See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/244957729

A History of Vector Analysis

Article  in  American Journal of Physics · August 1969

DOI: 10.1119/1.1975883

CITATIONS READS
119 5,121

1 author:

Michael Crowe
University of Notre Dame
48 PUBLICATIONS   595 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The Gestalt Shift in Conan Doyle's Sherlock Holmes View project

All content following this page was uploaded by Michael Crowe on 05 January 2015.

The user has requested enhancement of the downloaded file.

 A History of Vector Analysis
Michael J. Crowe
Distinguished Scholar in Residence
Liberal Studies Program and Department of Mathematics
University of Louisville
Autumn Term, 2002

Introduction
Permit me to begin by telling you a little about the history of the book1 on which this talk2 is
based. It will help you understand why I am so delighted to be presenting this talk.
On the very day thirty-five years ago when my History of Vector Analysis was published, a
good friend with the very best intentions helped me put the book in perspective by innocently
asking: “Who was Vector?” That question might well have been translated into another: “Why
would any sane person be interested in writing such a book?” Moreover, a few months later, one
of my students recounted that while standing in the corridor of the Notre Dame Library, he
overheard a person expressing utter astonishment and was staring at the title of a book on display
in one of the cases. The person was pointing at my book, and asking with amazement: “Who
would write a book about that?” It is interesting that the person who asked “Who was Vector?”
was trained in the humanities, whereas the person in the library was a graduate student in
physics. My student talked to the person in the library, informing him he knew the author and
that I appeared to be reasonably sane. These two events may suggest why my next book was a
book on the history of ideas of extraterrestrial intelligent life.
My History of Vector Analysis did not fare very well with the two people just mentioned,
nor did it until now lead to any invitations to speak. The humanities departments at Notre Dame
assumed that my subject was too technical, the science and math departments must have
assumed that it was not technical enough. In any case, never in the thirty-five intervening years
did I ever have occasion to talk on my topic. My response when recently asked to talk about the
subject was partly delight—I had always wanted to do this—but also some hesitation—this was a
topic I researched nearly forty years ago! But it has turned out to be fun.
Publishing the book has also proved interesting. Although it is not for everyone, the
hardbound printing of about 1200 copies gradually nearly sold out, based partly on a number of
very favorable reviews. It is rare that academic books sell that many copies. As it was about to
go out of print, I hit on the idea of asking Dover whether they would want to take it over. This
resulted in its re-publication in 1985 with a new preface updating the bibliography; by that time,
there had appeared a few dozen papers and books shedding new light on various aspects of the
subject. In the early 1990s, a curious development occurred. Nearly twenty-five years after the
book had been published, a research center in Paris (La Maison des Sciences de l’Homme)
announced a prize competition for a study on the history of complex and hypercomplex
numbers). As you can imagine, I was quite pleased to submit my book. Some months later I
was notified that I was being awarded a Jean Scott Prize, which included a check for $4000. At
this point, Dover decided to do a new printing of the book, which includes an announcement of
the prize. In any case, the book has now been continuously in print for 35 years and has led to
all sorts of interesting letters and exchanges.

1This talk is based on the following book: Michael J. Crowe, A History of Vector Analysis: The Evolution of the
Idea of a Vectorial System (Notre Dame, Indiana: University of Notre Dame Press, 1967); paperback edition with a
new preface (New York: Dover, 1985); another edition with new introductory material (New York: Dover, 1994).
Quotations not fully referenced in this paper are fully referenced in that volume.
2Warm thanks to Professor Richard Davitt of the Department of Mathematics at the University of Louisville for his
very helpful comments on drafts of this presentation.
 9/24/08 2

Section I: Three Early Sources of the Concept of a Vector and of Vector

Analysis
Comment: When and how did vector analysis arise and develop? Vector analysis arose only in
the period after 1831, but three earlier developments deserve attention as leading up to it. These
three developments are (1) the discovery and geometrical representation of complex numbers,
(2) Leibniz’s search for a geometry of position, and (3) the idea of a parallelogram of forces or
velocities.

1545 Jerome Cardan publishes his Ars Magna, containing what is usually taken to be the
first publication of the idea of a complex number. In that work, Cardan raises the
question: “If someone says to you, divide 10 into two parts, one of which
multiplied into the other shall produce 30 or 40, it is evident that this case or
question is impossible.” Cardan then makes the surprising comment:
“Nevertheless, we shall solve it in this fashion,” and proceeds to find the roots 5 +
–15 and 5 – –15 . When these are added together, the result is 10. Then he
stated: “Putting aside the mental tortures involved, multiply 5 + –15 by 5 – –15 ,
making 25 – (–15) which is +15. Hence this product is 40.”3 As we shall see, it
took more than two centuries for complex numbers to be accepted as legitimate
mathematical entities. During those two centuries, many authors protested the use
of these strange creations.

1679 In a letter to Christiaan Huygens, Gottfried Wilhelm Leibniz proposes the idea (but
does not publish it) that it would be desirable to create an area of mathematics that
“will express situation directly as algebra expresses magnitude directly.” Leibniz
works out an elementary system of this nature, which was similar in goal, although
not in execution, to vector analysis.

1687 Isaac Newton publishes his Principia Mathematica, in which he lays out his version
of an idea that was attaining currency at that period, the idea of a parallelogram of
forces. His statement is: “A body, acted on by two forces simultaneously, will
describe the diagonal of a parallelogram in the same time as it would describe the
sides by those forces separately.” Newton did not have the idea of a vector. He
was, however, getting close to the idea, which was becoming common in that
period, that forces, because they have both magnitude and direction, can be
combined, or added, so as to produce a new force.

1799 Caspar Wessel, a Norwegian surveyor, publishes a paper in the memoirs of the
Royal Academy of Denmark in which he lays out for the first time the geometrical
representation of complex numbers. His goal was not only to justify complex
numbers, but also to investigate “how we may represent direction analytically.”
Not only does Wessel publish for the first time the now standard geometrical
interpretation of complex numbers as entities that can be added, subtracted,
multiplied, and divided, he also seeks to develop a comparable method of analysis
for three-dimensional space. In this, he fails. Moreover, his 1799 paper fails to
attract many readers. It becomes known only a century later, by which time various

3Girolamo Cardan, The Great Art or The Rules of Algebra, trans. and ed. by T. Richard Widmer (Cambridge:
Massachusetts Institute of Technology Press, 1968), pp. 219–20.
 9/24/08 3

other authors had also published the geometrical representation of complex

quantities.
Comment: It seems somewhat remarkable that in three cases in the period from
1799 to 1828 two authors independently and essentially simultaneously work out
the geometrical representation of complex numbers. This happened in 1799
(Wessel and Gauss), 1806 (Argand and Buée) and 1828 (Warren and Mourey). In
fact, we shall see other cases of independent simultaneous discovery in this history.

1799 Around this time, Carl Friedrich Gauss works out the geometrical interpretation of
complex quantities, but publishes his results only in 1831. Like Wessel, Gauss is
seeking entities comparable to complex numbers that could be used for three-
dimensional space.

1806 Jean Robert Argand publishes the geometrical interpretation of complex numbers,
and in a follow-up publication of 1813 attempts to find comparable methods for the
analysis of three-dimensional space. Also in 1806, the Abbé Buée publishes a
somewhat comparable essay in which he comes close to the geometrical
representation of complex numbers.

1828 England’s John Warren and France’s C. V. Mourey, both writing independently of
the authors who had already published the geometrical representation of complex
numbers, publish books setting forth the geometrical representation of complex
numbers. Warren does not discuss extending his system to three dimensions,
whereas Mourey states that such a system is possible, but does not publish such a
system.

1831 Carl Friedrich Gauss publishes the geometrical justification of complex numbers,
which he had worked out in 1799. Whereas the former five authors on this subject
attracted almost no attention, the prestige and proven track record of Gauss ensures
the widespread acceptance of this representation followed upon his publication.
Ironically, Gauss himself did not accept the geometrical justification of imaginaries
as fully satisfactory. It is also interesting to note that Felix Klein argued in 1898
that Gauss had anticipated Hamilton in the discovery of quaternions, which claim
Peter Guthrie Tait and C. G. Knott vigorously disputed. Grassmann learns of
Gauss’s paper only in 1844 and Hamilton in 1852.

Section II: William Rowan Hamilton and His Quaternions

Comment: Hamilton searched for thirteen years for a system for the analysis of three-
dimensional space, that search culminating in 1843 with his discovery of quaternions, one of the
main systems of vector analysis. This section treats the creation and development of the
quaternion system from 1843 to 1866, the year after Hamilton had died and the year in which his
most extensive publication on quaternions appeared.

1805 Birth of William Rowan Hamilton in Dublin,

Ireland.

1818 Hamilton at age thirteen attains fame for many

intellectual achievements, including being “in
different degrees acquainted with thirteen
languages,” including Greek, Latin, Hebrew,

Hamilton
 9/24/08 4

Syriac, Persian, Arabic, Sanskrit, Hindoostanee, Malay, French, Italian, Spanish,

and German.

1823 Hamilton enters Trinity College, Dublin, placing first in the entrance exam.

1826 Even before the end of an undergraduate career, which had merited him many
awards, Hamilton is named Andrews Professor of Astronomy in the University of
Dublin and Royal Astronomer of Ireland. He holds these positions for the
remainder of his life.

1832 Verification by Humphey Lloyd of Hamilton’s mathematical prediction of internal

and external conical refraction, one of the most famous scientific predictions of the
century. This discovery, which comes out of Hamilton’s very important papers on
“Theory of Systems of Rays,” further enhances his fame.

1835 Hamilton knighted.

1837 Hamilton publishes a long paper interpreting complex numbers as ordered couples
of numbers, an alternate justification of such numbers, which now is seen as
preferable. Hamilton also argues that algebra can be understood as the science of
pure time as geometry is the science of pure space. In that paper, Hamilton
mentions his hope to publish a “Theory of Triplets,” i.e., a system that would do for
the analysis of three-dimensional space what imaginary numbers do for two-
dimensional space. Hamilton had been searching for such triplets from at least
1830. It is significant to note that in this paper Hamilton makes clear that he
understands the nature and importance of the associative, commutative, and
distributive laws, an understanding rare at a time when no exceptions to these laws
were known.

1843 Having searched for his triplets for thirteen years, Hamilton discovers quaternions.
In a letter he later wrote to one of his children about the discovery, he recounts that
his children used to ask him each morning at breakfast: “Well, Papa, can you
multiply triplets?” To this he would reply, “No, I can only add and subtract them.”
On 16 October 1843, his search ends with his discovery of mathematical entities he
calls “quaternions.” These are higher complex numbers of the form a + xi + yj +
zk, where a, x, y, z are real numbers and i, j, and k are three distinct imaginary
numbers obeying the following rules of multiplication: ij = k, jk = i, ki = j, ji = –k,
kj = –i, ik = –j, ii = jj =kk = –1. From this we see that for two quaternions in which
the first part, the real number, is equal to zero
Q = xi + yj + zk and
Q´ = + x”i + y”j + z”k,,
their product
QQ´= – (xx´ + yy´+ zz´) + i(yz´ – zy´) + j(zx´ – xz´) + k(xy´ – yx´).
Hamilton immediately becomes convinced that he had made an important
discovery, stating that “this discovery appears to me to be as important for the
middle of the nineteenth century as the discovery of fluxions [the calculus] was for
the close of the seventeenth.” He proceeds to devote the remaining twenty-two
years of his life to writing one hundred and nine papers and two immense books on
his quaternions.
 9/24/08 5

Comment: One good way (especially in the present context) of describing

Hamilton’s search for quaternions is to state that his search was for numbers with
the following six characteristics, all of which are found in ordinary complex
numbers: (1) associativity for multiplication and division, (2) commutativity for
addition and multiplication, (3) the distributive property, (4) the property that
division is unambiguous, (5) the property that the numbers obey the law of the
moduli,4 (6) the property of being useful for the analysis of three-dimensional
space. Quaternions possess all of six characteristics, with the exception that they
are not commutative for multiplication. One can get a sense of why quaternionists
objected to modern vector analysis when it is noted that modern vector analysis
involves two forms of multiplication, the scalar (dot) and vector (cross) products.
For the scalar product, associativity is irrelevant, and both the law of the moduli and
unambiguity of division must be abandoned. For the vector product, the associative
and commutative properties must be abandoned, division is not unambiguous, and
the law of the moduli fails as well.
Comment: It was around this time that the ideas of the founders of non-Euclidean
geometry, Nicholas Lobachevski and Janos Bolyai, were becoming known. It is
important to realize that Hamilton, by creating the first extensive and consistent
algebraic system that departed from at least one of the standard properties of
traditional mathematics issued in a development that was probably as significant for
algebra as the non-Euclidean systems were for geometry. Perhaps the most
significant message carried by Hamilton’s creation is that it is legitimate for
mathematicians to create new algebraic systems that break traditional rules.
Although some mathematicians resisted this claim, others soon took advantage of it
by creating new algebraic systems.

1846 Hamilton publishes a paper in which he introduces the terms scalar and vector,
referring respectively to the real and the imaginary parts of his quaternion. Thus he
writes regarding a quaternion Q = a + bi + cj + dk, that “Q = Scal. Q + Vect. Q =
S.Q + V.Q or simply Q = SQ + VQ.” In other words, SQ = a, whereas VQ = bi +
cj + dk. This led to quaternionists writing equations such as the following: if we
have two quaternions both having their scalar parts equal to 0, Q = xi + yj + zk and
Q´ = x´i + y´j + z´k, then the laws of quaternion multiplication dictate that SQQ´ =
–(xx´ + yy´+ zz´) and VQQ´ = i(yz´ – zy´) + j(zx´ – xz´) + k(xy´ – yx´). What is
important to note about this is that the scalar portion of this new quaternion can be
seen as mathematically equal to the negative of the modern scalar or dot product,
and the vector part as equal to the modern cross product. This will be very
significant historically; in fact, it was precisely along this path that modern vector
analysis originated.

1847 By this year, Hamilton receives prizes for his discovery from the Royal Irish
Academy and the Royal Society of Edinburgh and publishes at least thirty-four
papers on quaternions, which had been endorsed by some leading mathematical and
scientific figures, including John Herschel.

1853 Hamilton publishes his Lectures on Quaternions, a 737–page volume, not counting
its 64–page largely philosophical preface and 72–page table of contents.

4This law as applied to imaginaries specifies that if three complex numbers combine so that (a1 + b1i)(a2 + b2i) =
(a3 + b3 i), then (a12+ b12)(a22+ b22) = (a32+ b23) ..
 9/24/08 6

1865 Death of William Rowan Hamilton, who by this time had published 109 of the 150
papers that had been published on quaternions. During this period, quaternion
analysis had been much praised but little practiced. Hamilton had, however,
secured one energetic and talented disciple, the Scottish mathematician and scientist
Peter Guthrie Tait, who took up Hamilton’s mantle—or was it, as some thought,
Hamilton’s mania?

1866 Publication of Hamilton’s Elements of Quaternions, which was one and one half
times longer than Hamilton’s immense Lectures on Quaternions.

Section III: Other Early Vectorial Systems, Especially Grassmann’s Calculus

of Extension
Comment: Hamilton was not alone in creating a vectorial system during the period around 1843.
In fact, in that period six other authors from four countries were developing systems that were
more or less vectorial in character. The six men were August Ferdinand Möbius, Giusto
Bellavitis, Comte de Saint-Venant, Augustin Cauchy, Matthew O’Brien, and above all Hermann
Günther Grassmann.

1809 Hermann Günther Grassmann was born in Stettin

in Pomerania, where he spent the majority of his
life. He was a son of a mathematics teacher at
Stettin Gymnasium, Justus Günther Grassmann.

1827 Grassmann enters the University of Berlin, where

he studies mainly theology and philology. After
graduating, he returns to Stettin to study various
subjects including mathematics in preparation for
taking the state examination to qualify as a
teacher.

1827 August Ferdinand Möbius of Leipzig University

publishes Der barycentrische Calcul (The Grassmann
Barycentric Calculus), a system developed around
the idea of taking the centroid of a system of weighted points.

1835 Giusto Bellavitis publishes his first exposition of his system of equipollences,
which has some features in common with the now traditional vector analysis, as is
suggested in his definition of equipollent: “Two straight lines are called equipollent
if they are equal, parallel and directed in the same sense.” His lines in fact behave
in exactly the same manner as complex numbers behave, but it is important to note
that he viewed his lines as essentially geometric entities, not as geometric
representations of algebraic entities; in fact, he was opposed to complex numbers as
“unworthy to belong to a science based on reason alone.” Bellavitis devoted a long
period to an unsuccessful attempt to extend his system to three dimensions.

1836 Grassmann takes a teaching position in Stettin, which he teaches for the remainder
of his life.
 9/24/08 7

1840 Grassmann completes the writing of his Theorie der Ebbe und Flut (Theory of the
Ebb and Flow) and submits this 200+ page essay as evidence of his competence for
teaching. This work on tidal theory contains the first system of spatial analysis
based on vectors and is reasonably close to the modern system. Grassmann dated
the origin of these ideas to 1832 and traced his fundamental idea to reflections on
negative numbers and to the idea of adding and subtracting directed lines. He
traced his idea of a geometrical product to textbooks written by his father and
entitled Raumlehre (Space Theory) and Trigonometrie, the first having been
published in 1824, the latter in 1835. In particular, his father had written in 1824:
“The rectangle itself is the true geometrical product, and the construction of it ... is
really geometrical multiplication.” H. G. Grassmann’s work did not come out of
the geometrical representation of complex numbers tradition; in fact, he learned of
that representation only in December, 1844.
Comment: Grassmann’s 1840 Theorie der Ebbe and Flut presents among other
matters the addition and subtraction of lines (strecken) and also what is numerically
equivalent to the modern cross product, with this difference that whereas the
product of two vectors in the modern system is another vector, in Grassmann’s
system it is a geometrical entity, the directed area of the parallelogram between the
two strecken or vectors. Grassmann also presents in this treatise the “linear
product” of two strecken, this being identical to the modern dot or scalar product.
He also treated vector differentiation.

1844 Grassmann publishes the first full exposition of his system, his Die lineale
Ausdehnungslehre, ein neuer Zweig der Mathematik dargestellt and durch
Anwendungen auf die ubrigen Zweige der Mathematick, wie auch auf der Statik,
Mechanik, die Lehre vom Magnetismus und die Krystallonomie erläutert. Whereas
on the title page of Hamilton’s Lectures on Quaternions, Hamilton was identified
by a large array of titles and memberships, Grassmann’s title page identified him
only as “Lehrer an der Friedrich Wilhelms Schule zu Stettin.” The book attracts
almost no attention and about 600 copies of it were in 1864 used for waste paper.
Comment: Grassmann’s Die lineale Ausdehnungslehre (Linear Extension Theory)
demonstrated deep mathematical insights. It also in one sense contained much of
the modern system of vector analysis. This, however, was embedded within a far
broader system, which included n-dimensional spaces and as many as sixteen
different products of his base entities (including his inner and outer products, which
are respectively somewhat close to the our modern dot and cross products).
Moreover, Grassmann justifies his system by philosophical discussions that may
have put off many of his readers. The abstractness of his presentation and the
originality of his insights also contributed to the difficulties readers had in
comprehending Grassmann’s message, as is evident from comments made by
various mathematicians who had struggled with the book. Möbius, for example,
labeled it unreadable, Baltzer reported that reading the book made him feel “dizzy
and to see sky blue before my eyes,” and Hamilton wrote De Morgan that to read
the Ausdehnungslehre he would have to learn to smoke.

1845 After Grassmann’s unsuccessful efforts to persuade Möbius to write a review of his
book, Grassmann at Möbius’s urging, writes a review of his own book—the only
review his book received! He also publishes a paper containing a new discovery in
electrical theory that he had derived using his new methods. The result: more
 9/24/08 8

neglect, until the 1870s when Clausius made the same electrical discovery and
published it, only then realizing that Grassmann had preceded him.

1845 Adhémar Barré, Comte de Saint-Venant, publishes a short paper titled “Mémoire
sur les sommes et les différences géométriques, et sur leur usage pour simplifier la
mécanique,” in which he lays out a number of the fundamental ideas of vector
analysis, including a version of the cross product, the difference being that his
product was viewed not as another vector but as a spatially oriented area.
Grassmann and Saint-Venant correspond for a time, but Saint-Venant’s ideas do not
seem to have attracted significant attention. They do show, however, that the
search for a vectorial system was “in the air.”

1847 Möbius, hearing of a prize competition sponsored by the Jablonowski Gesellschaft

for investigations fulfilling Leibniz’s idea of a geometry of position urges
Grassmann to enter the competition. Grassmann submits an essay, which wins the
competition and is published in this year, along with some commentary by Möbius.
Result: more neglect. In the same year, Grassmann applies to become a university
professor. The mathematician E. E. Kummer is asked to evaluate Grassmann’s
writings, and finds them very unclear. Grassmann did not get a university position
at that time; in fact, none was ever offered him.

1853 The prominent French mathematician Augustin Cauchy publishes his “Sur les clefs
algébriques,” in which he presents methods he had devised for the solution of
various algebraic problems, for example, finding the roots of equations. Not only
had Grassmann published what were essentially the same methods in his
Ausdehnungslehre, but he had also sent two copies of that book to Cauchy in 1846
in an unsuccessful effort to get a copy to Saint-Venant. Friends alert Grassmann
that there is a priority issue regarding Cauchy’s methods, leading Grassmann to
write the French Academy to arbitrate the priority issue. A three person committee
is formed, which includes Cauchy! Nothing is ever settled, probably because
Cauchy died in 1857.

1852 Matthew O’Brien of King’s College, London publishes the most significant of a
number of his papers setting out in a less than satisfactory manner a system of
vector analysis, which was developed, it seems, partly in terms of Hamilton’s
quaternions. The most serious defect in O’Brien’s system is his failure to
investigate the associativity of his vectorial entities.

1853 Leaving aside Grassmann’s own writings, only two published comments on
Grassmann’s work appeared before the 1860s. One of these consists in the
commentary by Möbius included with the publication of Grassmann’s prize essay;
the other comment appears in the Preface to Hamilton’s Lectures on Quaternions,
published in this year. In 1852, Hamilton somehow learned of Grassmann’s book
and read through it, concluding with much relief that Grassmann had not discovered
quaternions. In his private correspondence he waxes and wanes about the merits of
Grassmann’s insights, but in his Preface he devotes a paragraph to Grassmann,
describing his book as “very original and remarkable” and its author as “profound
and philosophical,” but also stressing that their systems are “distinct and
independent of each other,” although sharing some features.
 9/24/08 9

1860 Luigi Cremona praises Grassmann’s ideas in a published article, thus bringing the
number of authors who had publicly noted his writings to a total of three.

1862 Grassmann, convinced of the merits of his ideas but frustrated by the almost total
neglect of his publications, publishes his system in a new form under the title Die
Ausdehnungslehre: Vollständing und in strenger Form bearbeitet. Three hundred
copies are printed in the shop of Grassmann’s brother and all at Grassmann’s
expense. For this volume, Grassmann wisely decides to remove the philosophical
discussions included in his earlier Ausdehnungslehre and to present his system in
Euclidean dress, a decision that Friedrich Engel, the editor of Grassmann’s works
and one of his two biographers, labeled a “disastrous mistake.” Grassmann himself
wrote in 1877: “this new work met with even less attention than the first.”

1867 In his Theorie der complexen Zahlensysteme, Hermann Hankel, a young and
promising mathematician who had studied with Riemann, praises Grassmann’s
ideas, but soon thereafter dies (1873).

1872 Rudolf Clebsch praises Grassmann’s work in his “Zum Gedachness an Julius
Plücker” published in this year, which is also the year of Clebsch’s death.

1872 Victor Schlegel, who had come to know Grassmann while teaching in Stettin from
1866 to 1868, publishes in this year his System der Raumlehre nach dem Prinzipien
der Grassmann’schen Ausdehnungslehre und als Einleitung in Dieselbe, with a
second part appearing in 1875. This re-presentation of Grassmann’s ideas in a more
elementary form was not, however, very successful. Engel argues that the
biography of Grassmann that Schlegel publishes in 1878 was more influential in
drawing attention to his work. Schlegel continued to champion Grassmann’s idea
for the rest of life (d. 1905), publishing over 25 papers in the Grassmannian
tradition.

1877 Death of Hermann Günther Grassmann, a brilliant if isolated schoolmaster whose

achievements extended far beyond mathematics. For example, in the 1870s, he
published his Wörterbuch zum Rig-Veda (1784 pages) and his translation of the
Rig-Veda (1123 pages), which led to his receiving an honorary doctorate from the
University of Tübingen. He also published textbooks on the Latin and German
languages, various religious and musical writings, and a book on German botanical
terminology.

1878 Because of the increased interest in Grassmann during the 1870s and the short
supply of the earlier Ausdehnungslehre, a second edition of the earlier
Ausdehnungslehre is published in this year.

Section IV: The Middle Period in the Development of the Modern System of
Vector Analysis.
Comment: It is useful to analyze the development of modern vector analysis in terms of three
periods, the first extending up to 1865, by which time the two main traditions, the Hamiltonian
quaternionic and the Grassmannian tradition had arisen. The second or middle period runs from
about 1865 to about 1880. By the beginning of this period, Hamilton (because of his death) and
Grassmann (who concentrated on other areas) had ceased to be major contributors. Other
mathematicians had gradually assumed positions of leadership. In the third period, which began
 9/24/08 10

around 1880, the modern system of vector analysis came into existence through the work of
Josiah Willard Gibbs and Oliver Heaviside and by 1910 had established itself as the dominant
system, although not without a struggle against the Hamiltonian and Grassmannian systems. The
leading figures in this middle period were Peter Guthrie Tait, Benjamin Peirce, James Clerke
Maxwell, and William Kingdon Clifford.
If one asks whether the Hamiltonian or Grassmannian systems was more vigorous in the
period from the early 1840s up to 1900, the answer is certainly the Hamiltonian. Quaternionic
publications in that nearly sixty year period numbered 594, whereas Grassmannian publications
came to 217. Looked at geographically, quaternionic interest was strongest in Britain and
Ireland, with the United States ranking next; Grassmannian publications were primarily written
in German, but both systems had followers far beyond the countries of their origin.
Two key questions should be kept in mind in the remainder of this discussion. It is true that
the Grassmannian system contained within it most of modern vector analysis. Consequently, it is
possible that the modern system of vector analysis could have originated from it. It is also true
that the quaternionic system was significantly different from modern vector analysis, but
possessed some similarities to it. From which system did the modern system originate? And
related to this question, how did this take place? The answers to these questions may prove quite
surprising.

1858 To the aging William Rowan Hamilton, it may

have seemed in 1858 that he would die without a
successor in the quaternionic cause. In that year,
however, he began to correspond with Peter
Guthrie Tait (1831–1901), a Scot who had
graduated from Cambridge University in 1853 as
Senior Wrangler and First Smith’s Prizeman, and
who in 1858 was teaching mathematics at Queen’s
College, Belfast, but who in 1860 became the
Professor of Natural Philosophy at Edinburgh
University. Tait had become very interested in
quaternions, especially in their usefulness for
physical science. One measure of the intensity of
their correspondence is that one letter alone runs Tait
96 pages. Hamilton could hardly have have hoped
for a more energetic successor. By 1859, Tait had published the first of his about
seventy papers on quaternions and soon began writing an elementary presentation
of that system, which he graciously held back from publication until after the 1866
publication of Hamilton’s Elements of Quaternions.

1867 Tait publishes his Elementary Treatise of Quaternions, which went through later
editions in 1873 and 1890 as well as translations into German and French. He also
co-authored with Philip Kelland An Introduction to Quaternions (1873; later
editions in 1882, and 1904). A noteworthy feature of Tait’s Treatise was the
extensive attention that he gave (as Hamilton had not) to physical applications.
Partly for this reason, his books tended to be filled with cases in which the scalar
portion or the vector portion of the full quaternion product was separated out, to the
point that those books look much like modern day vector analysis books, with of
course the major difference that the scalar part of the product of two quaternionic
vectors was the negative of the scalar product in modern vector analysis. Tait
 9/24/08 11

d d d
included extensive treatment of the operator ∇ = i dx + j dy + k dz . In fact,
Maxwell correctly if somewhat enigmatically described Tait as the “Chief Musician
upon Nabla.” All in all, Tait’s book looks very similar to modern vector analysis
books, even in much of its terminology, e.g., the scalar and vector products.

1867 Tait’s best known publication of 1867 was co-authored with William Thomson
(known from 1892 as Lord Kelvin). Sometimes referred to as the Principia of the
nineteenth century, it was their Treatise on Natural Philosophy. Correspondence
between these two leading British physicists shows that Tait had unsuccessfully
urged the inclusion of quaternionic methods in this major publication, Thomson
resisting. In fact, as Thomson wrote in 1901, he and Tait “had a thirty-eight years’
war over quaternions.” Tait also corresponded extensively about quaternions with
another great Scottish physicist and close personal friend, James Clerke Maxwell, in
this case with better but far from complete success (see 1873).

1870 Benjamin Peirce of Harvard publishes in lithograph his Linear Associative Algebra,
described by Dirk Struik as “the first major original contribution to mathematics
produced in the United States.” An early enthusiast for quaternions and the chief
source of the interest in quaternions in the U.S., Peirce in this publication, working
from Hamilton’s discovery of the possibility of new algebras, lays out and classifies
162 different algebras. He describes his goal as developing “so much of hyper-
complex numbers as would enable him to enumerate all inequivalent, pure,
nonreciprocal number systems in less than seven units.”

1873 James Clerke Maxwell (1831–1879) publishes his

Treatise on Electricity and Magnetism, arguably
the most important treatment of the dominant
physical science of the nineteenth century.
Although Maxwell had not used quaternionic
methods at all in working out this four famous
papers on electricity and magnetism, Maxwell, by
1870, partly under the influence of Tait, who had
been his friend since childhood, had begun to read
quaternionic works. Moreover, Maxwell expresses
many of the results presented in his Treatise not
only in Cartesian form, but also in their
quaternionic equivalent. This happened frequently Maxwell
enough that readers could easily assume that
Maxwell himself preferred these methods, but had decided not to force them upon
readers of his book. In fact, Maxwell’s position, stated explicitly in various
publications, was somewhat different. For example, in his famous paper of 1871
titled “On the Mathematical Classification of Physical Quantities,” Maxwell states:
“The invention of the calculus of Quaternions is a step towards the knowledge of
quantities related to space which can only be compared for its importance, with the
invention of triple coordinates by Descartes. The ideas of this calculus, as
distinguished from its operations and symbols, are fitted to be of the greatest use in
all sorts of science.” Some read this as an endorsement of quaternions as a
mathematical method; in fact, what Maxwell was saying was that vectorial methods
provide an invaluable way of thinking, but in actual practice, quaternionic methods
 9/24/08 12

are not satisfactory. What we shall see is that later authors, proceeding precisely
from such statements as this claim by Maxwell and from their own experience in
physical science, proceed to rearrange and shape the quaternionic system into the
modern system of vector analysis.

1877 William Kingdon Clifford (1845–1879) publishes his Elements of Dynamic, an

elementary treatise on mechanics. Clifford, a brilliant if short lived English
mathematician teaching at University College, London, was one of the few
mathematicians in this period who knew both the quaternionic and the
Grassmannian systems. He had lectured on quaternions at University College in
1877 and in 1878 had published a paper employing Grassmannian methods. What
is significant for our purposes in his Elements of Dynamic is that near the middle of
his book, in a section titled “Product of Two Vectors,” he actually introduces two
products, one of which calls the “vector product,” which is the standard vector
product as employed by both Grassmannians and Hamiltonians as well as in
modern vector analysis. His other product, which he calls the scalar product, he
describes qualitatively, but leaves the question of its sign undefined. As we have
seen, this product for Grassmannians was positive, whereas for quaternionists it was
negative. Not only that, it really was not a product at all, but rather nothing more
than a part of the full quaternion product. Nearly a hundred pages later he returned
to his scalar product and defined it as “as the negative sum of the products of their
components along the axes,” i.e., he specified it in a quaternionic manner. What we
see in Clifford is what was already beginning to appear in somewhat different forms
in Tait and Maxwell: a separation of the full quaternionic product into two parts,
along with the treatment of these parts as separate products.

Section V: The Creation of the Modern System of Vector Analysis.

Comment: Two individuals played a key role in the creation of modern vector analysis. They
were Josiah Willard Gibbs and Oliver Heaviside, who independently developed the system that
is almost universally taught at the present time.

V, Part One: Gibbs

1881 Josiah Willard Gibbs (1839–1903) prints the first

half of his Elements of Vector Analysis, which
presents what is essentially the modern system of
vector analysis. This had its basis in the course he
had begun to give at Yale in 1879 on vector
analysis. The draft of a long letter that Gibbs
wrote in 1888 to Victor Schlegel makes very clear
how he came to create his system. From reading
Maxwell’s Treatise on Electricity and Magnetism,
“where Quaternion notations are considerably
used, I became convinced that to master those
subjects, it was necessary for me to commence by
mastering those methods. At the same time I saw,
that although the methods were called
quaternionic, the idea of the quaternion was quite Gibbs
foreign to the subject. I saw that there were two important functions (or products)
called the vector part & the scalar part of the product, but that the union of the two
to form what was called the (whole) product did not advance the theory as an
 9/24/08 13

instrument of geom. investigation.” Gibbs then “began to work out ab initio” a new
form of vector analysis that involved two distinct products as well as various other
features of modern vector analysis. Gibbs then explains to Schlegel that after this,
he came to learn something of Grassmann’s work, first through a paper by
Grassmann on electricity. After learning a little of Grassmann, “I saw that the
methods wh. I was using, while nearly those of Hamilton, were almost exactly those
of Grassmann.” Gibbs recounts that he then procured copies of Grassmann’s
Ausdehnungslehre volumes, struggled with them, but adds: “I am not however
conscious that Grassmann’s writings exerted any particular influence on my VA,
although I was glad enough in the introductory paragraph to shelter myself behind
one or two distinguished names....” In summary, Gibbs adds that that he hopes
Schlegel will be interested to know “how commencing with some knowledge of
Ham[ilton]’s methods & influenced simply by a desire to obtain the simplest
algebra ... I was led essentially to Grassmann’s algebra of vectors, independently of
any influence from him....” Although the point is not mentioned by Gibbs, a side-
by-side comparison of Gibbs’s Vector Analysis and the second edition of Tait’s
Treatise on Quaternions, makes it very clear that Gibbs had learned many of his
methods and much of his notation from that book, and then translated it into the
form of modern vector analysis. Keeping in mind that Gibbs called our scalar or
dot product the “skew product” and wrote it as “α.β,” whereas he called our cross
product the “direct product” and wrote it as “α x β,” we can see the closeness of the
treatments of Gibbs and Tait by examining a few of their equations:
Gibbs (2.21) α.β = β.α
Tait (3;43) Sαβ =Sβα
Gibbs (2.21) α x β = –β x α
Tait (3;43) Vαβ = –Vβα

1884 Gibbs publishes the second half of his Elements of Vector Analysis, which
concentrates on the more advanced parts of vector analysis, especially linear vector
functions, that is, vector functions of such a nature that a function of the sum of any
two vectors is equal to the sum of the functions of the vectors. In doing this, Gibbs
introduces the terms and concepts of “dyad” and “dyadic.” Moreover, during the
1880s Gibbs frequently teaches a course on vector analysis, and does so every year
during the 1890s. In later years, the course consists of as many as 90 lectures.

1886 Gibbs publishes one of his most important and creative papers in mathematics.
Entitled “On Multiple Algebra,” it makes a case for increased attention to multiple
algebra, praises Grassmannian methods, and concludes with the famous line “We
begin by studying multiple algebras; we end, I think, by studying MULTIPLE
ALGEBRA.”

1901 Edwin Bidwell Wilson, a former student of Gibbs, publishes the first book-length
formally published presentation of modern vector analysis in English: Vector
Analysis: A Text Book for the Use of Students of Mathematics and Physics and
Founded upon the Lectures of J. Willard Gibbs, which becomes a classic.

1903 Death of Josiah Willard Gibbs.

V, Part Two: Heaviside

1883 Oliver Heaviside (1850–1925), an Englishman

whose schooling had ended at age 16, begins to
introduce vectorial methods into his writing on
 9/24/08 14

electrical theory, which were in the Maxwellian tradition; indeed, Heaviside is

frequently seen as the chief successor to Maxwell in electromagnetic theory.

1885 Heaviside in one of his electrical papers gives his first unified presentation of his
system of vector analysis, which is essentially identical to that of Gibbs and to the
modern system. Heaviside, known for his wit, later explained how he came to
develop his system, beginning by describing the experiences of a boy who,
enchanted by the word quaternion, tried to learn its meaning by reading Hamilton’s
books.
He took these books home and tried to find out. He succeeded after some trouble, but found
some of the properties of vectors professedly proved were wholly incomprehensible. How
could the square of a vector be negative? And Hamilton was so positive about it. After the
deepest research, the youth gave it up [and] died.
My own introduction to quaternions took place in quite a different manner. Maxwell
exhibited his main results in quaternionic form in his treatise. I went to Prof. Tait’s treatise to
get information, and to learn how to work them. I had the same difficulties as the deceased
youth, but by skipping them, was able to see that quaternions could be explored consistently
in vectorial form. But on proceeding to apply quaternionics to the development of electrical
theory, I found it very inconvenient. ... So I dropped out the quaternions altogether, and kept
to pure scalars and vectors....
Heaviside then recounts that in 1888, he received a copy of Gibbs’s privately
printed text, a sort of “condensed synopsis of a treatise,” finding that it was
essentially the same system to which he had been independently led. He then adds:
“I appeased Tait considerably ... by disclaiming any idea of discovering a new
system. I professedly derived my system from Hamilton and Tait by elimination
and simplification....”
In short, by an almost identical path to that followed by Gibbs and in entire
independence of him, Heaviside had arrived at essentially the same system. It
appears that Heaviside first learned of Grassmann’s system only in 1888 when he
found Grassmann’s name in Gibbs’s text.. There is no reason to think Heaviside
ever read any of Grassmann’s writings.

1893 Heaviside publishes the first volume of his Electromagnetic Theory, which contains
as Chapter 3, “The Elements of Vectorial Algebra and Analysis,” a 173-page
presentation of the modern system of vector analysis. This is the first extensive
published treatment of that system and contains an endorsement of Gibbs’s
presentation although not of his notation. Of course, Heaviside’s presentation
appeared in a specialized book on electrical theory; on the other hand, Heaviside’s
association of vector analysis with the ever expanding area of electrical science was
very helpful in ensuring its spread, as will be evident in what follows.

1925 Heaviside, by then suffering from poverty, deafness, and isolation, dies.

Section VI: A “Struggle for Existence” in the 1890s among the Systems of
Vector Analysis.
Comment: In an 1888 letter, Gibbs predicted that “a Kampf ums Dasein [struggle for existence]
is just commencing between the different methods and notations of multiple algebra, especially
between the ideas of Grassmann & of Hamilton.” Gibbs’s prediction was fulfilled: In the years
1890 to 1894, a widespread and vigorous debate on vectorial methods took place. No less than
eight journals, twelve scientists, and thirty-eight publications came forth. Lord Rayleigh aptly
 9/24/08 15

characterized the spirit of the debate by a paraphrase of Tertullian: “Behold how these vectorists
love one another.” Sample quotations follow in chronological order.

Peter Guthrie Tait, who wrote with vigor but also a certain impatience:

Even Professor Gibbs must be ranked as one of the retarders of Quaternion progress, in virtue of his
pamphlet on Vector Analysis, a sort of hermaphrodite monster, compounded of the notations of Hamilton
and of Grassmann.

Josiah Willard Gibbs, who wrote with almost unwavering tact and good sense:
The merits or demerits of a pamphlet printed for private distribution a good many years ago do not
constitute a subject of any great importance, but the assumptions implied in the sentence quoted are
suggestive of certain reflections and inquiries which are of broader interest, and seem not untimely at a
period when the methods and results of the various forms of multiple algebra are attracting so much
attention. It seems to be assumed that a departure from quaternionic usage in the treatment of vectors is an
enormity. If this assumption is true, it is an important truth; if not, it would be unfortunate if it should
remain unchallenged, especially when supported by so high an authority. The criticism relates particularly
to the notations, but I believe that there is a deeper question of notions underlying that of notations. Indeed,
if my offence had been solely in the matter of notations, it would have been less accurate to describe my
productions as a monstrosity, than to characterize its dress as uncouth.

Oliver Heaviside, who wrote with much insight and at time with scarcely less wit:
... the invention of quaternions must be regarded as a most remarkable feat of human ingenuity. Vector
analysis, without quaternions, could have been found by any mathematician by carefully examining the
mechanics of the Cartesian mathematics; but to find out quaternions required a genius.

Peter Guthrie Tait, who showed the penchant of vectorists for metaphors when late in the debate
responding to Arthur Cayley’s claim on behalf of the importance of using Cartesian coordinates
by comparing quaternions to a pocket map that repeatedly needs to be unfolded:

A much more natural and adequate comparison would ... liken Co-ordinate Geometry ... to a steam-
hammer, which an expert may employ on any destructive or constructive work of one general kind, say the
cracking of an egg-shell, or the welding of an anchor. But you must have your expert to manage it, for
without him it is useless. He has to toil amid the heat, smoke, grime, grease, and perpetual din of the
suffocating engine-room. The work has to be brought to the hammer, for it cannot usually be taken to its
work.... Quaternions, on the other hand, are like the elephant’s trunk, ready at any moment for anything, be
it to pick up a crumb or a field gun, to strangle a tiger, or to uproot a tree. Portable in the extreme,
applicable anywhere ... directed by a little native who requires no special skill or training, and who can be
transferred from one elephant to another without much hesitation. Surely this, which adapts itself to its
work, is the grander instrument! But then, it is the natural, the other the artificial, one.

Comment: One effect of this widespread and colorful debate was to alert the scientific public
that there were a number of vectorial systems, that they were somewhat different, and that major
mathematicians and physical scientists were concerned about the issues. Moreover, as Gibbs and
Heaviside became ever more widely known for their contributions to science, their mutually
supportive advocacy of what became the modern system must have been taken more seriously.
Another offshoot of the debate, or at least of the issues underlying it, was the successful issuing
in 1895 of a call for the formation of what became the International Association for Promoting
the Study of Quaternions and Allied Systems of Mathematics, which from 1900 to 1913
published a journal.5

5Bulletin
of the International Association for Promoting the Study of Quaternions and Allied Systems of
Mathematics.
 9/24/08 16

Section VII: Emergence of the Modern System of Vector Analysis: 1894–1910

Comment: We have seen that by 1893 three main systems of vector analysis had been created
and received substantial attention: the systems of Hamilton, Grassmann, and Gibbs-Heaviside,
the third being the most recent. Which would come to dominance and how would this happen?
The evidence indicates that by 1910, the Gibbs-Heaviside, i.e., the modern system had won out
in the struggle for existence. This section provides some of the evidence for that claim.
Moreover, it suggests that although Gibbs had presented his system with more sophistication and
originality, it was Heaviside’s writings, particularly his association of vector analysis with
Maxwellian electrical theory, that was more influential in establishing the modern system. The
evidence will take the form of a brief examination of eleven major publications presenting the
modern system of vector analysis in the period from 1894 to 1910, and also enough information
on the publishing history, if any, for these books to allow an assessment of their success.

1894 August Föppl publishes his Einfuhrung in die Maxwell’sche Theorie de Elektricität.
Its first three chapters (84 pages) consist of an exposition of Heaviside’s
presentation of vector analysis. Very influential, both in electrical theory and
applied mathematics. Föppl also used vector analysis in other publications.
Subsequent history: 2nd. ed, 1904; 3rd., 1907, 4th, 1912, 1st English ed., 1932,
16th German ed., 1957.

1899 Galileo Ferraris publishes his Lezioni di Elettrotechnica, presenting both electricity
and vector analysis in the Heaviside tradition.

1901 Edwin Bidwell Wilson publishes Vector Analysis: A Text Book for the Use of
Students of Mathematics and Physics and Founded upon the Lectures of J. Willard
Gibbs. This is the first formally published book devoted entirely to presenting the
modern system of vector analysis. Trained as a Harvard undergraduate in
quaternions by J. M. Peirce, Wilson, upon graduation in 1899, proceeded to Yale
where he reluctantly took Gibbs’s vector analysis course, and agreed to write this
book, which became quite successful.
Subsequent history: 2nd. ed, 1909, 8th printing, 1943, paperback reprinting, 1960.

1903 Alfred Heinrich Bucherer publishes his Elemente der Vektor-Analysis mit
Beispielen aus der theoretischen Physik. Author had background in electricity.
This is the first German book devoted solely to presenting the modern system of
vector analysis.
Subsequent history: 2nd. ed, 1905.

1905 Eugen Jahnke publishes his Vorlesungen über die Vektorenrechnung mit
Anwendungen auf Geometrie, Mechanic und mathematische Physik. Jahnke, a
mathematician, draws on both the Grassmannian and Gibbs-Heaviside
formulations.
Subsequent history: no later editions known.

1906 Gibbs’s Elements of Vector Analysis published as part of his collected works, which
also reprinted some of his creative and his polemical articles on vector analysis.
Subsequent history: Paperback reprinting, 1961.

1907 Pavel Osipovich Somoff publishes in Russian the first book in that language on
vector analysis. Explicitly states that he is following in the tradition of “Maxwell,
Heaveside [sic], Gibbs, and Föppl.”
 9/24/08 17

1907 Siegfried Valentiner publishes his Vektoranalysis. A physicist, he includes many

applications. Eclectic in approach, drawing on both the Gibbs-Wilson and the
Heaviside-Föppl formulations.
Subsequent history: 2nd. ed, 1912; 7th ed., 1950; reprint of the 7th ed., 1954.

1909 Cesare Burali-Forti (a mathematician) and Roberto Marcolongo (a physicist)

publish their Elementi di calcolo vettoriale con numerose applicazioni alla
geometria, alla meccanica e alla Fisica-Matematica. Written to a substantial extent
from a Grassmannian perspective, but with some attention to the Gibbs-Heaviside
formulation.
. Subsequent history: No later editions, but a French translation, 1910.

1909 Joseph George Coffin publishes his An Introduction to Vector Methods and Their
Various Applications to Physics and Mathematics. The American physicist Coffin
supplied with this book the need for a shorter and more elementary presentation in
the Gibbs and also Heaviside tradition.
Subsequent history: 2nd. ed, 1911; 1st French ed., 1914; 2nd ed., 6th impression,
1923; 9th reprinting, 1959.

1909–10 W. V. Ignatowsky publishes in two parts his Die Vektoranalysis und ihre
Anwendung in der theoretischen Physik. This book is chiefly in the Heaviside
tradition.
Subsequent history: 3rd ed., 1926.

Comment: It is striking that whereas most of the books using the Gibbs-Heaviside approach
went into a number of later editions, those using the Grassmannian approach attained no later
editions. This suggests that not only in the creation, but also in the acceptance of modern vector
analysis, the Grassmannian tradition played no major role. Moreover, the evidence provided
above shows that although both the Gibbs and the Heaviside traditions were quite influential, the
Heaviside tradition, with its association with electromagnetic theory, was more important.

View publication stats