the process of imitation.
In the concept of imitation lies the role of myth as truth. We know
something is not true, such as the Aesop’s fable in which a grasshop-
per is speaking with an ant. We know ants and grasshoppers cannot
speak, but the parable contains a truth within the fiction. We call this
the fictive distance that art possesses that elevates representation to the
level of art. In architecture it is the mythic origins of classicism and
how they are represented that concern us.
The traditional city as a work of art contains many such truths,
myths, and fictive representations. We can imagine in Paris the origins
of its streets in the animal trails of the primordial forests that preceded
them and the origin of the squares as clearings in that forest. Of course
we know that is not true specifically for any particular city, but the
mythic beginnings of the city are not far off. With respect to buildings,
we can imagine the interiors of the Bibliothèque Nationale in Paris as
an imitation of the canopy of trees in a forest.
In the city of Epidauros in Greece, the cavea is shaped by the val-
ley that contains it, almost improving upon nature, and by the need
for humans to gather around a performance that allows our aspira-
 The Architecture School within the University 297

tions to take flight. Architectural form here imitates both physical na-
ture and human nature.
Nature informs our construction, our knowledge and wisdom,
elevates shelter to craft, while myth elevates building to architecture.
Consider the relationship of the rustic hut to the vernacular building
and then to the classical. We see how the materials in the rustic hut are
used raw, without refinement, to make shelter. We see the vernacular
materials, more processed and refined, and in the classical condition
the craft being transformed into a language.
For example, the mythic origins of classical architecture can be
seen as the representation of the building’s making. The column flutes
are actually bundled tree trunks, the triglyphs are the ends of beams,
the metopes are the spaces between those beams, and the moldings are
ropes and other fasteners that mediate and connect the column shafts
and entablatures. Looking at a Corinthian base, we see the allusion to
ropes tied together to keep the stresses of the building from pulling
the column apart. We also see how classicism expresses principles of
physics, deforming the material after pressure is exerted on it and cre-
ating the molding known as a taurus. Similarly, the wall and its open-
ings, once examples of a simple vernacular construction, are elevated
to a classical language through idealization, representation, and imita-
tion. Roofs give protection to the building and depending on the cul-
ture contain their own narrative or myths.
Recently America marked the hundredth anniversary of the sinking
of the Titanic. That event is a parable of the collapse of a culture that did
not heed the warnings that were visible before the fall. We have built
such a system; its momentum cannot be changed easily. The principles
of the traditional city and its architecture can help restore the balance in
how we live together and inhabit our planet. There are other disciplines
that have to play their role in this effort, but how we live together and
how we build—urbanism and architecture—should help account for
much of the solution in the coming age of austerity and energy limits.
If the traditional city and its architecture embody the idea of the
unity of knowledge, what then about the modern city? It is clear that we
are better off today than we were in 1350 for all sorts of reasons, but we
have also built more buildings since 1950 than in the entire history of hu-
manity. Why then are the cities and buildings we have built over the past
 298 Michael Lyko udis

sixty years so much poorer than those of the past? Before World War II,
one had to travel far to find a bad building. Now one has to travel far
to find a good one. On the one hand, the great modern masters make
impressive buildings, sometimes very beautiful buildings, but in an age
when we are trying to find ways of living today without sacrificing the
opportunity for future generations to reach their potential, it seems a
pity to waste resources on something that is a consumable object of en-
tertainment rather than an investment in an economy of conservation.
As the world needs to accommodate a population of nine billion
or more in the coming decades or centuries, it will be impossible to
live like we did in antiquity or even close to modern times. Much as we
might like to, we cannot replicate the eighteenth century or any other
earlier period in history. But if we see history as a continuum, we can
learn from the accumulated knowledge of the traditional city: regional
and local economies and methods of building; buildings and cities built
in accordance with the limits of their respective climates, geology, flora,
and fauna; denser, more compact cities with mixed use that enable most
of life’s needs to be accommodated on a pedestrian scale; more durable
buildings that can be adapted successfully to new uses; and buildings
built of local materials that have low embodied energy and are durable
for centuries.
So now we come full circle to examine the city as the place where
the human search for truth takes place and the unity of knowledge is
there for all to see. While the modern experiment, with its promise of
limitless and cheap energy, is showing signs of weathering, its contri-
butions are not lost but folded back into a culture of tradition. Because
of it, we managed to make better institutions and political economies.
Because of it, we see the world as a whole made up of many parts, and
within each part we see a piece of ourselves. We are unified as a planet
but celebrate our individual cultural identities. While this utopia may
seem far off, we may be closer to at least part of it than we think.

N ot e s

1. Robert Jan van Pelt and Carroll William Westfall, Architectural Prin-
ciples in the Age of Historicism (New Haven: Yale University Press, 1991).
 The Architecture School within the University 299

2. Peter Murphy, “The City of Justice,” in Building Cities, ed. Norman

Crowe, Richard Economakis, and Michael Lykoudis (London: Artmedia
Press, 1999), 24.
3. U.S. Department of Energy, Buildings Energy Data Book, 1:1.3. 2011.
http://buildingsdatabook.eren.doe.gov/TableView.aspx?table=1.1.3.
4. Demetri Porphyrios, Foreword to Crowe, Economakis, and Lyk-
oudis, Building Cities, 7.
 C h a p t e r F o u r t e e n

How Is Theology Inspired by the Sciences?

-
Celia Deane- Drum m o nd

This essay aims to set out a tentative map of some of the different ways
in which theology might be inspired by the sciences. I am going to limit
my discussion to the biological sciences, having once been a practicing
biologist. I am also going to limit the range of theological ideas to those
that I consider most relevant in relation to the explicit scientific ex-
amples and practices. The idea that theology might have something to
say in relation to the sciences is certainly not new, and Thomas Aquinas’s
position on theology as “queen of the sciences” should be seen in rela-
tion to the perceived unity of knowledge in the premodern era.1
In a post-Enlightenment world, it becomes necessary for theology
to be rather more self-conscious in its engagement with the sciences,
in that while theology draws on philosophical traditions, those tradi-
tions are distinct from those that inform the sciences, with theology
committed to metaphysical theism as a starting point, while science gen-
erally eschews theism in favor of methodological naturalism. Aquinas
recognized that, compared to other sciences, theology was of a differ-
ent kind, but evidence or principles taken from revelation by faith were
not as problematic then as a basis for an argument. However, the kind
of accommodation between theology and science that Aquinas man-

300
 How Is Theology Inspired by the Sciences? 301

aged to achieve still places science in the service of theology, so that

natural reason serves to inform knowledge achieved through revela-
tion. This is not so much “natural theology” understood as science
serving a theistic apologetics; instead, he argues that a mind trained by
the natural and theoretical sciences is better able to perceive revealed
knowledge. In the first book of Summa Theologiae, Q.1.5, for example,
Aquinas considers whether theology is superior to other sciences in
light of the criticism that theology should be less certain as it is based
on faith rather than on other sciences “the premises of which are indu-
bitable.” This is just the kind of argument that contemporary new
atheists make in claiming the superiority of scientific knowledge, a
topic that I return to below. But instead Aquinas argues that theology
is superior in terms of certitude because the other sciences are able to
make mistakes, since they are based on human reason, while theology
is held in the light of divine knowledge. He also argues that the worth
of theology’s subject matter is superior since “it leads to heights that
reason cannot climb.” Theology, therefore, for Aquinas, does not so
much rely on other sciences as use their guidance in order to help weak-
ness of understanding in matters considered above reason by being me-
diated “through the world of natural reason from which the other sci-
ences take their course.”2 This is, therefore, an argument that theology,
even in matters of revealed knowledge, is actually enabled to perform its
task better as a result of engagement with natural sciences. He also be-
lieved that such a view applies to both theoretical and practical issues;
that is, theology is relevant to practical considerations of ethics.
The post-Enlightenment fracturing of the sciences and their dis-
location from a sense of historical rootedness in Christian tradition,
which permitted the burgeoning of experimental science, makes such an
approach harder to sustain in a contemporary context but, I argue, not
necessarily impossible, given certain qualifications. Certainly, we need
to be wary of claims that are made by either science or theology without
proper analysis of their presuppositions and methodological starting
points. Alasdair MacIntyre, for example, in his Three Rival Methods of
Moral Inquiry, points to the different styles of authority in the medieval
writers who assumed the authority of texts, and whose interpretations
were tested through disputation, through to the nineteenth-century
 302 Celia Deane-Drummond

encyclopedic positions, where authority lies in the individual, through

to deconstruction in the wake of the critique of Nietzsche, whose search
for the “truth” finally led him to resign his university post.3 Nietzsche
believed that an unrecognized motivation and unacknowledged pur-
pose led to blindness about the multiplicity of possible perspectives
in a manner that led MacIntyre to characterize his position as “gene-
alogical.” Nietzsche’s claim that “truths are illusions that we have for-
gotten are illusions, worn out metaphors now impotent to stir the
senses, coins which have lost their faces and are considered now as metal
rather than currency,” was highly controversial at the time but, ar-
guably, left a profound mark on the intellectual landscape, particularly
in the humanities.4
In light of the above, I suggest that as a prolegomena to any discus-
sion of unity of knowledge there needs to be an analysis of tensions in
the contemporary debate over what might be called alternative “truth
claims” in theology and science. Such an analysis comes before laying
out a map of common elements. I suggest that these tensions in them-
selves are not fruitless for the discussion but serve to inspire further
research in theology and arguably the sciences as well. My premise is
that, notwithstanding MacIntyre’s helpful typology of the different
ways of knowing and inquiry, it is possible to seek common under-
standing while acknowledging different presuppositions and starting
points. It therefore remains a self-conscious search and aspiration for
unity of knowledge rather than the bolder claim that such unity can
easily be achieved. My argument in this essay is that a shared search for
commonality between what seem to be highly disparate traditions is
best grounded in practical considerations of the practices of science and
the practices of religion, or their practical concrete focus recognized as
specific social goals.
If an attempt to search for a unity of knowledge comes through
human experiences of the practice of science and religion, then wonder
experienced through science or scientific discoveries may border on
an implicit religious experience, pointing to a common language. This
notion informs the second section of this essay. If searching for unity
of knowledge is mediated in shared practical concerns through com-
mon challenges to the global community, then it follows that both
 How Is Theology Inspired by the Sciences? 303

theology and science may be able to contribute to a shared goal. For

example, ecological sensitivity has provoked many theologians to ex-
press their ideas in ways that make interrelationships between the natu-
ral world and theology much more explicit. “Ecotheology” has grown
and developed into a field in its own right, and those adhering to this
view range from the most conservative to the most liberal from a the-
ological perspective. In the context of the ethical challenges arising out
of the theory and practice of science, particularly the biosciences, the-
ology is inspired to engage with specific scientific aspects of human
societies. Sustainability is perhaps a good example of a nexus of highly
complex problems where the ethical task must be multidisciplinary,
and the search for common understanding should not be abandoned
because it seems too difficult or complex. Finally, if the search for a
unity of knowledge is practiced in an educational setting, then theology
can be provoked by the natural and social sciences to find its place in
university education more broadly, so that the particular agenda of any
one of the sciences does not take hold of the student imagination. I
want to suggest that the language of wisdom provides one way in which
theology and the sciences can be thought of as having mutual but dis-
tinct roles to play in relation to building up their mutually enriching
conceptions of truth. These conceptions are not illusions, as Nietzsche
claimed, but they are situated in particular traditions built up through
particular practices.

Fi ndi ng Th e o l o gy b e t we e n Two Cultures

C. P. Snow imagined two cultures, humanities and science, where the

gap between them poses a huge problem for communication between
different disciplines.5 After years of dialogue between science and reli-
gion, what has become of the two cultures? Has Snow’s thesis been
forgotten in the haste to find common ground and build bridges be-
tween various disciplines? Historically, this was posed as the tension
between Jewish and Hellenistic traditions in Christianity; as the early
church historian Tertullian notoriously asked, “What can Athens have
to do with Jerusalem?”6 The evolutionary biologist Stephen Jay Gould
 304 Celia Deane-Drummond

certainly imagined science and religion as two nonoverlapping Magis-

teria, but in his final works he seemed to point toward some common
ground.7 I think we need to take this gap seriously even as we seek to
build dialogue and find forms of unity of knowledge that are still
meaningful for scientists and for those in the humanities.
Evolutionary biologists, for example, have a particular kind of
questioning that seeks explanations according to Darwinian concepts
of evolution by natural selection. When it comes to a discussion of the
origins and practices of religion, an evolutionary biologist will most
commonly try to “explain” religion (or, ironically, perhaps, its oppo-
site, atheism) in evolutionary terms.8 Such explanations presuppose the
Darwinian model, and religion presents a human behavioral puzzle to
be “explained” in evolutionary terms.9 Perhaps on the one hand reli-
gion is carried along by other traits that have selective advantages now
but were not particularly beneficial in the past, known as exaptation.10
On the other hand, religion may appear as a spandrel that lacks selec-
tive advantage both historically and in the present.11 For the biologist
Nikolaas Tinbergen, the “four causes”— ultimate, proximate, devel-
opmental and phylogenetic—of evolutionary explanation may be more
or less dominant, but the “ultimate” is the evolutionary one assumed
to be lurking in the background of other explanations.12 However, in
such discussions speech about “ultimate” explanations in evolution-
ary terms is surprising for theologians, who consider, following Diet-
rich Bonhoeffer’s categorization that echoes to some extent Aquinas’s
views discussed above, that all explanations based on natural science
are penultimate rather than ultimate, since ultimate explanations are
about God rather than human reasoning.13
This example illustrates that evolutionary biologists draw on key
narratives in order to express meaning, so when they use “ultimate
language,” this creates in their listeners a symbolic world, but that is
not necessarily what biologists are attempting to claim. Some scien-
tists are beginning to object to the use of ultimate language as it seems
to not only promise more than it can deliver but also, perhaps, detract
from important issues that are in danger of being ignored.14 An evo-
lutionary biologist’s reply may be that language about God has itself
evolved for specific purposes, hence the use of the “supernatural pun-
ishment” device and other tools to study religion.15 One of the dif-
 How Is Theology Inspired by the Sciences? 305

ficulties with this language is that what many evolutionary biologists

are trying to do is to find a way to operationalize and test religious be-
lief in a particular way according to particular traits that can then be
studied from the perspective of evolutionary biology. Yet such a posi-
tion tends to miss the other ways in which a phenomenon as complex
as religion might be analyzed even within the sciences, including, for
example, neuroscience and cognitive psychology, as well as “softer”
sciences such as linguistics, social science, anthropology, and so on.
Further, as I have hinted at already, even evolutionary biological sci-
ence is grounded in its own cultural world, a world that biologists
often seem not to recognize as important, both in terms of informing
its explanations of reality and in terms of the cultural factors that make
up human experience. In other words, in isolating biological factors as
a means of analysis and explanation, other possible sources of vari-
ation tend to be pushed to one side or dismissed.
What are the philosophical assumptions in making claims about
religious beliefs understood through scientific methods? Naturalism is
the term sometimes used to capture the presupposition that particular
human behaviors can be “explained” (some might say “reduced”) to
categories based on scientific methodology. This includes Darwinian
or other evolutionary explanations such as genetic drift. Those theolo-
gians who are rather more forthcoming about their philosophical pre-
suppositions recognize that metaphysical claims about God do not “fit
into” scientific patterns of explanation that presuppose naturalistic
interpretations. So can theological “God language” ever be analyzed
according to the methodologies of natural science if, by definition, it
is about second-order metaphysical claims about God that cannot be
tested? Is it ever meaningful to say such second-order claims have
“evolved,” or does “evolution” mean something very different from
Darwinian forms of evolution by natural selection? Could scientists
simply “adopt” evolutionary language about God for their purposes,
as a heuristic tool? If they do so, what does this do to public perception
about science, or theology for that matter? And what might be the
ethical implications of making such a shift?
Once specific issues are discussed around a common theme, such
as that of evolution and human nature, for example, gaps open up that
relate to the specifically theological claims about what it is to be human:
 306 Celia Deane-Drummond

talk of soul, image of God, and so on. If theologians start to use scien-
tific language in their more constructive theologizing, is this for the
purpose of new theological reflections? In other words, does it matter
that the starting point (gap) is there? Above all, I suggest we need to
be completely honest about what we are doing and recognize the pre-
suppositions in our decisions about which science and which theology
to use in speaking about the relationship between them. Hence if
theology is to be inspired by the sciences, then we need to question
which particular science we are drawing on and why. Just as MacIn-
tyre asked, Whose justice? Which rationality?,16 we can add, Which
science? Whose theory? Whose theology? What fragments remain after
what might be termed “the univocal” language of some sciences, or
theologies for that matter? Or, to put this in much more general terms,
who has authority, and whose voice do we listen to and why?
Scientists, like theologians, in their theoretical reflections often
seek unifying frameworks on the premise that there is an underlying
truth about reality that can be discovered. In the evolutionary world
this often boils down to evolution by natural selection as a foundational
stance. But such a “unifying” framework echoes some of the dangers
that Karl Rahner picks up in his discussion of theology and natural sci-
ence.17 For him, theology has a better chance than another branch of
science of acting as mediator between disciplines, as it is itself a conglom-
erate of ideas, drawing on history, philosophy, and so on. Of course,
there may be sciences that Rahner acknowledged, such as anthro-
pology, but where he too readily assumed that secular anthropology
was a disguised form of theology.18 In the case of anthropology, given
its plural disciplinary basis, there is arguably rather less tendency to
take over other disciplines or subsume them under a particular domi-
nating discipline in the way Rahner feared, but forcing one type of ex-
planation into another is a view that emerges in E. O. Wilson’s Con-
silience, for example.19
But there are also dangers in Rahner’s elevation of a specific form
of Christian theology, in his case, Roman Catholic theology, dangers
that may be equally problematic compared to the elevation (reifica-
tion) of, for example, evolution by natural selection in the biological
realm. How can we navigate the boundary and allow theology to be
 How Is Theology Inspired by the Sciences? 307

inspired by the sciences without absorbing one framework into an-

other? But the challenge remains: in transdisciplinary conversations
how can we be true to our own disciplines (whatever they might be)
without taking over others? In light of the above I suggest three pos-
sible strategies. First, find common ground in the experience of doing
science and the experience of practicing faith through metareflection
on “wonder.” Second, find a common task to which theology and sci-
ence might contribute. As an example, I refer here to global questions
of sustainability, where theological and scientific insights are relevant.
Finally, theology and science can come together and energize one an-
other in formulation of strategies for higher education, where both the-
ology and science need to be given a place at the table in developing
wisdom for the twenty-first century.

Jo urne y i nto Wonder

That science might be inspired by religious or theological insights is

well known in the history of science. We can consider, for example, the
natural theology of the early thirteenth century, which moved from
merely a contemplative activity to one that was imbued with an effort
to understand how and why creation worked the way it did. Such natu-
ral theologies persisted to the eighteenth century, so that in the botani-
cal realm William Paley’s Natural Theology or John Ray’s The Wis-
dom of God was intent on finding evidence of how God thought in
the workings of the natural world.20 Charles Darwin was apparently
inspired by such natural theologies while a student at Cambridge, but
his growing confidence in the explanatory powers of the scientific ac-
count left any theistic underpinning somewhat fragile.21 In the secular
climate today, the question becomes inverted, so that it now asks how
theology might be inspired by the sciences, instead of how can science
be inspired by theology. Indeed, some might consider that this ques-
tion can no longer be asked, as the motivated quest for pure scientific
discoveries as representing truth about the world tends to give way
to pragmatic technological solutions to problems within the world.
But if we hold onto the ideal of science as a search for truth about the
 308 Celia Deane-Drummond

world for a moment, then a common desire underlying such a search is

one motivated by wonder at the complexity, beauty, and diversity of the
world in which we live. Richard Dawkins, well known for his hostile
representation of theism, acknowledges, in his book Unweaving the
Rainbow, the wonder that arises from scientific discovery. He claims:

The impulses to awe, reverence and wonder which led Blake to

mysticism (and lesser figures to paranormal superstition, as we
shall see) are precisely those that lead others to science. Our inter-
pretation is different but what excites us is the same. The mystic is
content to bask in the wonder and revel in the mystery that we
were not “meant” to understand. The scientist feels the same won-
der, but is restless, not content, recognizes the mystery as pro-
found, then adds, “But we are working on it.”22

Of course, he is quite wrong here to suppose that such a religious view,

which he represents as stereotypical mysticism, is the only religious
view possible in relation to a natural phenomenon. A scientist can just
as easily be motivated to explore reasons why the mystery is there
for explicitly religious reasons as for reasons of pure curiosity. Here
we can think of wonder as a kind of preliminary elative experience that
joins with intellectual curiosity to drive the search for the truth behind
particular natural phenomena.
Is such an experience of wonder in the scientific quest at all analo-
gous to specific religious experiences? The answer to this question is
somewhat ambiguous and depends to a large extent on how far and to
what extent natural theology is acceptable as a way of thinking theo-
logically. By way of clarification, natural theology is most often un-
derstood as a methodological approach, a way of moving toward an
understanding of God from close observation of the natural world.
There are present-day examples of this position, such as Paul Davis’s
reflection on modern, contemporary cosmology, where, for him, the
precise laws of our own universe point to the Mind of God.23 This is a
good example, as it shows up the limitations of such a method, for am
I really inclined to believe in or be in a relationship with a God who is
represented simply by such mathematical laws?
 How Is Theology Inspired by the Sciences? 309

Those who hesitate to attach themselves to natural theology are

also aware of the difficulties of bringing together belief in a good God
with the apparently cruel and wasteful processes found in the natural
world. This is particularly true for biologists, which may be one reason
that theism is less common among prominent contemporary biologists
compared to physicists. There seems no need for the God hypothesis,
and, if anything, religious views are thought of as evolved capacities,
selected either directly during the process of hominid evolution or in-
directly as a result of other evolutionary advantages such as agent
detection capabilities that were then conveniently transferred to be-
lief in gods. More hostile elements associated with biology simmer
here as well, not least by philosophers of biology, such as Christopher
Hitchens, Sam Harris, Richard Dawkins, and Daniel Dennett, who ad-
vance rather more aggressive versions of atheism along with their bio-
logical explanations.24
Given this trend, it is hardly surprising that, following Karl Barth’s
famous “Nein” to his fellow theologian Emil Brunner, many Protes-
tant theologians, at least, are wary of most forms of natural theology.25
Such a skeptical view becomes reinforced by historical, cultural, and
racial problems associated with linking belief in a given nature with
religious perfection, such as in the blood and soil mentality of Nazi
Germany. Accordingly, the revealed theology that puts the most em-
phasis on the revelation of God in the Bible, and more specifically in
Jesus Christ, takes precedent. But then another problem arises, namely,
how and in what sense might there be any relationship between revealed
theology and the sciences? Can “natural theology” be dispensed with
so easily?26
In order to address this question, it is a good idea to consider the
way in which early theologians developed forms of natural theology
rather than take examples simply from contemporary science or dis-
torted versions of natural theology in cultural history. A good place to
start is with the Franciscan writers who developed a tradition of natural
philosophy. Saint Bonaventure, a Franciscan writing in 1259, claimed:

The supreme power, wisdom and goodness of the Creator shine

forth in created things in so far as the bodily senses inform the
 310 Celia Deane-Drummond

interior senses. . . . In the first way of seeing, the observer con-

siders things in themselves. . . . [ T]he observer can rise, as from a
vestige, to the knowledge of the immense power, wisdom and
goodness of the Creator. In the second way of seeing, the way of
faith[,] . . . we understand that the world was fashioned by the
Word of God.27

Although Bonaventure believed that the Word of God as revealed in

Jesus Christ is “superior” compared to the vestiges of God found in
the natural world, for him, God as Creator was a cosmic presence,
since God is the Creator of all that is.
It would be a mistake, therefore, to think that this form of natural
theology was a way of finding God in the natural world somehow in-
dependent of faith in God, or that those who were habitual sinners
could somehow just “see” God in the natural order of things without
prior experience of God in prayer. Bonaventure also seems to go fur-
ther than this in suggesting that only those who are acting out their
faith through actions of justice and only those who already have some
knowledge of God through intense meditation can begin the journey
and see with a pure heart those vestiges of the wisdom of God in the
creaturely world. Given that the believer could reach such heights of
contemplative grace, it is hardly surprising that such a view, at first
sight, appears antithetical to experimental science. But it is significant
that Bonaventure also claimed we need to move beyond mere vestiges,
for in creatures “He is present in them by his essence, His power and
His presence.”28
Bonaventure, heavily influenced by Augustine in his Platonic de-
scriptions of creatures as “shadows” of that perfect wisdom found in
God, also encouraged careful understanding of the truth, including the
truth that could come from scientific activity. Such scientific activity is
in his case put to a particular goal, namely, the goal of mystical union
with God. He believed that the contemplation of the insights of the
various sciences takes place in charity. In other words, all knowledge
and human endeavor is instilled with the spirit of love. He links various
aspects of human endeavor together: reading with fervor, speculation
with devotion, investigation with admiration, observation with exulta-
 How Is Theology Inspired by the Sciences? 311

tion, industry with piety, knowledge with love, understanding with

humility, study with grace, and, finally, the mirror with wisdom.29

E co l o gical Praxis

If, as Bonaventure suggests, wonder is a common language through

which believing scientists express a kind of implicit religious experi-
ence, filled out by closer attention to natural theology, then a more di-
rect inspiration of theology from natural science comes from practical
concerns about environmental issues and ecology. Ecology as science
can be distinguished from ecology as a social and political movement,
and ecotheology draws inspiration from all elements associated with
ecology, not just the scientific ones. Nonetheless, scientific accounts
about the natural world and close, detailed attention to the behavior of
other species act as sources of inspiration and motivation. Different
theologians will focus on different aspects of the problem and draw on
different elements of the sciences involved, ranging from climate sci-
ence to discussions about animal suffering, biodiversity loss, and so
on. The inspiration for such a reflection has as much to do with the
wonder in that biodiversity as a need to address important ethical is-
sues that science on its own does not seem able to solve. Ecotheology
forces theological reflection away from the temptation to be narrowly
anthropocentric, that is, focusing on humanity in a way that excludes
consideration of other creatures and ecological contexts. The variety
of forms that ecotheology might take are far too extensive to discuss
in any detail here, but it is significant inasmuch as it embraces a huge
range of methodological starting points, from traditional Eastern Or-
thodoxy to ecofeminism. Eastern Orthodox engagement with ecology
retains a strong focus on the Logos (Word) as revealed in logoi (words)
of the natural world and the place of human beings as microcosm and
mediator between heaven and earth. By putting emphasis on asceti-
cism and liturgical practices, it seeks to address the fundamental prob-
lems associated with a lack of sustainability, namely, the human ten-
dency to take more than is required for a flourishing life and to define
that flourishing in material rather than in spiritual terms.
 312 Celia Deane-Drummond

The ecumenical patriarchate of Constantinople, Bartholomew II,

also known in the media as the “Green Patriarch,” has been active in
promoting environmental responsibility throughout the world. While
the extent of environmental damage as shown in scientific study in-
spires such activism, the underlying motivation is one that seeks to
turn human beings away from selfish desires to live more responsibly
in communion with God and other creatures. Symposia and confer-
ences repeat the need for changes in practice and human responsibility,
as the opening summit of the June 2012 conference in Halki indicated.

Permit us to propose that perhaps the reason for this hesitation and
hindrance may lie in the fact that we are unwilling to accept per-
sonal responsibility and demonstrate personal sacrifice. In the Or-
thodox Christian tradition, we refer to this “missing dimension” as
ascesis, which could be translated as abstinence and moderation,
or—better still— simplicity and frugality. The truth is that we re-
sist any demand for self-restraint and self-control. However, dear
friends, if we do not live more simply, we cannot learn to share.
And if we do not learn to share, then how can we expect to survive?
This may be a fundamental religious and spiritual value. Yet it is
also a fundamental ethical and existential principle.30

Bartholemew claims, then, that a weakened sense of social responsi-

bility stems from a lost ascetic spirituality. Roman Catholic social teach-
ing also issues a similar call to environmental responsibility; in the writ-
ings of Pope John Paul II, we find a modification of the term human
ecology used in social science to describe, as in the Orthodox Church, a
way of emphasizing the importance of recognizing humanity’s place
within a wider ecological network of relationships.31 The pope also
used the term ecological conversion, referring to an understanding that
all of creation is in one sense linked into a cosmic Christology.32 While
the idea of a cosmic liturgy goes back to the early church, for example,
Maximus the Confessor, recovering this thread in contemporary dis-
cussion gives ecological language a significant place: an association
with what is arguably the heart of faith in Jesus Christ. The Word made
flesh becomes echoed in the logoi of creation. My own preference is
 How Is Theology Inspired by the Sciences? 313

to use the language of wisdom here: Christ as the Wisdom of God is

found as in a mirror in the wisdom of the creaturely world.
My preference for wisdom relates to one aspect that ecological
study consistently shows, namely, the suffering, predation, and death
of myriad creatures, both in present ecological systems and recorded
throughout evolutionary history. Some theologians have argued that
we need a new kind of theodicy as a result of such insights.33 Such limi-
tations are important to recognize in order to remind ourselves that
the temptation to hubris is always present, not least in the manner in
which we might presume to be able to solve the difficult global prob-
lems of sustainability and what this might mean in terms of balancing
different pressures in a truly human ecology, one that aims for what
Pope Benedict XVI and his predecessors have called authentic or inte-
gral human development.34
Sustainability, therefore, in its complex demand on human soci-
eties, reaches beyond what might be termed a “univocal” response
shaped by specific branches of modern science, if that science is thought
of as cutting out aspects of human experience. Here, not just theology
is inspired by the sciences, but different sciences come together with
other social and human disciplines in order to seek real solutions to
practical problems. Sustainability, therefore, when it is defined as in-
clusive of creatures other than human beings, represents what might
be termed ecological praxis directed toward a particular end or goal,
namely, the flourishing of the earth as such. Although science usually
tries to avoid teleological language, it is clear that the agenda of sus-
tainability is directed toward future generations as well as present ones.
Part of the difficulty is knowing how to balance different demands,
and in this context theology contributes to the discussion alongside sci-
ence. I am convinced that the fruit of such interdisciplinary practices
goes beyond that of respective disciplines and is generative of questions
and their possible resolution in ways that would not be the case if we
just kept to our comfortable watertight areas of specialization.
We need that aspect of prudence that Thomas Aquinas termed
“docilitas”: the openness to learn, not simply in relation to our own
boundary of unknowing, but in relation to other scholars who join
with us to explore this important complex of boundaries.35 Prudence,
 314 Celia Deane-Drummond

or practical wisdom, is also, significantly, an intellectual virtue of prac-

tical reason, one that insists on right action following deliberation and
judgment. But other elements of prudence are important as well. In
the cognitive area, not only teachability but also memoria and solertia
are important.36 Part of the difficulty of ecological restoration, for ex-
ample, is in knowing which particular historical period to go back to
in order to construct a new landscape when its predecessor no longer
exists as a result of human activities.37 Memory that is “true to being”
will consider the social and historical aspects of ecological restoration
in sustainable projects and not simply an arbitrary historical baseline.
Biologists can try to discern what an ecological system may have been
like in the past, but they are not in a position to adjudicate which his-
torical period is the one that should be aimed at today. This is an ethi-
cal question that requires prudential judgment. Solertia, or the ability
to act right in the face of the unexpected, is also a skill that is not
necessarily commensurate with scientific prowess. Given the number
of environmental disasters that are also profoundly human tragedies
highlighted on a repeated basis in the news media, I will name just a
few: Hurricane Katrina on August 29, 2005; the earthquake in Haiti on
January 12, 2010; or Hurricane Sandy on October 29, 2012. All these
events involved a scientific account of what happened, but knowing
how to respond quickly and decisively requires solertia.
I could also go on to name the other aspects of imperative pru-
dence that arguably connect with practices of sustainability, including,
for example, caution, insight, reason, foresight, and circumspection,
which basically means taking account of what is known, in other
words, the insights portrayed in science. Foresight is interesting in this
respect: if we could have accurate foresight about, for example, climate
change, perhaps there would be less intense debate about the level of
scientific uncertainty in climate change models. For Aquinas, fore-
sight is linked with God’s Providence, and therefore it underlies one
important difficulty faced by the modern believer and scientist. Do
the changes that we are experiencing undermine our faith in divine
Providence? Or does Providence give us confidence that, in spite of
human sinfulness, or what might be called in this case anthropogenic
sin, Earth will still somehow be under divine protection? This is too
 How Is Theology Inspired by the Sciences? 315

large an issue to address here; my intention is to show how theological

reflection might be inspired by the uncertainties evident in modern
science as much as their discoveries.

Wi s do m i n the University

My concluding discussion reflects, perhaps, a wider sense that theo-

logical reflection should not just be seen as inspired by the sciences
but as actively making a contribution to the way truth is not only per-
ceived but also realized in practical ways.38 MacIntyre famously de-
scribed the normative principles of justice and rationality as culturally
laden.39 Much the same could be said about conceptions of truth but
in this case drawing on distinct epistemological bases. The fragmenta-
tion and specialization of that knowledge in the modern university has
tended to silence theological or religious discussion as not relevant or
pertinent to university education. The University of Notre Dame is an
important exception to that trend, by naming but not shaming its the-
ological values as central to its mission as a center of excellence in uni-
versity education. In such a context, we can recover, perhaps, a role for
theology that recenters the human enterprise in the search for truth on
the moral and religious claims that prioritize some fields of research
over others. It would be appropriate to name this an exercise in seek-
ing wisdom rather than simply knowledge, where wisdom is that clas-
sical sense of rediscovering the relationship between fields of study
and, in theological terms, inclusive of a relationship with God. What
might that wisdom look like in practice?
In the first place, theological wisdom in the university draws on
education that students have already received in the context of family
and community. In the Hebrew scriptures this way of learning was
practical and contextual long before personified wisdom came to the
fore and prior to contemporary uses of wisdom literature in theologi-
cal discourse. This praxis understood as theory informed by practices
is very different from utilitarian methods, which emphasize usefulness
for its own sake and attempt to control knowledge, detached from
other forms of knowing and contemplation. In the university setting
 316 Celia Deane-Drummond

itself, the context of students’ community life is just as important to

learning as the content of their courses.
Second, theological wisdom is expressed in the Hebrew Bible in
feminine categories. I am confining my discussion of wisdom to Chris-
tian theology, but this should not be taken to imply that I think that
other religions have little to offer to the debates on wisdom. I focus on
arguably the most dominant religious tradition in the Western world
in order to show what insights this tradition might lend to the univer-
sity as an institution for higher education. Christian theology has been
dogged in its history by interpretations of theology that are influenced
by patriarchal societies and assumptions. Such influences also creep
into a university setting in a way that is subliminal if not carefully ex-
amined. Catherine Keller, a leading feminist theologian, has drawn on
the idea of emancipatory wisdom as that which best describes the fu-
ture of theology in the university.40 It is wisdom that can straddle the
world of the academic and ecclesial communities to which theology
must give an account of itself. For Keller, wisdom, “at least as prac-
tised in the indigenous and biblical traditions, is irredeemably impli-
cated in the sensuous, the communal, the experiential, the metanoic,
the unpredictable, the imaginal, the practical.”41 Her use here of “meta-
nioic,” from metanoia, meaning “change of heart,” is significant as
it implies transformation. This differs significantly from the coercive
control of matter by the mind, which is the dominant agenda of mo-
dernity; rather it takes time to “let things become” and includes the
social as well as the cosmological. Theological wisdom, therefore, is
not individualistic but highly relational, operating from within the so-
cial context and reaching out to the natural world as well. It has the ca-
pacity, therefore, to enlarge a student’s horizons to consider issues that
are important not just to the human community, but to the commu-
nity of other creatures. Indeed, based on Proverbs 8, God could be
said to create the world in love but through wisdom.42 Hence wisdom
is a fundamental characteristic of the way God is perceived to create
and sustain the world, perceived as a child at play, ever present with
God at the dawn of existence. Yet such a theological interpretation of
creation in wisdom, although grounded in a different metaphysics, does
not need to be at loggerheads with cosmological and evolutionary ac-
counts of the origin of the Earth inasmuch as they remain material ac-
 How Is Theology Inspired by the Sciences? 317

counts of origins.43 Rather, it adds to such an account a dimension that

fills out an interpretation of human origins in a way that complements
the voice of science. Even many evolutionary biologists are now claim-
ing that being religious is part of what makes human beings human.
Third, a theological voice is one that needs to be heard, for without
it more extreme voices start to force their way into higher education’s
agenda. Such a worrying trend is only too apparent in those who wish
to promote creationism, the belief that the story of Genesis is literally
true and an alternative to the evolutionary account of science. While
creationist science’s voice has become rather more sophisticated through
the notion of intelligent design, it still seeks to provide through ide-
ology an alternative to neo-Darwinian notions of evolutionary science.
Even if many biologists believe that neo-Darwinian theory is insuffi-
cient on its own to explain all aspects of evolution, thus including now
more sophisticated accounts of epigenetics, behavioral inheritance, or
symbolic learning, it is still accepted as a key working paradigm. It is
hardly surprising that given this trend virtually all secular universities in
the United States wish to keep theology out of their agendas. Yet per-
haps it is for this reason that such counterreactions have found their
force. For if people are inculcated into utilitarian methods of learning
and thinking at universities, then a culture that is generally religious
will sense some disorientation and so be more inclined to an equally
narrow reaction to that utilitarianism. In other words, a narrowing of
epistemology through a secularist agenda such as that expressed in uni-
versity education all too easily leads to a counterreaction that is ironi-
cally a very reflection of such narrowness but now expressed in religious
terms. The narrowing of epistemology and counteraction in creation-
ist accounts highlights the importance of a rich understanding of the-
ological wisdom that will discourage such retreats into what seem to
be apparently safe havens.
Fourth, and more radical perhaps, New Testament theological wis-
dom finds expression through the paradox of suffering, rather than
a celebration of human wisdom, in the wisdom of the cross.44 While
not doing away with the wisdom of the sages, the wisdom of the cross
points to another way of being that makes most sense in the context of
the Christian community. Yet could the wisdom of the cross have wider
relevance as well? Certainly, it shows that a Christian image of God is
 318 Celia Deane-Drummond

one that is on the side of those who are suffering and in pain. One of the
important tasks of the university is pastoral; students do not achieve in a
vacuum but are enabled through their lived experiences. If such experi-
ences are too traumatic, learning may suffer, at least temporarily. It is
here that a university needs to include not just a curriculum but also
provision for the pastoral needs of its students through adequate coun-
seling and chaplaincy. Discussions of the wisdom of the cross in the
Epistle to the Corinthians are also set in the context of an early Chris-
tian community in which different groups vied for authority accord-
ing to different perceptions of wisdom. Instead of such rhetorical game
playing, the author of the epistle encourages reflection on the wisdom
of the cross. Such wisdom speaks of the need for humility rather than
jockeying for positions of power through clever forms of speech. Is
such a goal realistic in a university context? What would the shape of
university management be like if such an approach were adopted by
university presidents?
Let me end this essay with a citation from Thomas Aquinas, who
viewed theology as the scientia of sacred doctrine.

In man different objects of knowledge imply different kinds of

knowledge: in knowing principles he is said to have “understand-
ing,” in knowing conclusions “science,” in knowing highest cause
“wisdom,” in knowing human actions “counsel” or “prudence.”
But all these things God knows by one simple knowledge. . . .
Hence God’s simple knowledge may be called by all these names,
provided that in using any of them of God we exclude from their
meaning all that implies imperfection, and retain only what implies
perfection.45

N ote s

This is an expanded version of a paper first presented to the Notre Dame In-
stitute for Advanced Study conference, “Conceptions of Truth and Unity of
Knowledge,” held at the University of Notre Dame, April 12–14, 2012. I
would like to thank the organizers for inviting me to participate and my re-
search assistant, Rebecca Artinian Kaiser, for careful assistance in the final
preparation stages.
 How Is Theology Inspired by the Sciences? 319

1. Aquinas thought of theology as a science in a manner similar to the

way some sciences work from premises established in light of a higher science,
such as optics following from principles established in geometry. For Aquinas,
theology followed from God-given principles. Hence, “Christian theology
should be pronounced to be a science . . . for it flows from founts recognized
in the light of higher science, namely God’s very own which he shares with
the blessed. Hence as harmony credits its principles which are taken from
arithmetic, so Christian theology takes on faith its principles revealed by God.”
Thomas Aquinas, Summa Theologiae, vol. 1, Christian Theology, trans. Thomas
Gilby (Cambridge: Blackfriars, 1963), 1a Q.1.2.
2. Aquinas, Summa Theologiae, 1a Q. 1.5.
3. Alasdair MacIntyre, Three Rival Versions of Moral Inquiry: Encyclo-
pedia, Genealogy and Tradition (Notre Dame: University of Notre Dame
Press, 1990).
4. MacIntyre, Three Rival Versions of Moral Inquiry, 35.
5. C. P. Snow, The Two Cultures (Cambridge: Cambridge University
Press, 1998).
6. This was Tertullian’s famous question regarding the Hellenization of
Christianity which has served as an inspiration for scholars ever since. What in-
deed does Athens have to do with Jerusalem? What concord is there between
the Academy and the Church? Between heretics and Christians? Tertullian, De
praescriptione, Chapter VII; Tertullian, Prescription against Heretics (Kesinger
Publishing, 2004). English translation also available on-line, Prescription against
Heretics, New Advent Translation, www.newadvent.org/fathers/0311.htm.
7. See S. J. Gould: “Nonoverlapping Magisteria,” Natural History 106,
no. 2 (March 1997): 16– 22; Rocks of Ages: Science and Religion in the Fullness
of Life (New York: Ballantine, 2002); The Hedgehog, the Fox and the Magis-
ter’s Pox: Mending the Gap between Science and the Humanities (New York:
Three Rivers Press, 2003).
8. See, e.g., Dominic Johnson, “What Are Atheists For? Hypotheses on
the Function of Non-Belief in the Evolution of Religion,” Religion, Brain and
Behavior 2, no. 1 (2012): 48– 59.
9. C. S. Alcorta and R. Sosis, “Ritual, Emotion, and Sacred Symbols:
The Evolution of Religion as an Adaptive Complex,” Human Nature 16 (2005):
323– 59; P. Boyer, Religion Explained: The Evolutionary Origins of Religious
Thought (New York: Basic Books, 2001); S. Atran, In Gods We Trust: The
Evolutionary Landscape of Religion (Oxford: Oxford University Press, 2004);
J. Bulbulia, “Meme Infection or Religious Niche Construction? An Adapta-
tionist Alternative to the Cultural Maladaptationist Hypothesis,” Method and
Theory in the Study of Religion 20 (2008): 1– 42; J. M. Bering, The God In-
stinct: The Psychology of Souls, Destiny and the Meaning of Life (London:
 320 Celia Deane-Drummond

Nicolas Brealey, 2010); R. Sosis, “The Adaptationist-Byproduct Debate on

the Evolution of Religion: Five Misunderstandings of the Adaptationist Pro-
gram,” Journal of Cognition and Culture 9 (2009): 315– 32.
10. Stephen Jay Gould introduced the term exaptation to explain that
traits may evolve without any particular advantage but then become co-opted
later in a way that enhances evolutionary fitness. S. J. Gould and E. Vrba,
“Exaptation: A Missing Term in the Science of Form,” Paleobiology 8, no. 1
(1982): 4–15.
11. There is an intellectual difficulty with the spandrel concept, which
assumes that characteristics have evolved through natural selection and there-
fore must have attached to something else that is advantageous. This leads bi-
ologists to be more cautious about pressing adaptation too far. For discussion,
see M. Pigliucci and J. M. Kaplan, “The Fall and Rise of Dr Pangloss: Adapta-
tionism and the Spandrels Paper 20 Years Later,” Trends in Ecology and Evo-
lution 15, no. 2 (2000): 66– 70.
12. N. Tinbergen, “On Aims and Methods of Ethology,” Zeitschrift für
Tierpsychologie 20, no. 4 (1963): 410– 33.
13. See “Ultimate and Penultimate Things,” in Dietrich Bonhoeffer, Eth-
ics, ed. C. Green, trans. R. Krauss, C. West, and D. Stott, Dietrich Bonhoeffer
Works 6 (Minneapolis: Fortress Press, 2005), 146– 70.
14. In 1961 Ernst Mayr was the first to use the terms ultimate and proxi-
mate in biology, an idea he continued to develop until 1993. E. Mayr, “Cause
and Effect in Biology,” Science 131 (1961): 1501– 6; and “Proximate and Ulti-
mate Causation,” Biology and Philosophy 8 (1993): 93– 94. By “ultimate,” he
meant evolutionary in a historical sense. See, e.g., Kevin Laland, John Odling-
Smee, William Hoppitt, and Tobias Uller, “More on How and Why: Cause
and Effect in Biology Revisited,” Biology and Philosophy (August 7, 2012),
http://lalandlab.st-andrews.ac.uk/documents/Publication185.pdf.
15. The language of “supernatural punishment” appears in evolutionary
accounts of religious belief, where such tendencies, however they are derived,
are treated as modular psychological “objects” that can be analyzed by natu-
ral selection. See Dominic Johnson, “Why God Is the Best Punisher,” Reli-
gion, Brain and Behavior 1, no. 1 (2011): 77– 84.
16. Alasdair MacIntyre, Whose Justice? Which Rationality? (Notre Dame,
IN: University of Notre Dame Press, 1988).
17. Karl Rahner, Theological Investigations, vol. 13, Theology, Anthro-
pology, Christology, trans. David Bourke (New York: Crossroad, 1983), 80–102.
Rahner also acknowledges Aquinas, but what is interesting is that he consid-
ers him the pioneer of a basically anthropocentric approach to theology. But
he also acknowledged Thomas’s ongoing relevance as one who could think in
 How Is Theology Inspired by the Sciences? 321

a broad way, who was self-critical, who listened to the views of others and paid
attention to points that might at first seem insignificant (5– 7), all of which
might be termed the “virtues” required of a theologian, or of any scholar for
that matter.
18. Karl Rahner, Theological Investigations, vol. 17, Jesus, Man and the
Church, trans. M. Kohl (New York: Crossroad, 1981), 64. So for him, what are
apparently secular anthropologies are “secretly theological assertions,” and the-
ological statements are “the radical form of secular anthropological statements.”
19. E. O. Wilson, Consilience: The Unity of Knowledge (London: Aba-
cus, 2003).
20. William Paley, Natural Theology, or, Evidences of the Existence and
Attributes of the Deity Collected from the Appearances of Nature (Oxford:
Oxford University Press, [1802] 2006); John Ray, The Wisdom of God Mani-
fested in the Works of Creation (1691), www.jri.org.uk/ray/wisdom/.
21. Arguably his growing confidence in the scientific explanation was
not the direct cause of his eventual loss of faith, but other personal factors
such as the death of his child and his own received theological views that in-
terpreted different forms as a result of God’s direct creative activity were hard
to reconcile with his understanding of evolution by natural selection.
22. Richard Dawkins, Unweaving the Rainbow: Science, Delusion and
the Appetite for Wonder (London: Faber and Faber, 1998), 17.
23. Paul Davis, The Mind of God: The Scientific Basis for a Rational World
(New York: Simon & Schuster, 1992).
24. Daniel Dennett, Breaking the Spell (New York: Viking, 2006). In 2007,
Daniel Dennett, Richard Dawkins, Sam Harris, and Christopher Hitchens cre-
ated a video titled Four Horsemen of the Apocalypse, deliberately setting out to
discredit the credibility of religious belief and the religious attacks on their sci-
ence. See www.youtube.com/watch?v=MuyUz2XLp1E.
25. Karl Barth, Nein! Antwort an Emil Brunner (Munich: C. Kaiser, 1934).
26. A question also raised by Sarah Coakley in her recent Gifford Lec-
tures, Sacrifice Regained: Evolution, Cooperation and God, www.abdn.ac.uk
/gifford/. There is a difference in Coakley’s approach, however, since she seems
to argue for an apologetic version of natural theology based on a particular in-
terpretation of the fifth of Aquinas’s Five Ways to God. Coakley rejects, cor-
rectly in my view, pitching theisic teleological arguments against evolutionary
theories, since this smacks of a dualistic competition between God and scien-
tific interpretations and argues, further, that to portray evolution as simply
“random” is far too oversimplified. But she seems to want to not just permit
but also highlight an interpretation of the fifth teleological way as an explicit
argument for God’s existence (alongside other possible interpretations, such as
 322 Celia Deane-Drummond

to bolster existing revelatory faith or bring to consciousness in a believer what

may be hidden from view). But in finding the fifth way lacking in light of the
possibility of a purely naturalistic, evolutionary interpretation, she then pro-
poses an argument for God’s existence through the ecstatic or excessive altru-
ism of the saints, which, she argues, is necessarily and always outside the realm
of possible scientific explanations and therefore not in competition with it. See
Sarah Coakley, Lecture 5, “Teleology Reviewed: A New ‘Ethico-Teleological’
Argument for God’s Existence,” in Sacrifice Regained, 17–19. I am arguing for
something rather more modest, and more in line with Aquinas’s discussion in
the Summa of the relationship between theology and the sciences, namely, a
form of “natural” theologizing that is useful in order to clarify the theological
task but presupposes faith in God. I am not attempting, in other words, to
convince the nonbeliever or return to supposed arguments for God’s exis-
tence, but, like Coakley, I also believe that theology needs to be given a place
in the public sphere.
27. Bonaventure, The Journey of the Mind to God, ed. S. F. Brown, trans.
P. Boehner (Indianapolis: Hackett Publishing, 1993), chap. 1, § 10 –12, p. 8.
For further discussion of Bonaventure, see Celia Deane-Drummond, Wonder
and Wisdom: Conversations in Science, Spirituality and Theology (London:
Darton, Longman and Todd, 2006), 56– 58.
28. Bonaventure, The Journey of the Mind to God, chap. 2, § 1, p. 11.
29. Ibid., Prologue, § 4, p. 2. See also pp. 70 – 71 for commentary.
30. “Keynote Address of His Holiness Ecumenical Patriarch Bartholo-
mew at the Opening Ceremony of the Halki Summit.” Halki Theological
School, June 18, 2012. www.patriarchate.org/documents/2012halkisummit.
31. For a full discussion of this topic, see Celia Deane-Drummond,
“Joining the Dance: Ecology and Catholic Social Teaching,” New Blackfriars
93, no. 1044 (March 2012): 193– 212.
32. John Paul II, “General Audience Address,” January 17, 2001. www
.vatican.va/holy_father/john_paul_ii/audiences/2001/documents/hf_jp-ii_
aud_20010117_en.html.
33. See, e.g., Christopher Southgate, The Groaning of Creation: God,
Evolution, and the Problem of Evil (Louisville, KY: Westminster John Knox
Press, 2008).
34. Pope Benedict XVI, Caritas in Veritate, www.vatican.va/holy_father
/benedict_xvi/encyclicals/documents/hf_ben-xvi_enc_20090629_caritas-in
-veritate_en.html.
35. Aquinas, Summa Theologiae, vol. 36, Prudence, trans. Thomas Gilby
(Cambridge: Blackfriars, 1973), 2a2ae Q. 49.3.
36. For more detailed discussion of the different facets of prudence, see
Celia Deane-Drummond, The Ethics of Nature (Oxford: Blackwell, 2004), 9–15.
 How Is Theology Inspired by the Sciences? 323

37. For more on the practical and historical limitations of ecological res-
toration, see Stephen T. Jackson and Richard J. Hobbs, “Ecological Restoration
in the Light of Ecological History,” Science 325, no. 5940 (2009): 567; Stuart
Allison, “What Do We Mean When We Talk about Ecological Restoration?,”
Ecological Restoration 22, no. 4 (2004): 281– 86.
38. I have discussed this in more detail in Celia Deane-Drummond,
“Wisdom Remembered: Recovering a Theological Vision for Wisdom in the
Academe,” in Wisdom in the University, ed. Ronald Barnett and Nicholas
Maxwell (London: Routledge, 2008), 77– 88; and “The Amnesia of Modern
Universities: An Argument for Theological Wisdom in the Academe,” in Edu-
cating for Wisdom in the Twenty-First Century, ed. Darin Davis (Waco, TX:
Baylor University Press, 2013).
39. MacIntyre, Whose Justice, Which Rationality?
40. C. Keller, “Towards an Emancipatory Wisdom,” in Theology and
the University: Essays in Honor of John B. Cobb Jr, ed. D. R. Griffin and J. C.
Hough (Albany: State University of New York Press, 1991), 125– 47.
41. Keller, “Towards Emancipatory Wisdom,” 143.
42. A full discussion of this is outside the scope of this chapter. For more
detail, see C. Deane-Drummond, Creation through Wisdom: Theology and
the New Biology (Edinburgh: T & T Clark, 2000).
43. Deane-Drummond, Wonder and Wisdom.
44. The first letter of Paul to the Corinthians, e.g., esp. 1 Cor. 1:8– 2:5.
45. Aquinas, Summa Theologiae, vol. 4, Knowledge in God, trans. Thomas
Gornall (Cambridge: Blackfriars, 1964), 1a Q. 14.1.
 A b o u t t h e C o n t r i b u t o r s

Francisco J. Ayala is University Professor and Donald Bren Pro-

fessor of Biological Sciences and Professor of Philosophy at the Uni-
versity of California, Irvine. His research and teaching focuses on evo-
lution, genetics, and the philosophy of biology. Ayala is a member of
the National Academy of Sciences (NAS) and a recipient of the 2001
National Medal of Science. He served as chair of the Authoring Com-
mittee of Science, Evolution, and Creationism, jointly published in 2008
by the NAS and the Institute of Medicine. He has received numerous
awards, including the 2010 Templeton Prize for exceptional contribu-
tion to affirming life’s spiritual dimension, and twenty honorary de-
grees from universities in nine countries. He has been president and
chairman of the board of the American Association for the Advance-
ment of Science and president of Sigma Xi, the Scientific Research So-
ciety of the United States. Ayala is the author of numerous books and
articles on the intersection of science and the humanities, including Dar-
win’s Gift to Science and Religion (2007) and Am I a Monkey? (2010).

Celia Deane-Drummond is Professor of Theology at the Univer-

sity of Notre Dame, with a concurrent appointment in the College of
Science. She is Fellow of the Eck Institute for Global Health and the
John Reilly Center for Science, Technology and Values at Notre Dame.
She holds a PhD in plant physiology as well as a PhD in systematic the-
ology. Her research focuses on the engagement of systematic theology
and the biological sciences, as well as bioethics and environmental eth-
ics. Before coming to the University of Notre Dame in 2011, Deane-
Drummond held a professorial chair in theology and biological sci-
ences at the University of Chester and was director of the Centre for
Religion and the Biosciences, launched in 2002. In addition to more

324
 Contributors 325

than thirty scientific articles, she is the author or editor of twenty-two

books, as well as thirty-three contributions to books and forty-three
articles in areas relating to theology or ethics. Among her most re-
cent books are Creation through Wisdom (2000); Brave New World,
ed. (2003); coeditor, Re-Ordering Nature (2003); The Ethics of Nature
(2004); Wonder and Wisdom (2006); Genetics and Christian Ethics (2006);
coeditor, Future Perfect (2006; 2d ed., 2010); Ecotheology (2008); Christ
and Evolution (2009); coeditor, Creaturely Theology (2009); Seeds of
Hope (2010); editor, Rising to Life (2011); coeditor, Religion and Ecology
in the Public Sphere (2011); coeditor, Animals as Religious Subjects
(2013); The Wisdom of the Liminal (2014). Deane-Drummond served
as chair of the European Forum for the Study of Religion and Envi-
ronment from 2011 to 2015. She was editor of the international journal
Ecotheology (2000– 2006), and she was seconded to the spirituality team
at the Catholic Fund for Overseas Development, working in the area
of environmental justice and climate change (2009–10). She is also co-
editor of the journal Philosophy, Theology and the Sciences, launched
in 2014.

Nicola Di Cosmo is the Henry Luce Foundation Professor of East

Asian History in the School of Historical Studies at the Institute for
Advanced Study. His main areas of research are relations between
China and Central Asia from ancient times to the modern period, the
history of foreign dynasties in China, and, more generally, frontier re-
lations seen from archaeological, anthropological, and historical per-
spectives. Di Cosmo taught at Harvard University and at the Univer-
sity of Canterbury (New Zealand) before joining the faculty of the
School of Historical Studies in 2003. He has written on Inner Asian
history, Chinese historiography, and military history, and he is the au-
thor of several books, including Ancient China and Its Enemies: The
Rise of Nomadic Power in East Asian History (2002), A Documentary
History of Manchu-Mongol Relations (1616 –1626) (2003), and Diary
of a Manchu Soldier in Seventeenth-Century China (2006). Di Cosmo
has also edited or coedited books, including Political Frontiers, Ethnic
Boundaries and Human Geographies in Chinese History (2001), War-
fare in Inner Asian History, 500 –1800 (2002), The Cambridge History
 326 Contributors

of Inner Asia: The Chinggisid Age (2009), and Military Culture in Im-
perial China (2009). He has written numerous book chapters as well as
articles in such publications as the Journal of World History, Interna-
tional History Review, and Central Asiatic Journal. He is on the advi-
sory or editorial boards of the Journal of East Asian Archaeology, Asia
Major, and Inner Asia. Di Cosmo has received fellowships from the
New Zealand Royal Society, Harvard University’s Milton Fund, the
Chiang Ching-kuo Foundation, the Center for Chinese Studies in Tai-
pei, the Institute of Asian and Middle Eastern Studies in Rome, and
the Italian Ministry of Education. In addition, he leads Smithsonian
Study Tours to Mongolia.

Carsten Dutt is Assistant Professor of German at the University of

Notre Dame. His research focuses on the intersection of literary criti-
cism and philosophy. The author or coeditor of several books, Dutt
has published in the fields of hermeneutics (Hermeneutik, Ästhetik,
Praktische Philosophie, 2000, English translation 2001; Gadamers philo-
sophische Hermeneutik und die Literaturwissenschaft, 2012), the theory
and methodology of conceptual history (Herausforderungen der Be-
griffsgeschichte, 2003; Zwischen Sprache und Geschichte, 2013), and
modern German literature (Figurationen der literarischen Moderne,
2007; Die Schuldfrage, 2010; Zur Lyrik Gottfried Benns, 2014). He is
also coeditor of Karl Jaspers’s correspondence (2014, 3 vols.). His cur-
rent book project focuses on the epistemology of literary interpreta-
tion. He received fellowships from the Center for Literary and Cul-
tural Studies in Berlin, the Collegium Budapest, and the Notre Dame
Institute for Advanced Study.

Allan Gibbard is Richard B. Brandt Distinguished University Pro-

fessor of Philosophy at the University of Michigan, Ann Arbor, where
he has taught for three decades. His research focuses on ethical theory,
social choice theory and decision theory, evolutionary moral psy-
chology, philosophy of language, metaphysics, and epistemology. He
is the author of Wise Choices, Apt Feelings (1990), Thinking How to
Live (2003), Reconciling Our Aims (2008), and Meaning and Norma-
tivity (2012). Gibbard is also the author of numerous articles in journals
such as Philosophical Studies, Journal of Philosophical Logic, Econo-
 Contributors 327

metrica, Journal of Philosophy, Social Theory and Practice, Philoso-

phia, Ethics, and Theory and Decision. He has served as president of
the Central Division of the American Philosophical Association, and
he is a member of the National Academy of Sciences, a fellow of the
American Academy of Arts and Sciences, a member of the American
Philosophical Society, a fellow of the Econometric Society, and a mem-
bre titulaire of the Institut International de Philosophie. He has held
research fellowships from the National Endowment for the Humani-
ties, the Guggenheim Foundation, the Rockefeller Foundation, and
the American Council of Learned Societies.

Robert Hanna is Professor of Philosophy at the University of Col-

orado at Boulder. He is a philosophical generalist whose areas of re-
search include Kant’s philosophy and Kantian themes in contemporary
philosophy; the history of analytic philosophy; philosophical logic; the
philosophy of mind, cognition, and action; and ethics. Hanna is the au-
thor of Kant and the Foundations of Analytic Philosophy (2001); Kant,
Science, and Human Nature (2006); Rationality and Logic (2006); with
Michelle Maiese, Embodied Minds in Action (2009); with A. Chapman,
A. Ellis, T. Hildebrand, and H. Pickford, In Defense of Intuitions: A
New Rationalist Manifesto (2013); and Cognition, Content, and the
A Priori (forthcoming). His current book project, The Rational Human
Condition, is an attempt to work out a general theory of human ration-
ality in a fully natural and desperately nonideal world.

Vittorio Hösle is Paul Kimball Professor of Arts and Letters in

the Department of German and Russian Languages and Literatures and
Concurrent Professor of Philosophy and of Political Science at the Uni-
versity of Notre Dame, where he served as founding director of the
NDIAS from 2008 to 2013. In 2013, he was appointed to the Pontifical
Academy of Social Sciences by Pope Francis. Hösle’s scholarly inter-
ests include systematic philosophy (metaphysics, ethics, aesthetics, and
political theory) and history of philosophy (mainly ancient and mod-
ern). He is the author, editor, or coeditor of more than forty books,
which have appeared in twenty languages, including Morals and Poli-
tics (2004) and God as Reason (2011), and more than 140 articles. His
most widely published work (translated into fourteen languages) is The
 328 Contributors

Dead Philosopher’s Café (2000), an exchange of letters with a young

girl that offers an imaginative introduction to the world of philosophy.

Laurent Lafforgue is Professeur Permanent at the Institut des

Hautes Études Scientifiques in Bures-sur-Yvette, France. His research
focuses on number theory and analysis. Lafforgue is the recipient of nu-
merous awards and honors, including the Clay Research Prize in 2000
and the Grand Prix Jacques Herbrand de l’Académie des Sciences in
2001. In 2002, he was awarded the prestigious Fields Medal, the equiva-
lent of a Nobel Prize in the field of mathematics, at the International
Congress of Mathematicians in Beijing for his work related to the Lang-
lands Program in number theory. Lafforgue established the Langlands
conjectures for a much wider class of cases than previously known by
proving the global Langlands correspondence for function fields. In
2003, he was named a member of the French Academy of Sciences and
became Chevalier de la Légion d’Honneur. In addition to his accom-
plishments in mathematics, Lafforgue is actively engaged in the preser-
vation of the French school system from ill-advised reforms. His efforts
include significant support for the teaching of humanities in French sec-
ondary schools. In May 2011, he received an honorary Doctor of Sci-
ence degree from the University of Notre Dame.

Keith Lehrer is Regent’s Professor Emeritus of Philosophy at the

University of Arizona. His research focuses on autonomy, episte-
mology, consensus, Thomas Reid, and self-trust. He is the author of
Knowledge (1974), Rational Consensus in Science and Society: A Philo-
sophical and Mathematical Study (1981, with Carl Wagner), Thomas
Reid (1989), Metamind (1990), Theory of Knowledge (1990), Philosophi-
cal Problems and Arguments: An Introduction (1992, 4th ed., with James
Cornman and George Pappas), and Self Trust: A Study of Reason,
Knowledge and Autonomy (1997). His most recent book, Art, Self and
Knowledge (2012), focuses on the philosophy of art. He is also the edi-
tor of nine books and numerous journal articles. Lehrer, also a painter,
has an interest in choreography and has begun to use one art form to in-
terpret the other. His most recent research objective is to achieve a new
unity of art and epistemology. Lehrer is the recipient of numerous hon-
ors and awards, including his election in 2005 as fellow of the American
 Contributors 329

Academy of Arts and Sciences and his election as fellow at the Center
for Advanced Study in Behavioral Sciences, Stanford (2007– 8). He has
also taught at the University of Miami, Santa Clara University, Stanford
University, and University of Graz in Austria.

Michael Lykoudis is Professor of Architecture and the Francis

and Kathleen Rooney Dean of the School of Architecture at the Uni-
versity of Notre Dame. His research focuses on traditional architec-
ture and urbanism, and he has sought to link architectural tradition and
classicism to urbanism and environmental issues. Lykoudis is coeditor
of two publications, Building Cities (1999) and The Other Modern (ex-
hibition catalogue) (2000) and the author of Modernity, Modernism,
and the Other Modern (forthcoming). He is also the author of numer-
ous articles that have appeared in journals such as Traditional Building.
Prior to joining the Notre Dame faculty, he was a project designer and
architect for firms in Connecticut, Florida, New York, and Greece. He
has directed his own practice since 1983 in Athens, Greece, and Stam-
ford, Connecticut, and now has a practice in South Bend, Indiana.
Lykoudis has organized several major international conferences in
collaboration with the Classical Architecture League, the Institute of
Classical Architecture and Classical America, A Vision of Europe, and
the Congress for New Urbanism. The conference and exhibition titled
“The Art of Building Cities” took place in 1995 at the Art Institute of
Chicago and was the first event in the United States to link the prac-
tice of contemporary classicism with the new traditional urbanism.
An exhibition and conference titled “The Other Modern,” took place
in Bologna, Italy, in 2000, and a conference titled “Three Generations
of Classical Architects: The Renewal of Modern Architecture” was
held in October 2005 at Notre Dame.

Thomas Nowak is Professor Emeritus of Chemistry and Biochem-

istry at the University of Notre Dame. He specializes in structural bi-
ology and biophysics. A former researcher at the Institute for Cancer
Research, Nowak came to Notre Dame in 1972. His research group is
currently focusing on enzymes involved in metabolic energy utiliza-
tion and energy storage and enzyme mechanisms, their regulation and
structure, how they are interrelated, and their roles in the regulatory
 330 Contributors

and catalytic processes. His research has appeared in numerous jour-

nals, including Journal of the American Chemical Society, Biochemistry,
Journal of Biological Chemistry, and Archives of Biochemistry and Bio-
physics. Nowak is the recipient of a National Institutes of Health Re-
search Career Development Award. He has served as a visiting pro-
fessor at the University of Groningen in the Netherlands and at the
Pontifical Catholic University of Chile in Santiago. Nowak received the
Peter A. Burns, C.S.C. graduate school award in 2001 and the Grenville
Clark Award in 2008 from the University of Notre Dame.

Zygmunt Pizlo is Professor of Psychology and of Electrical and

Computer Engineering at Purdue University. His research interests in-
clude visual perception, motor control, and problem solving. Pizlo
earned PhD degrees in electronic engineering (1982) and psychology
(1991). Testing his theories in a robot, he formulated the first psycho-
logically plausible theories of how humans perceive 3D shapes, solve
combinatorial optimization problems, and trade speed for accuracy in
perception and motor control. In 1994, Pizlo received the New Inves-
tigator Award from the Society for Mathematical Psychology. He has
over one hundred publications, including papers in Computer Vision
and Image Understanding; Memory and Cognition; Journal of Prob-
lem Solving, Perception and Psychophysics; Journal of Vision; Vision
Research; Journal of Imaging Science and Technology; Pattern Recog-
nition Letters, Electroencephalography and Clinical Neurophysiology;
and Journal of Mathematical Psychology. Pizlo is also the author of
3D Shape: Its Unique Place in Visual Perception (2008), the first mono-
graph on human perception of three-dimensional shapes, as well as a
coauthor of Making a Machine That Sees Like Us, which tells the story
of a machine built by the authors that solves the computationally
difficult problem of seeing the way humans do. He served on the edito-
rial boards of Behavior Research Methods, Instruments, and Com-
puters (1995– 98) and the Journal of Mathematical Psychology (2003– 9)
and was president and vice president of the Society for Mathematical
Psychology (2008–10). He was the founding editor and currently serves
as editor in chief of the Journal of Problem Solving. Pizlo’s research
has been supported by the National Science Foundation, the National
Institutes for Health, the Department of Defense, the Department of
 Contributors 331

Energy, Hewlett-Packard, Sandia National Laboratories, and the Air

Force Office of Scientific Research.

Aviezer Tucker is Assistant Director of the Energy Institute at the

University of Texas in Austin. He specializes in philosophies of histo-
riography and history, with a particular interest in the epistemic as-
pects of scientific knowledge of the past and the problem of inference
of common causes or origins. His more recent work includes topics in
political philosophy and theory. He is the author of Our Knowledge
of the Past: A Philosophy of Historiography (2004) and The Legacies of
Totalitarianism (forthcoming). He is editor of The Blackwell Compan-
ion to the Philosophy of History and Historiography (2009). His numer-
ous articles have appeared in journals that include History and Theory,
British Journal for the Philosophy of Science, Studies in the History and
Philosophy of Science, Philosophy, Inquiry, and Erkenntnis. He is the
recipient of numerous research fellowships, including those held at the
University of Cologne, the Max Planck Institute, the Gvirtzman Me-
morial Foundation in Prague, the Australian National University, New
York University, Columbia University, and the Central European Uni-
versity in Prague.

Osborne Wiggins is Professor in the Department of Philosophy

and Associate Faculty in the Institute for Bioethics, Health Policy, and
Law at the University of Louisville. His areas of interest are phenome-
nology and philosophical anthropology (especially as they apply to
psychiatry), the history of ideas, and the philosophy of medicine. With
coauthor Michael Alan Schwartz, M.D., Wiggins has published numer-
ous articles on phenomenological psychiatry, addressing classification
in psychiatry, the lifeworld-science relationship, schizophrenic expe-
rience, and other topics. For this work and his contributions to phe-
nomenological psychiatry, in 1998 Wiggins received the Margrit Egner
Award. He is a member of the core faculty of the M.A. program in
Bioethics and Medical Humanities at the University of Louisville. With
Annette C. Allen, he edited Clinical Ethics and the Necessity of Stories:
Essays in Honor of Richard M. Zaner (2010).
 i n d e x

Academy: systematization of Summa Theologiae 1a Q.1.2.,

knowledge in the, 32 319n.1
Acts 4:12, 103 Summa Theologiae 1a Q.1.5, 301
adenosine diphosphate (ADP), 122 on theology, 23, 300 – 301, 304, 318,
adenosine triphosphate (ATP), 120, 319n.1, 320n.17, 321n.26
121, 122, 124 arbor porphyriana, 40
agriculture, 54 archaeology, 232, 252, 275
Alhazen, 161 architecture, 51, 60, 68n.64
Alston, William P., 230n.17 and anthropology, 290
altruism, 153 as art, 291, 296
animal cruelty, 201–10, 211, 212–13, Bacon on, 67n.51
214n.5 beauty in, 294 – 95
Ankersmit, Frank, 244 and biology, 22, 290
anthropology, 24, 31, 59, 193, 290, character in, 294
305, 306 and chemistry, 22, 290
apriority classical architecture, 22– 23, 291,
of inferential principles, 6, 7, 91, 295– 97
95, 96 and culture, 292, 293– 94, 297
Kant on synthetic a priori and human nature, 291– 93,
knowledge, 12, 54– 55 296– 97
role of a priori knowledge in and law, 291
perception, 170– 74 and mathematics, 22, 289– 90, 292,
Aquinas, Thomas 294
on ethics, 301 modern architecture, 22– 23,
on five ways to God, 321n.26 297– 98
on foundations of morality, 144 and music, 22, 291
on God’s knowledge, 318 and myth, 296, 297
on God’s Providence, 314 and nature, 296– 97
on human reason, 301, 304 and physics, 22, 290, 297
on prudence, 313–14 principle of equilibrium, 23
on science, 23, 300– 301, 304, private vs. public in, 22, 292– 94
321n.26 and psychology, 290

332
 Index 333

sacred buildings, 22, 293– 94 on theoretical knowledge, 4, 35,

solid-void relationships in, 22, 291 36– 37, 38, 39, 65n.33
sustainability in, 23, 295– 96, 297, on zoology, 37
298 Aristotle, works of
types in, 293– 94 Analytics, 67n.54
and unity of knowledge, 21– 22, The Mechanical Problems, 44
289, 297– 98 Metaphysics, 35, 36, 37, 57
Aristotle Oikonomika, 39
on astronomy, 37 On Sophistical Refutations,
on biology, 37, 45 67n.54
vs. Comte, 61 Organon, 36, 67n.54
on dialectic, 35, 36 Physics, 44
on economics, 36, 39 Poetics, 65n.23
on ethics, 4, 35, 36, 144 Politics, 39, 65n.26
on final causes, 37 Prior Analytics, 67n.50
on formal causes, 37 Protrepticus, 65n.33
on foundations of morality, 144 Topics, 35, 65n.30, 67n.54
on happiness, 36 art, 3
on the heavens, 119 architecture as, 291, 296
vs. Hegel, 57, 58, 59 Hegel on, 60
on human intellect, 37, 48 imitation in, 296
on humans as political animals, 36 painting, 51
on logic, 35, 36, 57 sculpture, 51, 68n.64
on mathematics, 36, 39, 45 artes liberales, 36
on metaphysics, 35, 39, 44, 52, 58 artificial intelligence, 165
on perception, 163, 171 ascetic spirituality, 312
on physics, 4, 35, 36, 37, 39, 44 astronomy, 34, 45, 49, 59, 269– 70,
vs. Plato, 34, 35 277
on poetical (productive) Aristotle on, 37
knowledge, 4, 35– 36, 38 in Babylon, 33
on poetry, 36, 59, 65n.23 Comte on, 62, 63
on politics, 36 d’Alembert on, 52, 53, 54
on practical knowledge, 4, 35, 36, of Kepler, 47
38 in Sumeria, 33
on praxis/actions, 35– 36 atheism, 301, 304, 309
on reasoning, 65n.30 Augustine, St., 310
on rhetoric, 36, 59 Australopithecus, 141, 142
on soul and body, 37 Avery, O., 120
on strategy, 36 Ayer, Alfred Jules
on theology/First Philosophy, 36, on expressivism, 16, 207, 212
38, 45, 58 Language, Truth and Logic, 207
 334 Index

Babylonian astronomy, 33 Barth, Karl: on natural theology, 309

Bacon, Francis Bartholomew II, 312
on architecture, 67n.51 Bayes, Thomas, 235
on causality, 45 Bayes rule, 166
vs. d’Alembert, 51, 62, 68n.61 beauty, 3, 20, 23
d’Alembert on, 42, 50– 54 belief
on disciplines and faculties, first principles as grounds of, 74
43– 44, 51, 54 Parmenides on opinion, 33
on divine learning, 43, 45, 51 as subjective truth, 115–16
on First Philosophy, 58 as subject to revision, 6
vs. Hegel, 57, 58 as true and justified, 16, 17, 94,
on history, 43– 44 208, 209–10, 211, 223, 224,
on imagination, 4, 51 230n.17
on liberal arts, 67n.51 Benacerraf’s Dilemma, 91, 92, 100n.6
on logic, 57 Benedict XVI: on integral human
on mathematics, 45, 53, 67n.51 development, 313
on mechanical arts, 44 Bentham, Jeremy, 195
on memory, 4, 43, 46, 51 Berkeley, George, 99, 161, 169
on mind and body, 46 Bevir, Mark, 233
on morality, 46– 47 biblical studies, 237, 253, 254
Of the Proficience and documentary hypothesis in, 251
Advancement of Learning, Bielenstein, Hans, 269
Divine and Human, 42– 47, Big Bang theory, 19, 118, 233– 34,
66n.47, 68n.61 237, 239
on philosophy, 43, 45– 47, 48, 51, 52 biochemistry, 9–10, 119, 120 – 39
on poetry, 43, 44– 45, 51 biodiversity, 311
on reason, 4, 51 biology
and scientific revolution, 31 and architecture, 22, 290
on system of disciplines, 4, 42– 47, Aristotle on, 37, 45
50– 54 DNA, 18, 120, 213, 235, 239, 240
on theology, 52 proximate causes in, 24, 304,
Bain, Alexander, 183 320n.14
Balazs, E., 261 relationship to ethics, 11, 140,
Ban Gu 143– 57
as founder of Chinese imperial relationship to humanities, 10 –11
historiography, 260, 269 relationship to language, 149
Hanshu, 260, 265, 269 relationship to social sciences,
Barkow, J., 148 10 –11
Barnes, Annette: On Interpretation, relationship to theology, 23– 25,
230n.7 300, 303– 5
 Index 335

sociobiology, 63, 148 Hopi Ethics, 214n.5

ultimate causes in, 24, 304– 5, on qualified attitude method, 16,
320n.14 205– 7, 208– 9, 210, 211, 212
See also evolution; natural Brindley, G. S., 163
selection British empiricism, 54, 183
Bjorklund, F., 151 British imperialism, 32
Blackburn, Simon Brosnan, S., 156
on expressivism, 213n4 Bruner, J. S., 164
on quasi-realism, 211 Brunner, Emil, 309
Ruling Passions, 213n4 Büchner brothers, 119
Spreading the Word, 213n4 Burckhardt, Jacob: on the
Blackmore, S., 148 Renaissance, 18, 233
Blake, Robert, 254
Bohr, N., 119 Calatrava, Santiago, 290
Bok, H., 151 Čapek (robot), 168, 172, 173
Bonaventure, St. Capella, Martianus: De nuptiis
De reductione artium ad Philologiae et Mercurii, 65
theologiam, 4, 38– 40 capitalism, 32
on faith, 310 Carr, David: on historiographic
on five senses, 39 narratives, 242– 43
on God as Creator, 309–10 Carr, E. H.: What Is History?,
on love, 310–11 234 – 35
on mechanical arts, 38– 39 Castaneda, Hector-Neri, 86
on philosophical cognition, causality
39– 40 developmental causes, 304
on science, 310 efficient causes, 45
on scripture, 38, 39, 40 as experienced in schizophrenia,
on truth, 310 184
on wonder, 310–11 final causes, 45
on Word of God, 310 formal causes, 45
Bonhoeffer, Dietrich and inference, 6, 7, 91– 92
on God, 24, 304 material causes, 45
on scientific explanations, 24, 304 and perception, 159– 60
Bopp, Franz, 60 phylogenetic causes, 304
botany, 54, 59, 62 proximate causes, 24, 304, 320n.14
Boyd, R., 11, 152 ultimate causes, 24, 304 – 5, 320n.14
Brandis, Christian, 67n.54 Cavendish, Henry, 118
Brandt, Richard Cela-Conde, C. J., 154
on cruelty to animals, 214n.5 Chalmers, David J.: extended mind
Ethical Theory, 205– 7 thesis, 189– 90
 336 Index

Chambers, Ephraim, 68n.61 objective record vs. philosophical/

Chan, Hok-lam, 280 moral foundation in, 263, 273,
chemistry, 25, 53, 62, 63 276, 282
and architecture, 22, 290 portents and omens in, 269– 70
China private vs. public historiography,
examinations system, 274 261, 263, 271, 272, 282
Former Han dynasty, 268 relationship to dynastic legitimacy
Jin dynasty, 273 (zhengtong), 273– 74
Ming dynasty, 272, 281 relationship to ritual reality, 266– 67
Qin dynasty, 277 role of the state in, 262– 63, 269,
Spring and Autumn period, 270 – 71, 272, 273, 275, 282
265– 66 sources in, 261, 264 – 65, 267– 69,
Sui dynasty, 281 270 – 72, 273, 275, 276, 278– 79,
Warring States period, 265– 66, 282
268, 278 truth in, 20 – 21, 263, 264 – 76,
Yuan dynasty, 275, 280 281– 82
See also Han dynasty; Qing types of knowledge in, 260 – 61
dynasty; Song dynasty unity in, 20, 263, 264, 276– 82
Chinese historiography Chisholm, R. M., 79
annalistic vs. narrative elements Christianity, 60
in, 261, 263, 265 Chrysippus, 37– 38
appropriate concealment in, Chunqiu (Spring and Autumn
266– 67 Annals of Lu), 265– 67, 268,
author’s opinion in, 265 274, 280, 280 – 81
cause and effect in, 261, 265, 267 Clark, Andy: extended mind thesis,
comprehensive model in, 263, 189– 90
276– 78, 279– 80 Cleland, Carol E., 239, 240, 248
court scribes (shi), 266 Coakley, Sarah, 181
and cyclical change, 21, 264, 276, on God’s existence, 321n.26
277, 279– 80 on natural theology, 321n.26
dao/universal principles in, 276, Sacrifice Regained, 321n.26
277 cognitive psychology, 24, 305
dynastic model in, 262– 63, 264, coherentism, 5, 6, 25, 73– 74, 76– 77,
268– 69, 270– 71, 275, 279, 280, 80
281 Collingwood, R. G., 239
and History Office, 20, 262, Collins, Harry, 192, 256n.3
270– 71, 272 Collins, H. M.: on collective tacit
moral judgments in, 261, 263, 264, knowledge, 234
265, 267, 271, 272– 75, 276, common law standards of evidence,
278– 79, 280, 281, 282 254
 Index 337

comparative historical philology, cosmology, 44, 49, 54, 56, 232, 236
232, 233, 252, 253 Big Bang theory, 19, 118, 233– 34,
computational algorithms, 166, 168, 237, 239
173– 74 Craik, K. J. W., 174
Comte, Auguste creationism, 317
vs. Aristotle, 61 cruelty to animals, 201–10, 211,
on astronomy, 62, 63 212–13, 214n.5
vs. Bacon, 62 cultural evolution
Cours de philosophie positive, vs. biological evolution, 30, 140,
60– 64, 69n.68 142, 143, 152– 53
vs. d’Alembert, 62 gene-culture coevolution, 152– 53
encyclopedic interests, 54 relationship to moral norms,
on engineering, 4, 62 11–12, 143, 145– 46, 154 – 57
vs. Hegel, 54, 63– 64 culture
on introspection, 61, 63 and architecture, 292, 293– 94, 297
on law of the three stages, 61 relationship to ethics, 11–12, 16,
on mathematics, 62– 63 140, 143, 145– 46, 154 – 57
on organisms and dead matter, 62 Vico on, 4, 48, 49
on physics, 62, 63
positivism of, 54 Dai Yi, 262
on the sciences, 4, 61– 63 d’Alembert, Jean Le Rond
on sociology, 61, 62, 63 on astronomy, 52, 53, 54
on theoretical vs. practical on Bacon, 42, 50 – 54
endeavors, 61– 62 vs. Bacon, 51, 62, 68n.61
Condillac, Étienne Bonnot de, 50 vs. Comte, 62
Confucianism vs. Diderot, 68n.61, 68n.67
Classics of, 20, 21, 263, 266, 267, “Discours préliminaire de
268, 272, 280– 81 l’encyclopédie”/Encyclopédie,
Confucius and Chunqiu, 266, 4, 42, 49– 54, 68n.62
267, 280 on ethics, 52– 53
and historical knowledge, 261, on fine arts, 51
263, 273, 276, 280– 81 vs. Hegel, 57, 58
neo-Confucianism, 21, 262, 274, on history, 50, 51– 52
275, 276 on logic, 50, 52– 53, 57
values in, 280 on mathematics, 53
consciousness, 74, 82, 84 on mechanical arts, 44, 50, 51
conservation laws, 10 on metaphysics, 51, 52, 58
Copp, D., 144 on rational philosophy, 52– 53
1 Corinthians 1:8– 2:5, 323n.44 on theology, 52
Cosmides, L., 148 on tree of knowledge, 49– 54
 338 Index

Danto, Arthur, 246 Diderot, Denis

Darnton, Robert: The Great Cat vs. d’Alembert, 68n.61, 68n.67
Massacre and Other Episodes in on d’Alembert’s tree of
French Cultural History, 68n.61 knowledge, 51
Darwin, Charles, 24, 63, 144, 233, “Explication détaillée du systeme
317 des connoissances humaines,”
The Descent of Man, 11, 145– 46, 49, 53– 54, 68n.66
155 “Prospectus” of 1750, 49
on moral behavior and Dilthey, Wilhelm, 63
intelligence, 11, 145, 149 Dimensions of Goodness, 2, 20
on moral norms, 11, 145– 46 direct/forward problems, 12, 160, 161
on moral sense/conscience, 11, divine command, 144, 156, 157
145– 46, 149 DNA, 18, 120, 213, 235, 239, 240
on natural selection, 145– 46, 155, double helix model of DNA, 239
251– 52, 304, 305, 321n.21 Dretske, Fred I.: on primary and
and natural theology, 307, 321n.21 secondary ways of epistemic
on tribal success, 155 seeing, 237– 38
Davis, Paul: on cosmology and God, Dummett, Michael
308 as antirealist, 245
Dawkins, Richard, 309, 321n.24 on direct observation vs.
Unweaving the Rainbow, 308 inference, 245
DeBlasi, Anthony, 273 on historical truth, 19, 245– 48, 249
delusion, 184 on subjunctive justification, 19,
Demiéville, P., 276 245– 48, 249
Dennett, Daniel, 309, 321n.24 Durrant, Stephen, 260, 268, 269
DeRose, Keith, 237, 253 Dworkin, Ronald, 202
Descartes, René
on the cogito, 47– 48 Eastern Orthodoxy, 311–12
vs. d’Alembert, 52 ecological praxis, 24
on extended substances, 4 economics, 4, 32, 49, 53, 60
on knowledge, 4 Ehrlich, P., 118
on metaphysics, 47 Einstein, Albert, 18, 233
on philosophy as a tree, 47 on imagination, 117
Les Principes de la Philosophie, 47 Ekstrom, L., 151
and scientific revolution, 31 Eleatic school, 3, 33
on soul and matter, 47– 48 elements, 118
on thinking substances, 4 Elman, Benjamin A., 281
on universals, 171 emotivism, 6– 7, 93
Desch, Michael, 2 empathy, 11, 153
de Wall, F., 156 empirical inductivism, 234
 Index 339

Encyclopédie/d’Alembert’s moral epistemology, 205– 7

“Discours,” 4, 42, 49– 54, 68n.62 moral judgments in Chinese
engineering, 4, 22, 62, 165, 290 historiography, 261, 263, 264,
Enlightenment, 49, 162 265, 267, 271, 272– 75, 276,
Enoch, David, 202 278– 79, 280, 281, 282
on pragmatic account of moral optimism, 155
justification, 96 moral principles vs. moral
Enquist, M., 152 attitudes, 206
enzyme kinetics, 123– 26, 127, 128, moral realism, 16, 144, 211–13
129, 132, 133, 134, 137– 39 moral sense/conscience vs. moral
Epidaurus theater, 23 norms, 146, 148– 52, 153, 154
epistemology, 3, 5– 8, 32 naturalistic definitions in, 15–16,
coherentism in, 5, 6, 73– 74 195– 97
evident data vs. interpretation, 9, non-naturalism in, 15, 195– 99,
115–17 198, 202– 5, 206, 207
foundationalism in, 5– 6, 74– 76 normative ethics, 144, 197, 198
intuition in, 73– 74 Plato on, 34, 35
as normative, 197, 198– 99 and reflective equilibrium, 203
as sensualist, 50 relationship to biology, 11, 140,
Eriksson, K., 152 143– 57
Ermodoro, Senocrate, 65n.19 relationship to consequences of
ether, 119 actions, 149– 50, 151, 152, 153
ethics relationship to culture, 11–12, 16,
Aquinas on, 301 140, 143, 145– 46, 154 – 57
Aristotle on, 4, 35, 36, 144 relationship to natural selection, 11
and conflicting moral claims, 15, role of moral affect, 16, 200,
201– 2, 210, 211, 212, 213 205– 6, 208
d’Alembert on, 52– 53 value judgments, 149, 150 – 51
ethical knowledge, 14–16, 54, See also moral norms
55– 56, 195– 96, 197, 199– 202, Euclid’s Elements, 34, 76
205– 7, 209–10, 211–13 European education, 36
expressivism, 15, 207– 9, 211–12, evolution
213n4 causality in, 24, 304
ideal response theories, 15, 204– 5 evolutionary psychology, 148
intuitionism in, 15, 16, 195– 96, exaptation, 23– 24, 304, 320n.10
197, 199– 205, 207, 211, 213 gene-culture coevolution, 152– 53
Kant on, 15, 54, 55– 56, 58, 89, 94, and genetic drift, 305
144 Lamarckian evolution, 11, 154
vs. mathematics, 202– 3 of organisms, 9, 30, 116, 140,
metaethics, 144, 200– 202, 211–13 141– 52, 232, 237, 247, 251– 52
 340 Index

evolution (cont.) French imperialism, 32

of religion, 304– 5, 317 French Revolution, 238, 241
spandrels in, 24, 304, 320n.11 fructose-1,6-diphosphate (FBP),
and theology, 23– 24, 303– 5 125– 26, 127, 129, 130, 132, 133,
thresholds in, 152 134 – 35
See also cultural evolution; natural
selection Gadamer, Hans-Georg, 16
Ewing, A. C., 213n.1 Galileo Galilei, 47, 118, 233
exaptation, 23– 24, 304, 320n.10 Gallagher, Shaun, 186
exemplar representation, 6, 83– 87 Gazzaniga, M. S., 153, 156
experimental data, 10, 162– 63, gene-culture coevolution, 152– 53
186– 87, 196, 238, 240– 42 Genesis 3:17, 40
from observation, 9, 15, 31, 115–17 genetics, 119– 20, 305
expertise in self-reflection, 14 genomes, 247
expressivism, 16, 207– 9, 211–12, 213n4 Gentz, Joachim, 266
geography, 31, 63, 275
faith, 41, 115, 116, 300, 310, 321n.26 geology, 54, 59, 251
Fan Ye: History of the Later Han German Idealism, 4
Dynasty (Hou Hanshu), 269– 70 germ theory of disease, 118–19
Fechner, Gustav, 12, 159– 60 Gestalt psychology, 13, 170, 171, 186
Fei-Fei, L., 168 Gettier, Edmund, 230n.17
Field, Hartry, 100n.6 Gibson, E., 165
on inferential principles, 95 Gillett, Grant, 181
figure-ground organization (FGO), givenness
167, 168, 170, 172 direct givenness in self-awareness,
Fischer, J. M., 151 14, 183, 185, 187– 88, 190, 191– 92
Five Agents theory, 280 vs. interpretation, 187– 88, 225, 227
Fleming, A., 118–19 Sellars on myth of the given, 6,
Fodor, J. A., 82, 87 76– 77, 84 – 85
foundational disciplines, 2– 3 glycolysis, 9–10, 120 – 39
foundationalism, 5– 6, 74– 76 God
Four Horsemen of the Apocalypse as creator, 40, 309–10, 316, 321n.21
(video), 321n.24 existence of, 24, 321n.26
free will goodness of, 24, 309, 310
and inference, 7, 90, 92, 98– 99, and laws of nature, 308
101n.17 and mathematics, 24, 111
Kant on, 89 perfect knowledge possessed by,
relationship to morality, 151– 52 318
Frege, Gottlob, 4, 59 providence of, 314 –15
The Thought, 57 as reason, 58
French Enlightenment, 49 as trinity, 40
 Index 341

Goethe, Johann Wolfgang von Hebb, D. O., 167– 68

“Das Göttliche,” 231n.24 Hegel, G. W. F.
“Ganymed,” 229 on the absolute, 57
Goldstein, Leon J., 239, 245, 250 absolute idealism of, 99
Goodman, C. C., 164 on architecture of knowledge, 25
Goodman, N., 83 vs. Aristotle, 57, 58, 59
goodness, 2, 20, 24 on art, 60
of God, 24, 309, 310 vs. Bacon, 57, 58
as non-natural property, 196– 97, on civil society, 59– 60
205, 207 as coherentist, 5
and warranted desire, 196– 97 vs. Comte, 54, 63– 64
Gould, Stephen Jay vs. d’Alembert, 57, 58
on exaptation, 320n.10 on dialectic, 56– 57
on science and religion, 303– 4 Encyclopedia of the Philosophical
grammar, 4, 36, 39, 53 Sciences, 56– 60
great chain of being, 41, 66n.45 encyclopedic interests, 54
Green, D. M., 163 on God, 57, 58
Greene, J. D., 151, 152 on history, 60
Griffith, F., 120 vs. Kant, 54, 56, 58, 59
Gurwitsch, Aron, 184, 186 on language, 59, 60
on logic, 4, 56– 59, 60
Hacking, Ian, 256n.6 on mathematics, 58– 59
Hadjadj, Fabrice on nature, 56– 57
L’Héritage et la promess, 102 on organisms, 59
on science and truth, 102, 103 as panentheist, 57
on theology, 102, 103 on philosophy, 60
Haidt, Jonathan, 151, 152 on physics, 58, 59
on ethic intuition, 200 vs. Plato, 58
The Righteous Mind, 199, 200 on Realphilosophie, 58
Haitian earthquake of 2005, 314 on religion, 58, 60
Halki conference of 2012, 312 on spirit, 56– 57, 58, 59– 60, 63
hallucination, 184 Heidegger, Martin: on
Hamlet, 225– 26 interpretation, 218
Han dynasty, 260, 261, 268, 269, 277, Helmholtz, H. von, 167– 68
278, 281 Henrich, J., 152
happiness, 36, 195, 201– 2 Herder, Johann Gottfried, 60
Harman, G., 79 hermeneutics, 17–18, 49, 60, 63, 193
harmonics, 34 Hess, E. H., 165
Harris, Sam, 309, 321n.24 higher education and wisdom, 307,
Hartman, Charles, 273, 274, 278 315–18
Hauser, M., 148, 151 Hill equation, 133, 137, 138– 39
 342 Index

historical comparative linguistics, consensus among historians, 233,

235, 251, 252 245, 249– 51, 252, 255
historical sciences documents as evidence in, 241– 42,
defined, 232 248, 254, 256n.3
evidence in, 247– 48 epistemic contextualism in,
founding revolutions in, 251– 52 19– 20, 232, 236, 237, 253– 54,
inference in, 232– 34, 235– 36 257n.7
information decay vs. scientific inference in, 18– 20, 43, 232– 33,
extraction, 247– 48 236, 239
methodology, 232– 33 information-bearing signals in,
vs. theoretical sciences, 18, 19, 238, 240
233, 238– 39, 240– 41, 250, and particular token events, 238
251– 52, 256n.4 philosophies of, 233– 35, 236– 49,
historical truth, 18– 21, 232– 55 256n.6
in Chinese historiography, rationalization among historians,
20– 21, 263, 264– 76, 281– 82 234 – 35, 236
and consensus among historians, role of probability in, 254 – 55
250, 251 teaching of, 234 – 35, 255n.1
constructivism regarding, See also Chinese historiography
239– 40 history, 41, 193
correlation theory of, 249 Bacon on, 43– 44
Dummett on, 19, 245– 48, 249 d’Alembert on, 50, 51– 52
vs. historical evidence, 238, Hegel on, 60
250– 51 philosophy of, 256n.5
as isomorphic communal self- Vico on, 48, 49, 276
consciousness, 242– 43 Hitchens, Christopher, 309,
justificationist view of, 19, 321n.24
245– 48, 249 Hitler, Adolf, 255
as overdetermined, 252 Ho, H. L., 254
as probabilistic, 18–19, 232, holism, 91, 95
235– 36 Homeric sagas, 233, 252
skepticism regarding, 19, 236, 237, Homo erectus, 141, 145, 153
256n.2 Homo habilis, 141– 42, 145, 152,
as theoretical truth, 239– 40 153
as underdetermined, 252– 53, Homo neanderthalensis, 141,
254– 55 144, 153
historicism, 21, 30, 263 Huang, Chun-chieh, 277
historiography Hugh of Saint Victor
aesthetic dimension of, 244 Didascalion, 4, 38– 39
antirealism regarding, 243, 245, on mechanical arts, 38– 39
249– 51, 255 Hulsewé, A. F. P., 262
 Index 343

human beings Huxley, Julian, on evolutionary

behavioral traits, 10–11, 142 ethics, 11, 147– 48
brain size, 10, 140, 141– 42 hydrology, 54
cryptic ovulation of, 10, 142 hypotheses, 116–17, 135, 252– 53,
distinctive traits of, 10–11, 140, 254 – 55, 256n.5
141– 43
erect posture/bipedal gait of, 10, Ichikawa, Jonathan Jenkins: The
140, 141, 142, 150 Analysis of Knowledge, 230n.17
human choice, 149, 151– 52 ideal response theories, 15, 204 – 5
human nature, 305– 6 illusions of sense, 82– 83, 87, 161,
intelligence of, 10, 11, 142, 145– 46, 169, 174
149– 50, 151, 152, 153, 154 Müller-Lyer illusion, 12–13, 164,
opposing thumbs of, 10, 142 165
origins of, 10, 141, 317 imagination, 4, 14, 43, 51, 117,
social organization, 142– 43 189– 93
symbolic language of, 10, 11, 142, Indo-European hypothesis, 233, 251
149, 154, 156, 247 Industrial Revolution, 233, 246,
toolmaking of, 11, 142, 150 256n.6, 295
vocal-tract modification of, 10, 142 inference
See also free will as abductive, 6, 90, 91, 94, 95, 99
humanities apriority of inferential principles,
as interpretive disciplines, 16–17, 25 6, 7, 91, 95, 96
and mental acts, 16 as Bayesian, 235
and philosophical categories, 1– 2 causal mechanism of, 6, 7, 77– 78,
relationship to biology, 10–11 91– 92, 98– 99, 101n.16
Humboldt, Wilhelm von, 60 and cognitive phenomenology, 91,
Hume, David, 78 96– 99
on foundations of morality, 144 Contemporary Kantian Moralism
on sensations as exemplars, 83, 86 about Inference (CKMI), 6,
Hurricane Katrina, 314 90 – 91, 92, 93– 94, 95– 96, 98– 99,
Hurricane Sandy, 314 101n.16
Husserl, Edmund, 4, 13 as deductive, 6, 89, 91, 94, 95,
on distinction between 105– 6
abstract/theoretical and defined, 89– 90
concrete/ontological sciences, vs. direct observation, 245
30– 31 emotivism regarding, 6– 7, 93
as foundationalist, 5 and free will, 7, 90, 92, 98– 99,
on intentionality, 184 101n.17
Logical Investigations, 30– 31, 57 and functionalism, 91, 97
“Prolegomena” to Logical in historiography, 18– 20, 43,
Investigations, 30– 31 232– 33, 236, 239
 344 Index

inference (cont.) interpretation

as inductive, 6, 89, 91, 94, 95 conflict of possible
justification of, 6, 7, 18–19, 90– 91, interpretations, 17, 220, 222– 23,
94 – 96 229
in logic, 5, 7, 9, 25, 29 as creative process, 225– 26
logocentric predicament of data, 9, 115–17, 187– 88, 218–19
regarding, 94– 95 vs. givenness, 187– 88, 225, 227
and mathematics, 7– 8, 91 of Hamlet, 225– 26
mechanism of, 6, 90, 91, 96– 99 intentionalist framework of, 222
metaphysics of, 6, 90, 91– 92 vs. preinterpretive perception, 17,
monotonic vs. nonmonotonic 218, 220
reasoning, 100n.2 role of truth and knowledge in,
noncognitivism regarding, 91, 95 218– 23, 230nn.9–10
vs. noninferential belief, 77– 78, of visual art, 219– 23
79– 80 See also literary interpretation
and Platonism, 6, 90, 91– 92 introspection, 3, 4, 15, 25, 61, 63
principles of, 7, 94– 96 cogito of Descartes, 47– 48
and psychologism, 6, 90 and mind/body problem, 182– 84
purpose of, 6– 7, 90, 92– 94 role in psychology, 13, 181– 85,
rational normativity of, 90, 92– 94, 188
95– 96, 98– 99, 101n.16 vs. self-awareness/first-person
relationship to metaphysics, 57 perspective, 13–14, 182, 183,
role in perception, 12, 160– 61 184 – 85, 192
and separatism, 91, 97– 98 intuition
as summum bonum of reasoning, vs. coherence, 5– 6, 25, 73– 74,
6 76– 80
and transcendental mentalism, 6, criticisms of, 74
16, 90, 91, 92, 96 in ethics, 15, 16, 195– 96, 197,
weak or counterfactual 199– 205, 207, 211, 213
transcendental idealism as evidence, 73– 76
(WCTI), 92, 96, 99 in mathematics, 202– 3
inferentialism, 91, 95 vs. reasoning, 200
Ingarden, Roman inverse problems, 12–13, 160, 167– 74
on concretization in literary irony, 223
interpretation, 217, 226 Ismael, J., 85
on indeterminacy in literary
works, 217, 226– 27 James, William, 63
innatism, 12 on introspection, 13, 183
instrumentalism, 6– 7, 93 on mind/body dualism, 182– 83
intentionality, 91, 184– 85 Principles of Psychology, 182– 83
 Index 345

Jesus Christ, 309, 317–18 Third Critique, 54, 59

on being the truth and the life, 103 on time, 12, 161
as Truth, 103 Keller, Catherine: on emancipatory
as Word, 310, 312–13 wisdom, 316
John (Gospel of) 14:6, 103 Kepler, Johannes, 47
John Paul II, 312 keystone principles, 5– 6, 81, 85– 87
Joyce, Richard, 203 knowledge
as common sense, 115–16
Kahneman, Daniel, 213n.2 departmentalization of, 1, 29– 31
Kane, R., 151 discursive knowledge, 6, 81– 82,
Kant, Immanuel 84, 86, 87
on a priori principles, 89 evidence required for, 5– 6, 9, 18,
on architectonic unity, 55 73– 76
vs. Bacon, 55 growth of, 5, 29– 30
on causality, 12, 161 historical development of, 2, 3– 4,
on classification of sciences, 54– 56 29– 30
vs. d’Alembert, 55 vs. imagination, 117
on ethical knowledge, 15, 54, and inference, 5, 6
55– 56, 58, 89, 94, 144 as justified/warranted, true belief,
First Critique, 54, 55, 89, 100n.6 16, 17, 94, 208, 209–10, 223, 224,
on free will, 89 230n.17
on general vs. applied logic, 89 Parmenides on opinion vs., 33
Groundwork of the Metaphysics primitive knowledge, 6, 81– 83,
of Morals, 37 84 – 85, 86, 87
vs. Hegel, 54, 56, 58, 59 rise of new disciplines, 29– 32
letter to Marcus Herz, 100n.6 role in literary interpretation,
on logic, 15, 89 216–17, 218, 223– 29
on mathematics, 15, 55 tacit vs. explicit knowledge, 234
on metaphysics, 55, 56 tree of, 2, 3, 30, 31, 38, 40 – 41, 42,
on metaphysics of morals, 94, 96 47, 49, 50 – 54, 61, 63– 64
on philosophy, 54– 56 unity of, 2, 21– 22, 23, 25, 276– 77,
on punishment, 203 289, 297– 98, 300, 302– 3, 304
on reason, 55, 56 See also apriority
vs. Reid, 161 Koffka, K., 170
separation of normative and Köhler, K., 170
descriptive disciplines, 63– 64 Kopernik, N., 118
on space, 12, 161 Kosso, Peter
on synthetic a priori knowledge, on information-bearing signals, 238
12, 54– 55 on scientific observation, 237– 39,
on systematization of knowledge, 4 241
 346 Index

Krafft-Ebing, Richard von: as creative process, 17–18, 217–18,

Psychopathia sexualis, 32 225– 26
Kriegel, U., 84 literary work vs. text, 227
Kuhn, Thomas S., 9, 244 role of authorial intent in, 18,
Kusch, Martin, 256n.3 227– 28
Kusukawa, Sachiko, 67n.49 role of knowledge in, 216–17, 218,
223– 29
Lamarckian evolution, 11, 154 role of preinterpretive perception
Lanczos, C., 175 in, 17, 218, 220
language, 149, 154, 156, 247 role of truth in, 216–17, 218,
as distinctive human trait, 10, 11, 223– 29
142 schools of, 217
words for mental processes, 187 web of questions in, 223– 24
Latin language, 50 and zones of indeterminacy in
Laudan, Larry, 251 literary works, 217, 226– 27,
Lavoisier, A., 118 231n.24
law, 22, 65n.32, 291 Liu Zhiji
laws of nature, 9, 61– 62, 67n.52, 117, on official historiography, 21,
308 271– 72
Lee, Thomas H. C., 273, 279, 280 Shi Tong, 271– 72, 280
Leibniz, Gottfried Wilhelm von, 45, Llull, Ramon
54 Arbor scientiae, 4, 38, 40 – 41, 51
Lewis, David, 237 on faith and reason, 41
Lewis, Mark, 266 on theology, 41
Li, Wa-yee, 265, 267 Locke, John, 50
Li, Y., 170, 172 logic, 3, 8, 49, 115
libertarianism, 144 as abductive, 6, 90, 91, 94, 95, 99
light, 119 as a priori, 6, 7, 91, 95, 96
linguistics, 24, 305 Aristotle on, 35, 36, 57
historical comparative linguistics, d’Alembert on, 50, 52– 53, 57
235, 251, 252 as deductive, 6, 89, 91, 94, 95,
Lister, J., 118 105– 6
literary history as discipline of trivium, 4, 36, 39,
Bacon on, 43– 44 41, 46
d’Alembert on, 51 Hegel on, 4, 56– 59, 60
literary interpretation, 25, 230n.8 as inductive, 6, 89, 91, 94, 95
comprehensive true accounts of inference in, 5, 7, 9, 25, 29
literary works, 216–17, 218, Kant on, 15, 89
228– 29 law of noncontradiction, 33
concretization in, 217–18, 226– 27, in mathematics, 105– 6
231n.24 as necessary, 6, 91, 95, 96
 Index 347

as objective, 6, 91, 95, 96 mathematics

sound arguments, 93– 94 algebra, 53
valid inference in, 9 and architecture, 22, 289– 90, 292,
logical positivism, 9 294
Long, F., 168 Aristotle on, 36, 39, 45
Lotze, H.: theory of local signs, arithmetic, 33, 34, 45, 58– 59
167– 68 Bacon on, 45, 53, 67n.51
Lovejoy, Arthur A.: The Great calculus, 31, 58
Chain of Being, 66n.45 Comte on, 62– 63
Lyell, Charles, 251 vs. ethics, 202– 3
Lyons, William geometry, 33– 34, 41, 45, 51, 58, 76
The Disappearance of and God, 24, 111
Introspection, 183 Hegel on, 58– 59
on James, 183 incommensurable magnitudes in,
33
Machiavelli, Niccolò: Discourses on and inference, 7– 8, 91
Livy, 44 infinitesimal calculus, 53
MacIntyre, Alasdair inspiration in, 8, 111
on authority, 301– 2 intuition in, 202– 3
on justice and rationality, 306, Kant on, 15, 55
315 motivation of mathematicians, 8,
Three Rival Methods of Moral 108–10, 111
Inquiry, 301– 2 objects of, 33– 34
Mackie, John, 203 and physics, 1, 31, 33
Ma Duanlin: Wenxian tongbao Plato on, 3, 33– 35, 53
(Comprehensive Investigation relationship to natural world, 53,
of the Literary Traditions), 203
280 relationship to time, 105– 6, 110
Maffei, Francesco: Judith, 17, speculation and narration in, 7– 8,
220– 23 102–11
Maienschein, J., 148 stereometry, 34
Maine de Biran, Pierre, 61 symmetry in, 171
Mann, Thomas: The Magic teaching of, 103– 4, 107– 8
Mountain/Mme Chauchat, 226 truth in, 8, 103, 109–10
Mansvelt Beck, B. J., 269 written texts, 105– 6, 110 –11
Many Faces of Beauty, The, 2, 20, 21 Matthews, Robert: on literary and
Margolis, Joseph: on literary art interpretation, 17, 224 – 25,
interpretation, 224– 25 226, 227
Martineau, Harriet: Illustrations of Maximus the Confessor, 312
Political Economy, 144 Mayr, Ernst: on proximate vs.
mass media, 2 ultimate causes, 23, 320n.14
 348 Index

McCullagh, C. Behan: on historical on ethical properties as non-natural,

truth, 249 15, 195– 96, 198, 202, 205, 206
means and ends, 150 on skepticism, 79
mechanical arts, 38– 39, 44, 47, 50, moral affect, 16, 200, 205– 6, 208
51, 61 moral norms, 140, 147, 207– 8
medicine, 53, 54 application to particular
memetics, 148 situations, 144
memory, 247, 253– 54, 314 vs. moral sense/conscience, 146,
Bacon on, 4, 43, 46, 51 148– 52, 153, 154
Reid on, 74, 82 relationship to biological fitness,
Mendelian evolution, 154 11, 153, 154
mental properties vs. physical relationship to cultural evolution,
properties, 7 11–12, 143, 145– 46, 154 – 57
Merleau-Ponty, Maurice, 184 Morley, E., 119
Mesopotamia, 32– 33 Morrison, V., 166
metaphysics Müller-Lyer illusion, 12–13, 164, 165
Aristotle on, 35, 39, 44, 52, 58 Mulligan, Kevin, 197
Bacon on, 45, 49, 51 Murphey, Murray G., 244, 256n.2
d’Alembert on, 51, 52, 58 on constructivism, 240
Descartes on, 47 on Gospel of John and Gospel of
meteorology, 54 Thomas, 253, 254, 257n.7
Michaelis-Menten equation, 133, on historical skepticism and naive
138 realism, 249
Michelson, A., 119 on replication in historiography,
microbiology, 31 241– 42
microorganisms, 118–19 on testimonies, 247
microscopes, 31, 118 on theoretical concept of
Mill, James, 183 historiography, 239
Mill, John Stuart, 183, 195 on types of historiographic
mineralogy, 53, 54 theories, 256n.5
misery, 201– 2 on veracity of historiography,
Mittag, Achim, 276, 281 239– 40
Mongols, 274 Murphy, Peter: on beauty, 294 – 95
monogamy, 155– 56 Murphy, S., 151
Montaigne, Michel de music, 33, 45, 51
Essais, 44 and architecture, 22, 291
on monsters and nature, 44 musical performance, 17
Montesquieu, Baron de, 59 mysticism, 308
Moore, George Edward myth
on ethical intuition, 15, 195– 96, and architecture, 286, 297
197, 199, 202, 205 as truth, 296
 Index 349

Nagel, Jennifer, 257n.7 nuclear magnetic resonance (NMR)

Nagel, Thomas, 182, 186 spectrum, 116
natural history, 31, 43, 44, 51– 52,
232, 233 occasionalism, 48
naturalism, defined, 305 Old Tang History, 273
natural kinds, 42– 43, 50 Olshausen, B. A., 168
natural sciences, 18, 44, 48 ontology, 3, 56
vs. historical sciences, 18, 19, 233, optimization, 175
238– 39 Oriental studies, 32
theories in, 8– 9, 10, 19, 116–19, Osborn, Grant, 25
237, 238– 42, 251– 52 Ouyang Xiu, 274
natural selection, 30, 150, 203, 212, History of the Five Dynasties (Xin
213, 306 Wudai shi/Wudai shiji), 273
Darwin on, 145– 46, 155, 251– 52, New Tang History, 273
304, 305, 321n.21 Ovid, 32
and evolution of religion, 304– 5,
309, 317, 320n.15 paleontology, 18
relationship to moral behavior, 11, Paley, William
143, 145– 46, 147, 148, 153, 155– 57 Natural Theology, 307
natural theology, 24, 45, 301, 307, The Principles of Moral and
308–10, 311, 321n.21, 321n.26 Political Philosophy, 144
Nazi Germany, 309 Panofsky, Erwin
Neanderthals, 141, 144, 153 on history of types
Neoplatonism, 38 (Typengeschichte), 222
neuroscience, 305 on Maffei’s Judith, 17, 222
New Look movement, 164– 65 Papineau, D., 86
Newton, Isaac, 18, 31, 47, 119, 233, 251 paradigm shifts, 9
Ng, On Cho, 267, 271, 272, 273, 275, parental care, 11, 149, 154, 155
279, 280 Parfit, Derek
Nicomachus of Gerasa: Introduction On What Matters, 202
to Arithmetic, 34 on punishment, 203
Nienhauser, William H., Jr., 260 on reasons, 197
Nietzsche, Friedrich on Sidgwick, 195
on interpretation, 218 Park, Seungbae, 251
on truth as illusion, 302, 303 Parmenides, 33
non-natural properties Pascal, Blaise, 47
in ethics, 15, 195– 99, 198, 202– 5, Pasteur, L., 118
206, 207 patriarchy, 316
goodness, 196– 97, 205, 207 Paul, St.: on wisdom of the cross,
Moore on, 15, 195– 96, 198, 202, 318, 323n.44
205, 206 penicillin, 118–19
 350 Index

perception Hegel on, 60

definitions of, 159– 61 and humanities, 1– 2
direct problem regarding, 12, 160, relationship to science, 31, 32
161 subdivision of, 4
and distal stimuli, 12, 159, 160 phlogiston theory, 9, 118
and Fechnerian causal chain, 12, phosphoenolpyruvate (PEP), 122,
159– 60 123, 124, 125– 26, 127, 128, 129,
vs. introspection, 182 130, 132, 133, 134 – 35
inverse problem regarding, 12–13, phylogeny, 232, 233, 235, 241, 251
160, 166, 167– 74 physics, 25, 45, 47, 49, 53, 119
Müller-Lyer illusion, 12–13, 164, and architecture, 22, 290, 297
165 Aristotle on, 4, 35, 36, 37, 39, 44
nativism vs. empiricism in, Big Bang theory, 19, 118, 233– 34,
163– 66, 167– 68, 171, 174 237, 239
as preinterpretive, 17, 218, 220 Comte on, 62, 63
and proximal stimuli, 12, 159– 60 experimental data, 10
role of a priori knowledge in, and God, 24
170– 74 Hegel on, 58, 59
role of inference in, 12, 160– 61 vs. historiography, 18
role of learning in, 163– 64, 166, least action principle, 175
167– 68 and mathematics, 1, 31, 33
role of universals in, 13, 171 model of the atom, 239
scientific study of, 162– 63, quantum theory, 9, 119
165– 66, 172– 75 relativity, 9
vs. sensation, 160– 61 subatomic particles, 31
and symmetry, 171 theories in, 9, 10
as veridical, 12, 160– 62, 169– 70, truth in, 251
171, 174 physiology, 62, 63
peripatetic axiom, 163 Pinker, Steven, 156– 57
Peripatos: systematization of Planck, M., 119
knowledge in the, 32 Plato
Perona, P., 168 analogy of the line, 33– 34
Peter, St., 103 vs. Aristotle, 34, 35
phenomenology, 13–14, 59, 184, 217 on dialectic, 34, 35, 64n.11
inference and cognitive on epistemology, 5
phenomenology, 91, 96– 99 on ethics, 34
See also Husserl, Edmund on Form of the Good, 34
philosophy, 146, 276, 282 vs. Hegel, 58
Bacon on, 43, 45– 47, 48, 51, 52 inference and Platonism, 6, 90,
d’Alembert on, 52– 53 91– 92
 Index 351

on knowledge as justified, true prudence, 313–14

belief, 17 psychologism, 6, 57, 90, 91, 97
on mathematics, 3, 33– 35, 53 psychology, 22, 25, 37, 56, 59, 63,
on piety, 196 196
on poetry, 44 and architecture, 290
on political philosophy, 34 cognitive psychology, 24, 305
on reason, 35, 116 evolutionary psychology, 148
on soul and body, 35 Gestalt psychology, 13, 170, 171,
on understanding and thought, 186
33– 34 intuition in, 199, 200
on universals, 13, 171 psychopathology, 184
Plato, works of role of imagination in, 189– 93
Charmides, 34, 64n.11 role of introspection in, 13, 48,
Critias, 35 181– 85, 188
Euthydemus, 34, 55 role of self-reflection in, 185,
Euthyphro, 196 187– 93
Laws, 65n.19 social psychology, 199, 200
Parmenides, 38 subjective brain in, 181– 82
Republic, 33– 34, 62 subject matter of, 181– 82, 183, 184
Theaetetus, 5 public intellectuals, 2
Timaeus, 35, 38 Pulleyblank, E. G., 261, 271, 279
poetry, 32, 60 Purves, D., 168
Aristotle on, 36, 59, 65n.23 pyruvate (pyruvic acid), 120, 121,
Bacon on, 43, 44– 45, 51 122, 123
Polanyi, Michael, 192 pyruvate kinase, 10, 121, 122– 35
politics, 4, 49
polygamy, 156 Qian Daxin, 275
Porphyrios, Demetri: on classicism, Qing dynasty
295 historical criticism during, 21,
Porphyry, 40 263, 275– 76, 281
positivism, 144, 183, 234 philological school during, 262,
postmodernism, 234, 249 272, 275
pragmatism, 6– 7, 93 quadrivium, 33, 62
private property, 201 quantum theory, 9, 119
probability, 15, 19, 197– 98, 253– 54 Quetelet, Adolphe: Sur l’homme
Problem of Cognitive-Semantic et le développement des ses
Luck, 92 facultés, 31
promise keeping, 195 Quine, W. V. O.
Proverbs 8 on God and wisdom, on theories, 241
316 on truth system, 81
 352 Index

Rahner, Karl: on theology, 24, 306, on trustworthiness of perception,

320n.17 12
Ranke, Leopold von, 233, 234, 237, on universals, 171
250, 251 relativity, 9
rationality, 5, 145, 154, 188 religion, 3, 32– 33
Rawls, John: on reflective evolution of, 304 – 5, 317
equilibrium, 203 truth in, 115, 116
Ray, John: The Wisdom of God, 307 See also biblical studies; faith;
reason, 38, 43, 75 scripture
abstraction, 10–11, 13, 50, 51, 62, Renaissance, 162, 233, 246, 256n.6
91, 142, 151, 152, 168, 171 revelation, 45, 51, 52, 58, 300– 301, 309
Bacon on, 4, 51 rhetoric, 4, 39, 41, 45, 53, 59, 115
and faith, 41, 301 Richerson, P. J., 11, 152
Kant on, 55, 56 Robinson, D. W., 281
Plato on, 35, 116 Roman Catholic social teaching, 312
rational normativity of inference, Roth, Paul A., 245, 246, 256n.6
90, 92– 94, 95– 96, 98– 99, Rousseau, Henri: Representatives of
101n.16 Foreign Powers Arriving to
reasons-internalism, 90, 94 Hail the Republic as a Sign of
truths of, 115, 276 Peace, 234 – 35
See also logic Rowlands, Mark, 184, 191
reductionism, 23, 29– 30, 63, 97 The New Science of the Mind, 189
Reid, Thomas Ruse, M., 148
on common sense, 162
on conception and belief, 77– 78 Scanlon, T. M., 197, 202
on evidence, 74– 76 Schaberg, David, 265
on First First Principle, 5– 6, Schechter, Joshua: on pragmatic
74 – 76, 78– 79, 80, 81, 82 account of justification, 96
as foundationalist, 5– 6, 74– 76 schizophrenia, 14
on intuition, 74– 76 Schlegel, August Wilhelm, 60
vs. Kant, 161 Schlegel, Friedrich, 60
on memory, 74, 82 on architecture as frozen music, 22
on perceptions vs. sensations, Schleiermacher, Friedrich Daniel
160– 61 Ernst, 60
on principles of faculties, 5– 6, Schmidt-Glintzer, Helwig, 264
74 – 76, 78– 79, 81, 82 Schusterman, Richard, 218
on reasoning, 75– 76 Schwartz, Michael Alan, 184
on sensation and conception, 83, scientific ethics, 116
86 scientific laws, 117
on signs, 86 scientific method, 61, 116–17
 Index 353

scientific revolution, 30, 31, 42, 233 Shanhaijing, 265

scientific wonder, 302– 3 Shepard boxes, 169
scripture, 315–16, 317 Sherman, Carolyn, 25
biblical studies, 237, 251, 253, 254 Shijing (Book of Odes), 268
Bonaventure on, 38, 39, 40 Sidgwick, Henry
New Testament, 233 on ethical intuition, 15, 195– 96,
Old Testament, 156, 233 199– 200, 202, 205, 211
Ten Commandments, 156 The Methods of Ethics, 195
wisdom of the cross, 318, 323n.44 as utilitarian, 195
sculpture, 51, 68n.64 Sieyès, Emmanuel Joseph, 79n.79
self-awareness/first-person signal detection experiments, 163,
perspective 164, 172
direct givenness in, 14, 183, 185, Sima Guang: Zizhi Tongjian
187– 88, 190, 191– 92 (Comprehensive Mirror for
general claims based on, 188– 89 Guide in Government), 274,
vs. introspection, 13–14, 182, 183, 278– 79
184– 85 Sima Qian
as pre-scientific/everyday, 185– 87 as founder of Chinese imperial
as scientific self-reflection, 185, historiography, 260 – 62, 268,
187– 93 277– 78, 282
Sellars, Wilfrid on portents and omens, 269– 70
as coherentist, 5, 76– 77, 80 on Shanhaijing, 265
on description of mental states, 86 Shiji, 20, 260 – 62, 264, 265, 268,
on Explanatory 269, 277– 78
System/background system, 6, on sources, 267– 68
78, 80 on unity of historical knowledge,
on language and conceptions, 276– 77
76– 78 Sima Tan, 260 – 61, 264, 268
on myth of the given, 6, 76– 77, Simoncelli, E. P., 168
84– 85 Simpson, O. J., 254
on noninferential belief, 77– 78, skepticism, 16, 183
79– 80 regarding historical truth, 19, 236,
on normative claims, 197 237, 256n.2
on revision of beliefs, 6 regarding inference, 6, 79– 80
on sensory stimuli vs. conception regarding moral knowledge,
of stimuli, 76– 77 203– 4
sexuality, 32 regarding non-natural properties,
Shafer-Landau, Russ, 202 15, 198
Shangshu (Book of Documents), regarding perception, 12
268, 280– 81 Slater, A., 166
 354 Index

Smith, Adam: The Wealth of Stoppard, Tom: on authors vs.

Nations, 53 critics, 228
Snow, C. P., 303 Strimling, P., 152
Sober, E., 148 Strozzi, Bernard, 230n.11
social Darwinism, 147– 48 structuralism, 234
social psychology, 199, 200 subjective idealism, 99
social sciences, 44, 53, 61, 63, 305 subjectivity of others, 14, 185– 86,
and human ecology, 312 187, 188– 89
and imagination, 14 suffering, 317–18
relationship to biology, 10–11 Sumerian astronomy, 33
statistics in, 31 sustainability, 23, 24, 303, 307, 311,
theories in, 256n.5 313–14
sociology, 4, 25, 61, 62, 69n.79, 193 Swets, J. A., 163
solertia, 314
Song dynasty, 273– 74, 278, 280 Tang dynasty, 279, 281
dynastic history during, 275 History Office, 20, 262, 270 – 71,
neo-Confucianism during, 21, 272
262, 275 territorial unity during, 274
Northern Song, 279 Veritable Annals (shilu), 270 – 71
Southern Song, 274, 279, 281 Taylor, Alfred E.: A Commentary
Sosa, E.: on animal vs. reflective on Plato’s Timaeus, 64n.15
knowledge, 82 technology, 9, 61– 62, 67n.52, 117,
space, 184 118, 150
Kant on, 12, 161 telescopes, 31
spandrels, 24, 304, 320n.11 Tertullian: on Athens and Jerusalem,
Spencer, Herbert, 63 303, 319n.6
on evolutionary ethics, 11, textual criticism, 232, 235, 237, 252
146– 47 theatrical arts, 39
on human freedom, 147 theology, 4, 32, 39– 40, 41, 43, 48,
on pleasure and pain, 147 56, 57, 61, 102, 103
The Principles of Ethics, 146– 47 Aquinas on, 23, 300 – 301, 304,
Spinoza, Baruch de, 44 318, 319n.1, 320n.17
Stahl, Georg Ernst: phlogiston Aristotle on First Philosophy,
theory, 9, 118 36, 38, 45, 58
Steinman, Robert M., 175 ecotheology, 303, 311–15
Stelluto, Don, 25 as emancipatory wisdom, 316
Stern, D. G., 86 Hadjadj on, 102, 103
Steup, Mathias: The Analysis of natural theology, 24, 45, 301,
Knowledge, 230n.17 307, 308–10, 311, 321n.21,
Stoicism, 4, 37– 38 321n.26
 Index 355

relationship to biology, 23– 25, Jesus Christ as, 103

300, 303– 5, 309 and knowledge, 16, 17, 94, 208,
relationship to the sciences, 209–10, 223, 224, 230n.17
23– 25, 300– 315, 316–17, 319n.6, as lived vs. seen, 7, 102– 3
321n.26 in mathematics, 8, 103, 109–10
revealed theology, 45, 51, 52, 58, methods in various disciplines,
300– 301, 309 3, 25
and sustainability, 23, 24, 303, 307, myth as, 296
311, 313 objective vs. subjective, 9, 115–17
and wisdom, 24– 25, 307, 313, and Platonism, 91– 92
315–18 relationship to explanation, 80, 81
and wonder, 24, 302– 3, 307–11 in religion, 115, 116
thermodynamics in ligand binding, in the sciences, 8– 9, 102, 103,
126, 127, 129, 132, 134, 135– 37 115–17, 122, 135, 307– 8
Thorndike, E., 165 in sound arguments, 93– 94
time, 184 truths of reason, 115, 276
Kant on, 12, 161 unity of, 103
relationship to mathematics, as wisdom, 24 – 25, 307, 313,
105– 6, 110 315–18
Tinbergen, Nikolaas: on four causes See also historical truth; inference;
of evolutionary explanation, 304 intuition
Tolliver, J., 85 truth telling, 195
Tooby, J., 148 Turing computability, 97
transcendental mentalism, 6, 16, 90, Turner, Derek, 240
91, 92, 96 Turner’s Frontier Theory, 256n.5
trivium, 4, 36, 39, 41, 46 Twitchett, D. C., 270
truth type vs. token events, 238– 39,
as asymptotic search, 9, 117, 120, 240 – 42, 256n.5
122
commitment to, 25 universities, 30
as common-sense knowledge, departments in, 1
115–16 University of Notre Dame
correlation theory of, 249 Institute for Advanced Study,
correspondence theory of, 236, 1– 3, 25
240, 242, 245, 249, 265 School of Architecture, 21
deflationary account of, 212 theological values of, 315
and exemplar representation, utilitarianism, 144, 151, 195
83– 87
as illusion, 303 value judgments, 149, 150 – 51
vs. illusions of sense, 82– 83 van Leeuwenhoek, A., 118
 356 Index

van Pelt, Robert Jan: Architectural Wertheimer, M., 165, 170

Principles in the Age of Westfall, Carroll William:
Historicism, 293 Architectural Principles in the
Varro’s Disciplinae, 65n.32 Age of Historicism, 293
veridicality, 12, 160– 62, 169– 70, 171, Whewell, William, 195
174 White, Hayden: on three levels of
Vico, Giambattista historiography, 244
on culture, 4, 48, 49 Williams, Bernard, 216
on economics, 53 Wilson, D. S., 148
on history, 48, 49, 276 Wilson, Edward O., 148
Principj di scienza nuova, 48– 49, Consilience, 306
59 Winckelmann, Johann Joachim, 60
on tree of sciences, 49 wisdom, 24 – 25, 307, 313, 315–18
Vincent of Beauvais: Speculum Wolff, Christian, 54
Maius, 38, 41 wonder, 24, 302– 3, 307–11
Vogelsang, Kai, 271 Wundt, Wilhelm, 63

Waddington, C. H., 147, 148 Xenocrates: on logic, physics and

Walk, R., 165 ethics, 3– 4, 35, 37– 38
Wandschneider, Dieter: Xiongnu, 278
Technikphilosophie, 67n.52
Wang, Gungwu, 274 Zahavi, Dan, 186
Wang, Q. Edward, 267, 271, 272, Zeno of Citium, 37
273, 275, 279, 280 Zeno of Elea, 33
Wang Shizhen: “Critical Treatise on Zhang Xuecheng
Historical Errors” (Shisheng on Confucian Classics as history,
kaowu), 272 263, 281
warranted belief, 15, 204, 205–10, on cyclicality and evolution, 21,
212–13 276
and knowledge, 16, 17, 94, 208, “General Principles of Literature
209–10, 223, 224, 230n.17 and History” (Wenshi tongyi),
meaning of warranted, 207– 9 276
warranted desire and goodness, Zheng Qiao: Tongzhi
196– 97 (Comprehensive Treatises),
Weber, Max, 63 279– 80
Wedgwood, R., 101n.16 Zhu Xi: Tongjian gangmu, 274 – 75
Weigelin-Schwiedrizik, Usanne, 263 zoology, 53– 54, 59, 62
Wei Zheng, 281 Zuozhuan, 265, 267, 268
“This outstanding collection brings together first-rate scholars,
hailing from disciplines as varied as mathematics, biology, chemistry,
architecture, literary studies, engineering, and theology, and invites
them to reflect on the specific truth experience of their branch of
knowledge. By doing justice to the diversity of this experience and
at the same time to the unity of the quest for truth, this volume
stupendously renews the classical idea of the unity of knowledge,
indeed the beautiful metaphor of the tree of knowledge.”
—JEA N GR ONDIN , U N IVE RS IT Y OF MON TR E AL

The essays in Forms of Truth and the Unity of Knowledge are collected
into five parts, the first dealing with the division of knowledge into
multiple disciplines in Western intellectual history; the second with the
foundational disciplines of epistemology, logic, and mathematics; the
third with explanation in the natural sciences; the fourth with truth
and understanding in disciplines of the humanities; and the fifth with
art and theology.

Contri b utors : Vittorio Hösle, Keith Lehrer, Robert Hanna,

Laurent Lafforgue, Thomas Nowak, Francisco J. Ayala, Zygmunt Pizlo,
Osborne Wiggins, Allan Gibbard, Carsten Dutt, Aviezer Tucker,
Nicola Di Cosmo, Michael Lykoudis, and Celia Deane-Drummond.

Vittorio Hösle is Paul G. Kimball Chair of Arts and Letters in the

Department of German Languages and Literatures and concurrent
professor of philosophy and political science at the University of
Notre Dame. He was founding director of the Notre Dame Institute
for Advanced Study. He is the author or editor of numerous books,
including The Many Faces of Beauty (2013), The Philosophical Dialogue: A
Poetics and a Hermeneutics (2012), and Morals and Politics (2004),
all published by the University of Notre Dame Press.

THE UNIVERSIT Y OF
NOTRE DA ME PRESS
Notre Dame, Indiana 46556
undpress.nd.edu