What Did Mathematics Do To Physics?
What Did Mathematics Do To Physics?
What Did Mathematics Do To Physics?
, xxxix (2001)
INTRODUCTION
384
YVES GINGRAS
in their present disquisitions have rather an Eye to the physical causes of motion,
& the principles thereof, than to the mathematical Rules of it. He then commented
that he considered his hypothesis on motion to be
indeed of the Physical Laws of Motion, but Mathematically demonstrated. For I
do not take the physical & Mathematical Hypothesis to contradict one another at
all. But what is Physically performed is Mathematically measured. And there is
no other way to determine the Physical Laws of Motion exactly, but by applying
the mathematical measures & proportions to them.7
For Wallis the physics of motion was mathematical and he could simply not
understand what Oldenburg meant by separating them or even giving the impression
of opposing them. And Huygens, whose Orologium oscillatorium published in
1673 gave new examples of the geometrization of natural philosophy, was still
complaining to the Marquis de lHpital in December 1692 that:
We nd so few occasions to apply geometry to physics that I often nd that
surprising. For this, with mechanical inventions, is what merits most of our
attention; otherwise, as Seneca said somewhere, we lose our intelligence in
playing with futile calculations.8
Huygens thus makes explicit the relation between mathematics, mechanics and
practical utility that is often present in the tradition of the mixed sciences.
Finally, Newton himself, who followed Barrows lectures and succeeded him in
the Lucasian Chair at Cambridge University, made clear the continuity between the
mixed sciences and natural philosophy in his Optical Lectures of 167072 when he
said about the use of mathematics that
just as astronomy, geography, navigation, optics and mechanics are held
to be mathematical sciences, though they deal with ... physical things, so
although colours belong to physics, nevertheless scientific knowledge of
them must be considered mathematical, in that they are treated through
mathematical reasoning.9
Whereas discourses concerning the mathematization10 of nature have been largely
discussed, following Koyrs lead, through the lens of Platonic philosophical
inuences, the above considerations suggest that a more fruitful approach would
be to see this process as the extension to other elds of the tradition of the mixed
science.11 As recent work has shown, this process had important repercussions on
the transformation of the disciplinary boundaries between mathematics and natural
philosophy in the seventeenth century.12 But here, I would like to concentrate on what
could be called the long term unintended consequences of the use of mathematics
in physics, which have received scant attention from historians of science.13 It is
thus the effects rather than the causes (or reasons) of the mathematization of physics
that I want to follow in this paper.
My starting point will be the publication of Newtons Principia which marks,
conceptually, a radical departure from the then dominant tradition of a mechanical
385
386
YVES GINGRAS
whoever casts a brief glance at the waves striking a beach does not see the
tide mount; but under the supercial to-and-fro motion, another movement
is produced, deeper, slower, imperceptible to the casual observer; it is a
progressive movement continuing steadily in the same direction and by virtue
of it the sea constantly rises.17
This paper will try to describe the rising tide of the mathematization of physics by
calling attention to its largely unintended social, epistemological and ontological
consequences. But before beginning our enquiry, it will be useful to avoid
misunderstandings by saying a word about our methodology.
Over the past thirty years, the social history of science has developed, in
part, on the basis of a strong reaction to what was perceived as a whiggish and
anachronistic view of the development of science and has tended to make very
detailed examinations of episodes that were precisely delimited in space and time.
This microscopic view has greatly enlarged our understanding but it has also had
the consequences of focusing attention almost exclusively on actors categories and
of dissolving any analytical (conceptual) category forged for making sense of
the longue dure. Another aspect of these developments has been the tendency to
limit the analysis of events to an understanding of their immediate context and to
inquire only into the reasons the actors may have had to do what they did. Though
these questions are perfectly legitimate and very interesting in themselves, they
do not cover the whole range of possible questions that an historian of science
may wish to ask of his sources.
In this paper, I will for instance not address the question of what precisely natural
philosophy meant in Newtons days, or why different actors came to believe that
natural philosophy should be mathematical.18 As noted above, I wish to examine the
unintended consequences of this choice, irrespective of the reasons various actors
may have had in making it. For it is plain that actors intentions and programs do
not always square with the consequences of their actions. In fact, many sociologists
would tend to think that most effects result from the unintended consequences of
action.19 And because this question is not only local but in a sense global, since
these effects were not instantaneous but had repercussions over a long period of
time, we do need categories that transcend those used by the actors at a given time
and place. In the seventeenth and eighteenth centuries, for example, the French
used the words physique and physicien where the British still used natural
philosophy and natural philosopher. And though for Newton natural philosophy
may have been about God,20 it, as we shall see below, did not stop his Continental
readers from viewing his work as part of physique or geometry and, therefore,
from putting God aside. Thus, for present purposes, I will use the two words
interchangeably since the focus of our attention is not the meaning of those words
at a given time and place (a meaning that in fact changed over time and place) but
rather the reaction of various actors to the use of mathematics for understanding
inanimate matter, whatever the name that was attributed to a given domain at a
given time or place (physics, mechanics, dynamics, hydrodynamics, electricity,
387
magnetism, etc., are all domains that in the course of time came to be included in the
term physics).21 And what has been said of physics also applies to mathematics:
its content obviously changed over time as well, being essentially geometry and
proportion theory in the seventeenth century, calculus and differential equations
in the eighteenth century, and then vectors, tensors and matrices in the nineteenth
and twentieth centuries, to name only a few of the new tools used in physics at
different times. The term mathematics thus covers them all. In others words, the
analytical categories used in historical inquiry can differ from the categories used
by the historical actors for they depend on the nature of the questions framed. By
being attentive to the scale (micro-, meso-, macro- in space as well as in time) at
which a question is raised and by not confusing analytical categories with actors
categories and by not automatically transferring concepts or terms used at one level
to another, one could more easily eschew articial debates22 about whiggishness
or anachronism.23 This does not mean that the actors categories are not important
but that they are not sufcient for constructing historical narratives in the longue
dure. And only such an approach covering a period roughly from 1700 to 1900
can make visible the social, epistemological and ontological consequences of
the mathematization of physics.
MATHEMATIZATION AND EXCLUSION
Three weeks before the publication of Newtons Principia, Fatio de Duilliers, then
in London, wrote to Huygens that some members of the Royal Society thought that
since the meditations [of Newton] all physics has been changed.24
Notwithstanding the fact that Fatio saw physique where his colleagues saw
natural philosophy, his letter clearly suggests that there was already a feeling
that the publication of the Principia was somehow revolutionary. One thing is
certain: it launched a new phase in the ancient debate over the nature of the relations
between physics and mathematics. Though Galileo preceded Newton in applying
geometry to free fall, he did not concern himself with the efcient cause of that
fall and left that aspect outside his mathematics. The counter-intuitive effects of
the mathematization of physical phenomena only began to be perceived with the
development of dynamics, that is, the mathematization of the concept of force,
as the cause of change in the state of motion.25 Though discussions concerning
the social and epistemological consequences of the Newtonian approach were
(though implicitly) often discussed in the same texts, I will treat them separately.
In this section, I will concentrate on the threat that many practitioners saw in the
ever-growing use of mathematics for their legitimate participation in discussions on
matters of natural philosophy. It must be remembered that Descartess physics and
cosmology as exposed in his Principles of philosophy were essentially qualitative,
a mathematical physics without mathematics to use Mouys expression,26 and
were easily accessible to all those who liked to discuss natural phenomena in a
simple and literary manner. As we shall see, they thus sensed, often confusedly, the
threat that advanced mathematics posed to them, felt excluded from the discussion,
388
YVES GINGRAS
389
a better precision than the purely metaphysical forces of Newton which give
those laws with too much geometrical precision.32
Privat de Molires thus clearly preferred the peu prs of vortex theory to the
articial geometrical precision of Newtons metaphysical forces. His critique of
the excessive precision implicit in Newtons mathematization was well received by
the permanent secretary of the Academy, Fontenelle who, in his annual summary
of the activity of the Acadmie Royale des Sciences called attention to Moliress
argument that, applied to phenomena, physical principles did not have geometrical
precision: M. lAbb de Molires astutely criticizes Newton on his extreme
precision; physical principles are not so precise when we come to apply them
to phenomena.33
This notion that nature does not suffer too much precision, was reminiscent of
Aristotle who noted in the Metaphysics that the minute accuracy of mathematics
is not to be demanded in all cases, but only in the case of things which have no
matter. Hence its method is not that of natural science.34 This idea was still alive
in the eighteenth century. The abb Nollet, for example, explained to his audiences
that it is dangerous for a physicist to develop too great a taste for geometry since
in physics one never nds either precision or certainty.35 This was to take a road
opposite to Newtons, who explicitly stated in his Preface to the Reader that errors
do not come from the art but from those who practise the arts; and that if anyone
could work with the greatest exactness, he could develop an exact mechanic.36 For
only if nature is exact can it be understood through mathematics.
An excellent example of the growing difculties attending those who, in the
middle of the eighteenth century, wanted to contribute to the discussions on the
nature of physical phenomena without using a mathematical approach is offered
by the writings of Cadwallader Colden (16881776). A colonial administrator
and politician, he had studied medicine in London before migrating to America
where he had a successful career, becoming Lieutenant Governor of New York in
1761. A friend of Benjamin Franklin, he also had scientic and literary ambitions
and corresponded with many well-known gures of his time. His major book, The
history of the ve Indian nations depending on the Province of New York, was
published in 1727 and frequently reprinted. In 1745, he published in New York An
explication of the rst causes of action in matter: and of the cause of gravitation.
He sent copies to his London friend Peter Collinson, a Fellow of the Royal Society
of London, who gave one to the Society and distributed the rest across Europe
using his correspondence network.37
The interest shown in the book by a cultivated public is suggested by its being
reprinted in London the year after and by the publication in 1748 of a partial Dutch
translation printed in Leipzig and Hamburg.38 Colden also published an expanded
edition of his views in London in 1751 under the new title, Principles of action in
matter, the gravitation of bodies and the motion of the planets, explained of those
principles. A French translation was published that same year in Paris.39 Extracts
also appeared in the Gentlemans magazine.40
390
YVES GINGRAS
Probably conscious that his theories of matter and gravity would be perceived as
those of a dilettante, he insisted that he had been constantly employed in business
and so could not pursue his studies other ways than by way of amusement.41 In
a letter to Collinson, Colden even mentioned that times are very improper for the
speculations in which I employ my thoughts since we are all in great hurry in
preparing to attack the French Settlements in Canada.42 More importantly, Colden
insisted in his essay that with a competent skill in astronomy, equations may be
formed for the motion of the planets, without any aid of the conic sections, or any
other knowledge, besides the common rules of arithmetic and trigonometry. 43
Clearly distancing himself from the mathematical treatment of gravitation by
geometers, he insisted he had more in view to convey his conceptions clearly to
others, than to elegance in style; or the pomp of a demonstrative method.44 Despite
the interest shown in his ideas on the cause of gravity, however, Colden attracted
no disciple other than his son. In view of this, the fact that someone of Leonhard
Eulers stature took the time to mercilessly pull the book to pieces, to use the
words of Coldens biographer, is a signicant indication that the frontier separating
insiders from outsiders (the rst thus becoming the expert or professional,
the latter the amateur or dilettante), was still not clearly dened in the middle
of the eighteenth century.45
Euler had obtained a copy of the book from the Reverend Caspar Wetstein, a
member of Benjamin Franklins and Peter Collinsons circles. Eulers comments are
contained in his letter to Wetstein where he noted that the American philosopher
had little knowledge of the principles of motion and this entirely disqualies the
author from establishing the true Forces requisite to the motion of the planets from
whatever cause he may attempt to derive them.46 Eulers reaction is particularly
interesting for he was among those who believed in the existence of a uid whose
pressure could explain gravitation. But as he wrote to Le Sage, who had also tried
his hand at a mechanical model of gravitation, the theory of uid movement is
not enough studied yet to produce a complete explanation.47 In other words, a
mechanical explanation of gravity had to take, even for Euler, a mathematical form,
in his case through the laws of hydrodynamics.48 Though it was intuitive, verbal
physics was thus no longer a legitimate way of providing a physical explanation
when it was inconsistent with the mathematical laws of physics.
In excluding Colden as a man who has not entirely devoted himself to the study
of [the cause of gravity] and who ignored the rst principles of hydrostaticks,49
Euler made plain that the price of entry into the club of legitimate practitioners
was an adequate knowledge of mathematics. As the frontier delimiting the eld
of activity became better dened, contributors who proposed new explanations
of phenomena that did not take the elds implicit rules into account were met
with silence or dismissed without further argumentation as being beside the point
(and outside the eld). The historical dynamic of the creation of these boundaries,
constitutive of scientic elds, varies of course with domains and the nature of the
tools used to investigate objects. Whereas mathematics helped dene the boundaries
391
of rational mechanics (and later electricity and magnetism), in other elds, like
chemistry for example, the sophistication and cost of instruments and reagents
played a more important role than mathematics in dening the cost of entry to
the eld. Once the boundaries were well dened and the gate keepers controlled
access to the legitimate places of publication, outsiders could make their voices
heard only in books or in magazines of general interest. Thus, Colden could still
ask his friend Collinson to try to have his answer to Euler presented to the Royal
Society (Collinson had told him that Pro-Euler remarks were read before the
Royal Society), but having failed to have his views printed in the Philosophical
transactions of the Royal Society, he was reduced to urging him to do his best to
have them appear in the Gentlemans magazine.50
THE RISE OF A PRIVATE SCIENCE
For some, the critical stand toward the use of advanced mathematics in physics
was related to their conviction that science had to be accessible to a wide circle
of people. As noted by Shapin, Boyle repeatedly remarked upon the relative
inacessibility of mathematics, the use of which restricts the potential size of
the audience.51 In 1690, at a time when he was still trying to explain gravity by
mechanical means, Varignon also noted that treating physics through geometry
makes the former unintelligible to those untrained in the latter. Asking himself
rhetorically if one could understand the mysteries of nature without the help of
geometry, he concluded that the only thing he could do to help the general reader
was to provide a general plan of his treatise without giving the details of his
demonstrations. His Nouvelles conjectures sur la pesanteur were thus preceded by
a Discours sur la pesanteur with a different pagination.52
For the Jesuit Louis-Bertrand Castel, for example, who was part of the editorial
board of the Journal de Trvoux from 1720 to 1745, science had to be accessible to
the common man and thus not only higher mathematics but fancy experimentations
were to be excluded from proper scientic methods. One could read in his Journal,
probably under his pen, that the experiments capable of perfecting physics, ought
to be easy to make and to repeat at any time, and almost by everyone.53 His Vrai
systme de physique de M. Isaac Newton printed in 1743 was subtitled la porte
du commun des physiciens, that is, accessible to the common physicist.54 On
Newtons prism experiments he noted, no doubt with some exaggeration, that
in order to make these experiments on the refraction of light correctly one
must be a millionaire.55 In his comment on Castels Trait de la pesanteur
published in 1724, the abb de Saint-Pierre already insisted that it was important
to make discoveries accessible to the common reader; the thing was difcult, he
conceded, but necessary.56 As late as the 1770s, we still nd echo of this demand
for democratic apparatuses in Priestleys complaint that Lavoisiers instruments
were too costly and too delicate to constitute a norm for chemical proof. 57 The
privatization of science thus took many roads, reaching new limits today with
the construction of a single accelerator capable of producing the most esoteric
392
YVES GINGRAS
393
394
YVES GINGRAS
were by much the greater number in the Royal Society.73 In his Dictionnaire
de physique, published in 1761, Paulian had also taken notice of the fact that
it was now impossible to become a physicist without a minimum knowledge
of mathematics.74
THE RESENTMENT OF THE EXCLUDED
Though the mathematical way to physics represented rst and foremost by Newtons
Principia was largely accepted by those mid-eighteenth-century gomtres who
devoted themselves to extending its applications to terrestrial and astronomical
phenomena, there was still a vocal group of people who were more interested in
verbal explanations than in mathematical calculations. For them, less talented or
less interested in investing long hours in abstruse mathematical calculations, the
legitimacy of their verbal contributions to physics was threatened by views like
those put forward by dAlembert and they did not leave the stage without protest
even after 1750. Massire, for example, may have perceived himself as such an
outsider, for he admitted writing on a scientic subject without being a savant. In
1759, he published his critical reections on the system of attraction, a 400-page
book which, he admitted, owes its existence only to the boredom and idleness of
the countryside.75 Though he had not read the Principia but used Voltaires and
sGravesandes books on Newtonian physics as a substitute, he was nonetheless
shocked by what he found: I saw that this part of Newtons philosophy [on the
movement of the planets] consisted only in calculations; and it seemed to me that
this was not the trade of a philosopher.76 Having noted that Newtonians calculate
everything, he added: for me, who am not a calculator, I must admit that I felt
myself revolted against this new kind of philosophy.77
The ubiquity of geometry and equations in many physics treatises provoked
Massires into asking if there was not some affectation in this manner of
philosophizing on the part of their authors. And like Colden before him, he told his
readers that any man who knows what are right angles, obtuse angles and acute
angles can understand the essential part of his book and that he himself did not
know enough algebra to speak the language of the savants.78 Declaring, following
many others, the Newtonian conception of gravitation to be occult, he echoed
Castels view, observing ironically that to make the answer acceptable, there was
only to add to it some calculations.79
The Comte de Lacpde, who presented himself as a member of the Academies of
Dijon, Toulouse, Rome and many others, was also keen to ridicule attraction when
he wrote that, though a sensitive attraction may seem obscure,
following the example of the great Newton, I will envelop my hypothesis within
geometric and algebraic veils to make it invisible to the eyes and criticisms of
the uninitiated. If someone object to this obscurity, I will cite M. dAlembert and
others who modestly admit that there are propositions in the masterpiece of the
English philosopher that stop even the most gifted geometers.80
395
Citing dAlembert was deliberately ironic for, as we have seen, he was strongly
opposed to the physics of the peu prs. And as late as 1826, for example, in
an attempt to explain gravitation in mechanical terms, J. Mangin, member of the
Philomatic Society of Verdun, could still write:
I know that all the analytical calculations of the defenders of this system [of
attraction] are able to frighten many readers but it remains nonetheless true
that all these calculations are only based on suppositions since the physical
cause of attraction is unknown.81
About ten years later, Antoine-Louis Gunard Demonville tried in vain to have
his papers on the true system of the world accepted by the French Academy of
Science. Frustrated, he directed his venom at Denis-Simon Poisson, denouncing the
dictatorship of mathematics. As he wrote in the Preface to his book:
M. Poisson is a mathematical monomaniac who will admit no new truth if
he cannot nd it already sketched in one of his axioms of algebra ... under
how many mistakes will he not try to suffocate us in emptying his magazine
of formulas!82
When mathematics came to be applied to other elds such as architecture and
the design of bridges and buildings, there were also strong reactions. In 1805, for
example, the French architect Charles-Francois Viel even published a pamphlet
titled De limpuissance des mathmatiques pour assurer la solidit des btimens.
For him, the abuse of mathematics simply led to ill-constructed bridges that could
not withstand the test of time like those that had been constructed by the ancients,
who had never used sophisticated calculations to build their marvels.83 And as we
shall see below, even Faraday looked upon the mathematization of electricity and
magnetism with a sceptical eye.
All these adverse and often aggressive reactions (and many others could be
cited84) clearly show that there was real resentment from those who used to see
themselves as part of an open intellectual space in which they exchanged views
on matters of natural philosophy. Putting too much mathematics into physics
thus restrained not only potential readers, as Boyle observed, but also potential
contributors. In this process, the boundaries of what was a relatively open public
space of discussion were slowly redened in such a way that potential readers were
increasingly limited to potential contributors, that is, to those with the appropriate
training. In other words, mathematization contributed to the formation of a relatively
autonomous scientic eld, with its control of access mechanisms.85
We still nd an echo of the debates on the importance of mathematics in physics
in Jean-Baptiste Biots Introduction to his Trait de physique exprimentale et
mathmatique of 1816, where he wrote that:
Many people, in France and elsewhere, believe that Physics must be presented
in a purely experimental form, without any algebraic apparatus. The English,
so eminent in this science, think we use too much calculation and complicate
396
YVES GINGRAS
it with our formulas instead of clarifying it. Many of them, who are very
skilled and very exact, believe that the precision that we think we approach
[using calculations] is purely ideal, since it goes innitely beyond the limits to
which these experiments are inevitably subjected. This question is fundamental
and merits to be debated.86
He then went on to offer a detailed and strong twelve-page defence of the usefulness
of mathematics in physics to which we will return below.
isol at ing far aday
Though mathematics came to dominate rational mechanics, it must be emphasized,
as the time span covered here suggests, that the process was indeed very slow.
Moreover, the mathematization of other elds like electricity or magnetism also
had exclusionary effects which were again met with negative reactions.87 Up to the
middle of the nineteenth century, the study of heat, electricity and magnetism was
relatively accessible to empirically- or qualitatively-minded people of the kind who
liked to debate the nature of gravitation in the seventeenth and eighteenth centuries.
They were then, though only for a limited time, protected from the recondite
language of mathematics. In fact, the difference in the level of mathematical
codication between gravitation and electricity in the mid-eighteenth century
certainly explains in part the success of Benjamin Franklin as compared to the failure
of his friend Colden, who occupied a similar social position and had a comparable
intellectual training but chose to attack a eld already dominated by geometers,
instead of devoting himself to simpler empirical phenomena.
In 1831, the young James David Forbes, who two years later would become
Professor of Natural Philosophy at the University of Edinburgh, had already
understood that in the present state of Science, a liberal basis of mathematical
knowledge is indispensable to [the] successful prosecution [of the physical
sciences].88 At the time of his nomination, he confessed to his mentor William
Whewell that
any doubt as to the propriety of viewing mixed mathematics as belonging
to a natural philosophy class is at this moment peculiarly untenable: for the
whole progress of general physics is happily so fast tending to a subjection
to mathematical laws of that department of science, that in no very long time
magnetism, electricity and light may be expected to be as fully the objects of
dynamical reasoning as gravitation is at this present time.89
Mathematics was still making headway in physics and Forbess teaching would
prepare students accordingly.
Twenty years later, when Maxwell sent his paper On Faradayss lines of forces
to the eponymous natural philosopher, the level of mathematics was already
high and Faraday admitted to Maxwell that I was at first almost frightened
when I saw such mathematical force made to bear upon the subject and then
397
wondered to see that the subject stood it so well.90 Writing again eight months
later he asked Maxwell:
There is one thing I would be glad to ask you. When a mathematician engaged
in investigating physical actions and results has arrived at his conclusions,
may they be not expressed in common language as fully, clearly, and denitely
as in mathematical formulae?91
Recalling that Faraday fashioned himself as a natural philosopher92 and that he
never used any mathematics for arriving at his fundamental discoveries, we can
understand his uneasiness when confronted with Maxwells formidable treatment
of his intuitions. Like many of those who in the eighteenth century protested
against the use of advanced mathematics in physics, Faraday, a lecturer at the
Royal Institution, was accustomed to presenting his work to the public through
experiments and images and he must have felt the exclusionary effect of Maxwells
advanced mathematics. In his paper On the conservation of force, in which he
suggested that gravitation must be explained by lines of force or by an ether, he
was clearly on the defensive when he noted that it may be supposed that one who
has little or no mathematical knowledge should hardly assume a right to judge of
the generality and force of the principle of gravitation. But his answer was that
mathematics cannot of itself introduce the knowledge of any new principle.93
Criticizing Faraday for misunderstanding the concept of potential energy, Ernst
Brcke noted that it is a long time since such a far reaching physical question has
been touched upon wholly without the aid of mathematical apparatus without
the assistance of those wonder-working symbols whose brief rhetoric speaks more
convincingly to the mind than the tongue of Cicero or Demosthenes.94
The central place of mathematics in physics had been explicitly stated by
Maxwell just a year before Faraday expressed his views on the issue. In his
inaugural lecture at Aberdeen in November 1856, Maxwell insisted that for
him natural philosophy is and ought to be mathematics, that is, the science in
which laws relating to quantity are treated according to the principles of accurate
reasoning, a conviction he repeated word for word four years later in his inaugural
lecture at Kings College.95
Thus, over a period of nearly two centuries, the progressive mathematization of
various domains of physics had the effect of excluding as legitimate practitioners
most of the readers (and sometimes contributors) of the scientic and literary
magazines who used to talk about natural phenomena without using the language
of mathematics. But as we have seen, they did not leave the place without reacting
angrily to their progressive marginalization. By the middle of the nineteenth century,
with the development of energy physics, electrodynamics and thermodynamics,
a large part of physics had become the esoteric knowledge of a small group of
students trained by the incumbents of physics chairs then developing in European
universities and later by the members of physics departments of North American
universities.96 To quote Bachelard, where Castel and Nollet had readers, Biot,
398
YVES GINGRAS
Forbes and Maxwell had students.97 Combined with a tighter control of access to
the membership of scientic academies and the emergence of specialized scientic
disciplinary journals, these institutional settings would now dene the boundaries
of a private sphere where the legitimate practice of the discipline would carry
on. These developments simultaneously constructed an outside where all those
still interested in explaining gravity, squaring the circle or nding a perpetuum
mobile, could search an audience for their theories without interfering with what
was going on inside the eld. It is signicant that the Paris and Berlin Academies
ruled out such topics in 1775, the Royal Society having done so earlier in 1749.98
While Copernicus could defend himself against religious critiques of his system by
claiming that mathematics is for mathematicians, mid-nineteenth physicists could
silence amateurs by claiming that physics is for (mathematical) physicists.
THE CHANGING MEANING OF EXPLANATION
In addition to making the access to the practice difcult and time consuming, the
mathematization of physics also had a more subtle epistemological effect that
rst became perceptible in the debate over the mechanical explanation of gravity.
Though the debate over that question is now relatively well known,99 the approach
to this episode has concentrated on the various solutions and models proposed to
save the validity of a mechanical and plenist cosmology from the mathematical
objections rst put forward by Newton. I believe, however, that the profound
signicance of that debate lies in the fact that it was the very meaning of the term
explanation that was at stake in the discussions concerning the legitimacy of
hypotheses and in the contested interpretations of Newtons famous hypothesis
non ngo. This episode shows that the evaluation criteria for what was to count
as an acceptable explanation (of gravitation in this case) were shifting towards
mathematics and away from mechanical explanations.100 Confronted with a
mathematical formulation of a phenomenon for which there was no mechanical
explanation,101 more and more actors chose the former even at the price of not
nding the latter. This was something new. For the whole of the seventeenth
century and most of the eighteenth, to explain a physical phenomenon meant to
give the physical mechanism involved in its production. Hence, Descartes could
still reject Galileos law of free fall because it was not based on a mechanical
explanation of gravity.102
The publication of Newtons Principia marks the beginning of this shift where
mathematical explanations came to be preferred to mechanical explanations when
the latter did not conform to calculations. This shift in meaning helps explain the
strong reactions against what were perceived as occult explanations on the part
of partisans who based their cosmology on contact forces, which in turn implied
that the universe had to be a plenum (or else constantly bombarded by particles).
The nature of the shift of meaning implicit in Newtons approach to physics can
be compared to the shift Kepler induced when he invented a celestial physics,
which was clearly a contradiction in terms in the context of scholastic natural
399
philosophy wherein astronomy was distinct from physics.103 And in point of fact,
Maestlin plainly told Kepler that such a notion was contrary to common sense and
good philosophy.104 Likewise, in stating that since Proposition LII of Book II of his
Principia proved by geometrical calculations that no plausible vortex movements
could be made compatible with Keplers laws and that all celestial phenomena could
be understood using the law of universal gravitation, Newton was in practice saying
that mathematics was replacing verbal formulations as the nal arbiter and true
explanation of phenomena. As Kepler had mixed physics with astronomy, Newton
had mixed physics with mathematics and thus explained physical phenomena
mathematically.105 Whereas astronomers got used to the rst, natural philosophers
now had to adapt to the second.
Like Keplers before him, Newtons category mistake was not long in being
recognized.106 As is well known, the review of the Principia in the Journal des
savans in 1688 praised the authors geometry but concluded that to make his work
as perfect as it was possible Newton now had to produce a physics as exact
as his mechanics, which, it was added, could be done only by substituting true
motion in place of those he has supposed.107 In other words, only true mechanical
contact between the parts of a plenum could be considered a physical explanation,
whereas Newton had simply posited mathematical forces, which pertains to a
different order of things, namely geometry, and thus could not constitute a true
explanation of physical phenomena.
Though many authors criticized Newton for not giving a physical explanation
of gravity, they rarely went beyond denouncing him for bringing back occult
qualities or even miracles. It is probably the Jesuit Castel who analysed in
greatest detail what Newton was really doing with his new approach. As a devoted
Cartesian, he was in fact well placed to see how Newtons explanations differed
from those usually given by mechanical philosophers. He not only perceived the
exclusionary effect of the higher mathematics used by Newtonians, as we have seen,
but he also clearly pointed to the problematic nature of the explanations that Newton
offered. And though a Fellow of the Royal Society of London, he was sometimes
perceived as moiti fou, moiti sens, as Diderot said. 108 However, his often
extreme opposition to Newtonian physics, though perceived as an embarrassment by
many of his Jesuit colleagues and other less dogmatic followers of the Cartesian way
in physics,109 constitutes in fact an excellent condens of what was so disconcerting
from the point of view of the paradigm of mechanical explanations that dened
the dominant schemes of perception and evaluation, that is, the scientic habitus110
of those who reacted to Newtons Principia. Moreover, as his biographer pointed
out, Castel probably represents an historically signicant majority opinion111 of
those who read the Journal de Trvoux and similar literary journals and who were
actively interested in the sciences along the lines developed by Descartes in his
Principia philosophiae, that is, science as a kind of verbal physics based on clear
and distinct ideas. This is the universe that Castel realized was being threatened
by Newton and this is why he devoted an entire book to dissecting the logic of the
400
YVES GINGRAS
401
402
YVES GINGRAS
By explicitly excluding from the purview of his work the kind of mechanical
explanation that was still seen by many as a legitimate part of physics, dAlembert
was suggesting that physicists should resist the furor to explain everything,
introduced into physics by Descartes,127 thus claiming that from now on any
explanation had to take a mathematical form. For him, Newton had been the rst to
show the art of introducing geometry into physics and to form, through the union
of experience and calculation, an exact, profound, luminous and new science.128
By contrast, Castel was saying that the price to pay for such a science was too
high if it meant the abandonment of physical, that is mechanical, explanations
of physical phenomena.
In 1752, only a few seasons after dAlembert had declared that vague
conjectures were at last banished from physics, some Cartesians tried to provide
the public with the best preservative against the seduction of what we nowadays
call Newtonianism by printing Fontenelles Thorie des tourbillons cartsiens.129
Echoing Castels analysis, the editor wrote in his Preface:
There is, so to say, two very different worlds; one mathematical, the other
physical. The mathematical, which we can also call the metaphysical, only
exists in the ideas of the geometer: he supposes the innitely small, dots without
dimensions, lines without width...; as well as vacuum and gravitation. All
these suppositions are the basis of a calculation which without them could
not be exact and which without this exactitude could not be demonstrative.
But nothing of this can be found exactly in nature ... and this is a strange
illusion to abuse of the abstractions in transposing them in the physical world
as if they were real beings.130
Thirty years later, in their Physique du monde, De Marivetz and Goussier could
still complain that the phrase to calculate a phenomenom was very improper and
had been introduced into physics by people who are better at calculations than at
explanations.131 The central role of mathematics in physics was thus still being
contested and dAlembert was conscious of the revolution initiated by Newton. In
his Essai sur les lmens de philosophie, he explained that the generation which
opposed the revolution, and of which Castel was an outspoken representative,
nally died or stayed quiet in the academies while it was left to a new generation,
of which he was part, to complete the revolution:
when the foundations of a revolution are laid it is nearly always in the next
generation that the revolution is completed, rarely before because the obstacles
vanish instead of yielding; rarely long after because once the barriers are
traversed the human spirit develops rapidly until he meets another obstacle
that forces him to stop again.132
For his generation and the successive ones, the word explanation had simply
acquired a different meaning from the one it had previously held and the controversies
that had surrounded the existence of vortices and occult qualities were thenceforth
out of season. This new meaning was also closely associated with the redenition
403
404
YVES GINGRAS
arrives at the utmost limit of distinctness. We cannot express physical facts except in
a mathematical form.139 He wanted to teach his students not only the mathematical
accuracy of expression of which all physical facts are capable, but the mathematical
necessity of their interdependence.140 He believed that
transformation of symbolic expressions is most essential to physical science
but it is in reality pure mathematics. Everything connected with the original
[physical] question may be dismissed from the mind during those operations,
and the mathematician to whom they are referred may be doubtful whether
his results are to be applied to solid geometry, to hydrostatics or to electricity.
But as we are engaged in the study of Natural Philosophy we shall endeavour
to put our calculations into such a form that every step may be capable of
some physical interpretation, and thus we shall exercise powers far more
useful than those of mere calculations the application of principles, and
the interpretation of results.141
Through his teaching, Maxwell, like all those who before him promoted a
mathematical physics, was thus contributing to the institutionalization of a practice
of physics heavily connected with and dependant upon mathematics. But as the
mathematics (the syntax) became more developed, Maxwells wish to give a
physical interpretation (a semantic) to every step of the calculations, a view strongly
promoted by Whewell,142 became progressively attenuated in the work of successive
generations of physicists. Moreover, in addition to making possible abstraction and
generalization, the manipulation of symbols discussed by Biot and Maxwell were to
have an important if often undesired and disturbing ontological effect.
Albeit in a cryptic manner, Maxwell seems to have perceived the tendency toward
more and more abstract kinds of explanations provided in physics when he told
J. A. Fleming that the progress of science was indicated by our making our terms
mean less and less. By which he meant, according to Fleming in his Recollections
of Maxwell, that whereas older physicists talked of electric uids, caloric, etc.,
we speak simply of electrication, heat, etc and decline to commit ourselves as
to what electrication, etc., is.143 As P. M. Harman has shown, Maxwell makes
a disjunction between the nature of substances and the framework of dynamical
principles and he asserts the sufciency of a purely symbolic or functional mode
of representation.144 The basis of this formal approach was the use of Lagrangian
formalism, which as Joseph Larmor explained, allows one to ignore or leave
out of account altogether the details of the mechanisms, whatever it is, that is in
operation in the phenomena under discussion.145 Thus the use of more general
and more abstract mathematical formulations played against the tendency to
conceive matter in substantial terms, the emphasis being put on the relational
aspects, which for Maxwell were the most important things to know.146 In other
words, the syntactical structure of the theory affected and constrained the semantic
interpretation of its terms.
The long history of mechanical explanations of gravity provides a good example
405
406
YVES GINGRAS
CONCLUSION
In this paper, I have tried to show that the mathematization of physics had long-term
social, epistemological and ontological effects on the discipline. A similar
analysis could be made of the famous debate concerning the non-visualizability
of quantum mechanical phenomena in the 1920s. One would then see that it was
strictly analogous to the debate over vortices or the ether, for the disappearance
of these substances had the effect of making gravitation and light propagation
hardly anschaulich: their understanding depended essentially on mathematical
formalisms.155 Thus it is not very surprising that David Bohm, a strong advocate
of a realist (a better word would be substantialist) interpretation of quantum
mechanics, wrote in the mid-1980s that the current emphasis on mathematics has
gone too far and that physics may have taken a wrong direction in giving so much
emphasis to its formalism.156 Though Bohms views were very marginal at the
time,157 they remind us, in the end, that the question of the relationship between
physics and mathematics is still being debated158 and one could fruitfully follow its
effects in contemporary physics.159 And since there is no reason to think that
these effects were limited to physics, the framework of analysis suggested here could
be used to look at the effects of mathematics on other disciplines like chemistry
and biology. From J. J. Sylvester and A. Cayley in the 1870s, who used advanced
mathematics to understand molecules and isomers, to the emergence of quantum
chemistry and mathematical biology, mathematics has had the tendency of
redirecting the focus of inquiry towards the relational character of the elements, thus
contributing to the transformation of concepts and practices.160
But only a more detailed analysis could show that the desubstantialization of
matter was directly related to the mathematization process itself which distanced the
meaning of the concepts from their original intuitive referents. Through their formal
manipulation as mathematical symbols, concepts thus acquired a relational denition
and lost their original substantial quality while gaining in generality.161
ACKNOWLEDGMENTS
This paper was written during my stay in Boston as a Fellow at the Dibner Institute
for the History of Science and Technology in winter 2000. Special thanks are due
to Jed Buchwald and Evelyn Simha for their hospitality at the Dibner. Many of my
colleagues there and in the Boston area were generous enough to comment on a
previous draft of the paper: Babak Ashra, Davis Baird, Mario Biagioli, Kenneth
Caneva, Mordechai Feingold, Evelyn Fox Keller, Sam Schweber, Ana Simoes.
I thank all of them very much, as well as my colleagues Raymond Fredette,
Robert Nadeau, Claude Rosental and the two anonymous referees for their
stimulating suggestions. Needless to say, I am solely responsible for the views
presented here, and for the limitations and errors they may still contain. This
research has been supported by a grant from the Social Science and Humanities
Research Council of Canada.
407
REFERENCES
1. Aristotle, Physics II.2 194a 89.
2. W. R. Laird, Patronage of mechanics and theories of impact in sixteenth-century Italy, in Bruce
T. Moran (ed.), Patronage and institutions: Science, technology, and medicine at the European
court 15001750 (Rochester, 1991), 5166.
3. There is a considerably large literature on the mixed sciences, but, to my knowledge, there is still
no general account of its extension to elds other than the canonical ones (Astronomy, Optics,
Harmonics) as suggested here. For the case of Galileo, see Peter Machamer, Galileo and the
causes, in R. E. Butts and J. C. Pitts (eds), New perspectives on Galileo (Dordrecht, 1978),
16180; James G. Lennox, Aristotle, Galileo, and mixed sciences, in W. Wallace (ed.),
Reinterpreting Galileo (Washington, D.C., 1986), 2951; W. R. Laird, Galileo and the mixed
sciences, in Daniel A. Di Liscia, E. Kessler and C. Methuen (eds), Method and order in
the Renaissance philosophy of nature (Aldershot, 1997), 25370. More generally, see Paul
Lawrence Rose, The Italian renaissance of mathematics (Geneva, 1975); Peter Dear, Discipline
and experience: The mathematical way in the Scientic Revolution (Chicago, 1995); Jean-Marc
Mandosio, Entre mathmatiques et physique: Note sur les sciences intermdiaires la
Renaissance, in Comprendre et maitriser la nature au Moyen Age: Mlanges dhistoire des
sciences offerts Guy Beaujouan (Geneva, 1994), 11538; Richard D. McKirahan, Aristotles
subordinate sciences, The British journal for the history of science, xi (1978), 197220;
John J. Cleary, Aristotle and mathematics (Leiden, 1995); Gary I. Brown, The evolution of
the term mixed mathematics, Journal of the history of ideas, lii (1991), 81102; James
W. Garrison, Newton and the relation of mathematics to natural philosophy, Journal of the
history of ideas, xlviii (1987), 60927.
4. B. le Bovier de Fontenelle, Sur lutilit des mathmatiques et de la physique et sur les travaux
de lAcadmie des Sciences, Oeuvres compltes, vi (Paris, 1994), 3750; John Arbuthnot,
An essay on the usefulness of mathematical learning, in George A. Aitken (ed.), The life
and works of John Arbuthnot (Oxford, 1892), 40935; Boyles 1671 essay On the usefulness
of mathematics to natural philosophy is in vol. iii of The works of the Honourable Robert
Boyle, ed. by Thomas Birch (London, 1744), 42534. For Bacons views on mathematics, see
Graham Rees, Mathematics and Francis Bacons natural philosophy, Revue internationale
de philosophie, xl (1986), 399426.
5. For Boyles views on mathematics see Steven Shapin, Robert Boyle and mathematics: Reality,
representation and experimental practice, Science in context, ii (1988), 2358.
6. Cited by Antoni Malet, Isaac Barrow on the mathematization of nature: Theological voluntarism
and the rise of geometrical optics, Journal of the history of ideas, lvii (1997), 26587, pp.
2801. Barrows Essay was rst published in Latin in 1683. On the tradition of the mixed
sciences in England, see J. A. Bennett, Christopher Wren: Astronomy, architecture, and the
mathematical sciences, Journal for the history of astronomy, vi (1975), 14984.
7. Wallis to Oldenburg, 5 December 1668, in A. R. Hall and Marie B. Hall (eds), The correspondence
of Henry Oldenburg (Madison and London, 196586), v, 221.
8. Huygens to Marquis de lHpital, 29 December 1692, in C. Huygens, Oeuvres compltes (The
Hague, 18881950), x, 354; the quotation from Seneca was calculis ludimus, in supervacuis
subtilitas tertur. On Huygens, see Joella G. Yoder, Unrolling time: Christiaan Huygens and
the mathematization of nature (Cambridge, 1988).
9. Alan S. Shapiro (ed.), The optical papers of Isaac Newton (Cambridge, 1984), i, 86; cited
by Allan Gabbey, Newtons mathematical principles of natural philosophy: A treatise on
mechanics?, in P. M. Harman and Alan E. Shapiro (eds), The investigation of difcult
things (Cambridge, 1992), 31213.
10. I distinguish quantication from mathematization. The rst refers to the production of numbers
408
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
YVES GINGRAS
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
409
of physics, chap. 4 of her book Science in culture: The early Victorian period (New
York, 1978), 11136; Thomas Kuhn, The essential tension (Chicago, 1977), 6064; John
L. Heilbron, Experimental natural philosophy, in G. S. Rousseau and Roy Porter (eds),
The ferment of knowledge: Studies in the historiography of eighteenth-century science
(Cambridge, 1980), 35787.
I think, for example, that much of the discussion about the non-existence of the Scientic
Revolution is based on a confusion of levels and categories of analysis.
For a recent discussion of that question see Nick Jardine, Uses and abuses of anachronism in
the history of the sciences, History of science, xxxviii (2000), 25170. Though I essentially
agree with the author, I would not use the term anachronism to refer to the application of
sociological categories to past events (p. 261). Since analytical categories are dened by
analysts they need not have been part of the actors repertoire. It is thus a category mistake to
call them anachronistic since they are not used as if they were actors categories. Otherwise
we would have to call anachronistic any explanations of past diseases invoking virus or
microbes before the discovery of these entities. I know of no historian ready to pay that price.
And it should be clear that these categories do not prevent us from looking at how the actors
themselves explained these diseases in the absence of the categories of virus or microbes.
They are simply different levels of analysis.
Fatio to Huygens, 24 June 1687, in Huygens, Oeuvres (ref. 8), ix, 1678.
On the concept of force, see Richard S. Westfall, Force in Newtons physics: The science of
dynamics in the seventeenth century (London, 1971); Max Jammer, The concept of force: A
study in the foundations of dynamics (Cambridge Mass., 1957).
Paul Mouy, Le dveloppement de la physique cartsienne, 16461712 (Paris, 1934), 144.
Journal de Trvoux, x (1710), 356.
On Varignon, see Michel Blay, La naissance de la mcanique analytique: La science du mouvement
au tournant des XVIIe et XVIIIe sicles (Paris, 1992).
Jacqueline de la Harpe, Le Journal des Savants et lAngleterre, 17021789 (University of California
Publications in Modern Philology, xx, no 6; Berkeley, 1941); Alfred R. Desautels, Les Mmoires
de Trvoux et le mouvement des ides au XVIIIe sicle, 17011734 (Rome, 1956); John N.
Pappas, Berthiers Journal de Trvoux and the philosophes, Studies on Voltaire and the
eighteenth century, iii (1957), 1363; Jean Ehrard and Jacques Roger, Deux priodiques
franais du 18e sicle: Le Journal des Savants et les Mmoires de Trvoux, in G. Bolleme et al.
(eds), Livre et socit dans la France du XVIIIe sicle (Paris, 1965), 3359; C. Lennart Carlson,
A history of the Gentlemans Magazine (Providence, 1938); Albert Pailler, Edward Cave et le
Gentlemans Magazine (17311754) (Lille, 1975).
In his classic book La formation de lesprit scientique (Paris, 1938), Gaston Bachelard studied
typical works of 18th-century science but only to contrast them with modern 19th- and
20th-century science. Completing this approach, I wish to follow the process of exclusion that
transformed scientic practice and led to the situation described by Bachelard as scientic as
opposed to what he called the pre-scientic spirit of the 18th century. It is obvious that I do
not have to use these normative categories in order to describe that process. For a more detailed
discussion of Bachelards views in relation to our project, see Yves Gingras, Mathmatisation
et exclusion: Socio-analyse de la formation des cits savantes, in Jean-Jacques Wunenburger
(ed.), Gaston Bachelard et lpistmologie franaise (Paris, in press).
For a detailed analysis of Leibnizs reaction to Newtons mathematization of natural philosophy,
see Domenico Bertoloni Meli, Equivalence and priority: Newton versus Leibniz (Oxford,
1993). As he explains, Leibniz stressed the insufciency of purely mathematical laws [and]
the need for physical explanations ... (p. 24); see also Yves Gingras, La dynamique de
Leibniz: Mtaphysique et substantialisme, Philosophiques, xxii (1995), 395405. In the 1730s
410
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
YVES GINGRAS
Jean Bernoulli also devoted two prize-winning essays to the question of the physical cause
of gravitation, trying to reconcile vortex motions with the mathematical laws of Kepler and
Newton. For him, Newtons vacuum and attraction were incomprehensible for a physicist
who had to search the causes of the facts; Essai dune nouvelle physique cleste in
Opera omnia (Geneva, 1742), iii, 2667. For details, see William Shea, The unnished
revolution: Johann Bernoulli (16671748) and the debate between the Cartesians and the
Newtonians, William Shea (ed.), Revolutions in science: Their meaning and relevance
(Canton, 1988), 7092.
Mmoires de lAcadmie Royale des Sciences, 1733, 311.
Histoire de lAcadmie Royale des Sciences, 1733, 94.
Aristotle, Metaphysics, 995a, 1518.
Cited by R. W. Home, The notion of experimental physics in the early eighteenth-century, in J. C.
Pitt (ed.), Change and progress in modern science (Dordrecht, 1985), 10731, p. 124.
Isaac Newton, Principia, translated by I. Bernard Cohen and Anne Whitman (Berkeley,
1999), 381.
P. Collinson to C. Colden, 27 March 1747, in The letters and papers of Codwallader Colden,
iii: 17431747, Collections of the New York Historical Society for the year 1919 (New
York, 1920), 368.
Cadwallader Colden, Principles of action in matter, the gravitation of bodies and the notion
of the planets, explained of those principles (London, 1751), preface. Also mentioned in
C. Colden to Dr Betts, 25 April 1750, in The letters and papers of Codwallader Colden,
iv: 17481754, Collections of the New York Historical Society for the year 1920 (New
York, 1921), 204.
Cadwallader Colden, Explication des premires causes de laction de la matire et de la cause de
la gravitation (Paris, 1751). Since Colden was, through his relation with Franklin, in indirect
contact with Abb Nollet, the latter may have been at the origin of the translation.
Gentlemans magazine, December 1752, 499500, 5701, 58990; and January 1753, 6566.
Colden, Principles of action in matter (ref. 38), 3.
C. Colden to P. Collinson, 20 June 1745, in The letters and papers of Codwallader Colden,
iii (ref. 37), 119. The Anglo-American eet was then attacking the French settlement of the
LouisBourg fortress which capitulated on 26 June.
Colden, Principles of action in matter (ref. 38), 2. A similar statement is also found in the
1745 edition, p. v.
Ibid., 3.
Alice M. Keys, Cadwallader Colden: A representative eighteenth century ofcial (New York,
1906), 1314.
Euler to Wetstein, 21 November 1752, vera copia in the letter from P. Collinson to Colden, 7 March
1753, in The letters and papers of Codwallader Colden, iv (ref. 38), 356.
Euler to Le Sage, 16 April 1763, in Notice de la vie et des crits de George-Louis Le Sage.
Redige daprs ses notes par Pierre Prvost. Suivie dun opuscle de Le Sage sur les
causes nales; dextraits de sa correspondance avec divers savants et personnes illustres
(Geneva, 1805), 386.
On Eulers views on mechanical explanation, see Curtis Wilson, Euler on action-at-a-distance
and fundamental equations in continuum mechanics, in Harman and Shapiro (eds), The
investigation of difcult things (ref. 9), 399420.
Euler to Wetstein, op. cit. (ref. 38), 356.
On these exchanges, see The letters and papers of Codwallader Colden, iv (ref. 38), 378 (letter
from Collinson, March 10, 1754,), 395396 (Colden to Collinson, July 7 1753); the citation is
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
411
on p. 406 (Collinson to Colden Sept. 1, 1753). He seems not to have succeeded for I could nd
nothing on gravitation by Colden in the Magazine after the extracts from his book appeared in
January 1753. In his letter to Colden on 10 March 1754, Collinson noted that your answer to
pro: Euler is not yett publisd..., ibid., 378.
Shapin, op. cit. (ref. 5), 42.
P. Varignon, Nouvelles conjectures sur la pesanteur (Paris, 1690). It is interesting to note that
though historians have studied in detail Varignons contribution to analytical mechanics, they
pass over in silence this essay totally devoted to a mechanical explanation of gravity. Somehow
in the 1690s, Varignon seems to have had a conversion to the mathematical approach and to
have completely abandoned this project.
Journal de Trvoux, xlii (1742), 1093; George R. Healy, Mechanistic science and the French
Jesuits: A study of the response of the Journal de Trvoux (17011762) to Descartes and
Newton, Ph.D. dissertation, University of Minnesota, 1956, p. 198, attributes, plausibly, the
text to Castel, though it is not signed.
Pre Louis Castel, Vrai systme de physique gnrale de M. Isaac Newton. A la porte du
commun des physiciens (Paris, 1743).
Cited by Bachelard, op. cit. (ref. 30), 230.
Cited by Donald S. Schier, Louis-Bertrand Castel, anti-Newtonian scientist (Iowa, 1941), 113.
J. Golinski, Precision instruments and the demonstrative order of proofs in Lavoisiers chemistry,
Osiris, n.s., ix (1994), 3047 and idem, Science and public culture: Chemistry and enlightenment
in Britain, 17601820 (Cambridge, 1992), 138.
By contrast, the relatively wide distribution of X-ray apparatus at the end of the nineteenth
century made that phenomena accessible to non-professional physicists, see Yves Gingras,
La rception des rayons X au Qubec: Radiographie des pratiques scientiques, in Marcel
Fournier, Yves Gingras and Othmar Keel (eds), Sciences et mdecine au Qubec: Perspectives
sociohistoriques (Sainte-Foy, 1987), 6986.
Castel exchanged letters with Diderot on his Lettre sur les aveugles; see Schier, op. cit.
(ref. 56), 48.
See John Pappas, Lesprit de nesse contre lesprit de gomtrie: Un dbat entre Diderot et
dAlembert, Sudies on Voltaire and the eighteenth century, lxxxix (1972), 122953; and more
generally T. L. Hankins, Jean dAlembert: Science and the Enlightenment (Oxford, 1970);
Michel Paty, DAlembert (Paris, 1998).
D. Diderot, De linterprtation de la nature, Oeuvres philosophiques (Paris, 1961), 177244,
p. 216.
See for example ibid., 214 where he writes: nos faiseurs de cours dexprience ressemblent
un peu celui qui penserait avoir donn un grand repas parce quil aurait eu beaucoup
de monde sa table.
Histoire de lAcadmie des Sciences, 1745, 28; cited by Pierre Brunet, Les physiciens hollandais et
la mthode exprimentale en France au XVIIIe sicle (Paris, 1926), 132.
Journal encyclopdique, February 1769, 131.
J. R. DAlembert, Introduction aux recherches sur la prcession des quinoxes et sur la nutation
de laxe de la Terre dans le systme newtonien, in Oeuvres compltes (Geneva, 1967), i,
43750, p. 437. On Buffons critiques of mathematics, also published in 1749, see Jacques
Roger, Buffon (Paris, 1989), 2635.
DAlembert, Introduction, 450.
See for example ibid., 344, 353.
Ibid., 438.
DAlembert, Discours prliminaire ou analyse des recherches sur diffrents points importants du
412
YVES GINGRAS
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
413
the study of physical phenomena was a way of imposing the legitimate denition of a eld as
well as the legitimate methods of inquiry. On boundary work, see Thomas F. Gieryn, Cultural
boundaries of science (Chicago, 1999).
Jean-Baptiste Biot, Trait de physique exprimentale et mathmatique (Paris 1816), p. xi. On Biot,
see Eugene Frankel, J. B. Biot and the mathematization of experimental physics in Napoleonic
France, Historical studies in the physical sciences, viii (1977), 3372.
Heilbron, Experimental natural philosophy (ref. 21), 36775.
J. D. Forbes to W. Whewell, 29 May 1831, cited by Crosbie Smith, Mechanical philosophy and the
emergence of physics in Britain, Annals of science, xxxiii (1976), 329, p. 25.
J. D. Forbes to W. Whewell, 8 August 1833, cited by Smith, op. cit. (ref. 88), 27. On Whewells
attitude towards the relation between physics and mathematics, see Harvey W. Becher, William
Whewell and Cambridge mathematics, Historical studies in the physical sciences, xi (1980),
148; Menachem Fish, A philosophers coming of age: A study of erotetic intellectual
history, in Menachem Fish and Simon Schaffer (eds), William Whewell: A composite portrait
(Cambridge, 1991), 3166.
Faraday to Maxwell, 25 March 1857; P. M. Harman (ed.), The scientic letters and papers of
James Clerk Maxwell (Cambridge, 1990), 548.
Faraday to Maxwell, 13 Nov. 1857; ibid., 552, note 13.
Iwan Rhys Morus, Different experimental lives: Michael Faraday and William Sturgeon,
History of science, xxx (1992), 128.
M. Faraday, On the conservation of force, Philosophical magazine, 4th ser., xiii, no. 86,
April 1857, 22539, p. 238.
E. Brcke, On gravitation and the conservation of force, Philosophical magazine, 4th ser.,
xv, no. 86, February 1858, 8190, p. 82.
Harman, op. cit. (ref. 90), 429, 671.
On the physics discipline, see Kenneth L. Caneva, From galvanism to electrodynamics: The
transformation of German physics and its social context, Historical studies in the physical
sciences, ix (1978), 63169; David B. Wilson, Experimentalists among the mathematicians:
Physics in the Cambridge natural sciences tripos, 18511900, Historical studies in the
physical sciences, xii (1982), 32571; Olesko, op. cit. (ref. 84); Christa Jungnickel and
Russell McKormach, Intellectual mastery of nature: Theoretical physics from Ohm to Einstein
(Chicago, 1986); Robert Silliman, Fresnel and the emergence of physics as a discipline,
Historical studies in the physical sciences, iv (1974), 13762; R. Sviedrys, The rise of
physics laboratories in Britain, Historical studies in the physical sciences, vii (1976),
40536; D. Kevles, The physicists: The history of a scientic community in modern America
(New York, 1978); Yves Gingras, Physics and the rise of scientific research in Canada
(Montreal and Kingston, 1991).
Gaston Bachelard, Lactivit rationaliste de la physique contemporaine (Paris, 1951), 42.
James E. Mclellan III, Science reorganized: Scientic societies in the eighteenth century (New
York, 1985), 360 (his chap. 7 deals with tighter control of access to membership in scientic
academies); on specialized scientic journals, see ibid., 2579, and J. E. McLellan, The
scientic press in transition: Roziers Journal and the scientic societies in the 1770s, Annals
of science, xxxvi (1979), 42549.
Pierre Brunet, Lintroduction des thories de Newton en France au XVIIIe sicle: Avant 1738
(Paris, 1931); A. J. Aiton, The vortex theory of planetary motion (London, 1972); Henry
Guerlac, Newton on the Continent (Ithaca, 1981), chap. 3; Alexandre Koyr, Newtonian
studies (London, 1965).
For an excellent discussion of changing notions of explanation in the history of physics, see
Stephen Gaukroger, Explanatory structures: Concepts and explanation in early physics
414
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
YVES GINGRAS
and philosophy (Hassocks, 1978). For the case of mathematics, see Michael S. Mahoney,
Changing canons of mathematical and physical intelligibility in the later 17th century,
Historia mathematica, ii ( 1984), 41723.
For the different interpretations given by historians to the term mechanical philosophy, see Floris
Cohen, op. cit. (ref. 11), 1425. Here I mean by mechanical explanation one that provides an
efcient cause based on contact forces; two roads were open: if void was admitted, that force
could be obtained through corpuscular interactions (as in Lesages theory of gravitation); if
void was excluded it could be through the action of a uid (as in Eulers theory) or through
the movement of corpuscules of different size lling all space as in Descartess system. From
this point of view, Newtons mathematization of gravitation was a demechanization of the
world picture and not a mechanization as suggested by E. J. Dijksterhuis, The mechanization
of the world picture (Princeton, 1986).
William R. Shea, Descartes as critic of Galileo, in Butts and Pitt (eds), New perspectives
on Galileo (ref. 3), 13959.
R. S. Westman, The astronomers role in the sixteenth century: A preliminary study, History
of science, xviii (1980), 10547; for a detailed analysis of the sixteenth-century debate on
the relation between natural philosophy and astronomy and on the nature of astronomical
explanation, see N. Jardine, The birth of history and philosophy of science: Keplers A
defence of Tycho against Ursus with essays on its provenance and signicance, rev. edn
(Cambridge, 1988).
A. Koyr, La rvolution astronomique (Paris, 1961), 364.
On Newtons style of mathematical physics, see I. Bernard Cohen, The Newtonian revolution
(Cambridge, 1980); P. A. Kroes, Newtons mathematization of physics in retrospect,
in P. B. Scheurer and G. Debrock (eds), Newtons scientific and philosophical legacy
(Dordrecht, 1988), 25367.
This is a category mistake from the point of view of the actors. For us of course, it is no longer
a category mistake since we have accepted Newtons view of what physics is. The same
applies to Keplers celestial physics.
Journal des savans, 2 August 1688, 154; on the first reviews of the Principia, see I. B.
Cohen, The review of the rst edition of Newtons Principia in the Acta eruditorium, with
notes on the other reviews, in Harman and Shapiro (eds), The investigation of difcult
things (ref. 9), 32353.
Cited by Schier, op. cit. (ref. 56), 199.
Note that the actors used the categories of Cartesians and Newtonians in a variety of ways.
A common feature to all is, I think, that the rst must involve a mechanical explanation of
all phenomena in the sense dened above (ref. 101) while the other can content itself with a
mathematical formulation and leave unanswered the question of the mechanical action.
On habitus, see Pierre Bourdieu, The logic of practice (Stanford, 1990).
Schier, op. cit. (ref. 56), 58.
Ibid., 89.
Castel, op. cit. (ref. 54), 13.
Ibid., 37. On Huygenss reaction to Newtons Principia, see Roberto De A. Martins, Huygenss
reaction to Newtons gravitational theory, in J. V. Field and Frank A. J. L. James (eds),
Renaissance and revolution: Humanists, scholars, craftmen and natural philosophers in early
modern Europe (Cambridge, 1993), 20313.
Castel, op. cit. (ref. 54), 52.
Ibid., 94.
Ibid., 95, italics in the original.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
415
Ibid., 97.
Ibid., 9899.
Ibid., 121.
Ibid., 121.
Ibid., 253.
Ibid., 304.
Ibid., 302.
Ibid., 348.
DAlembert, Introduction aux recherches sur la prcession des quinoxes et sur la nutation de laxe
de la Terre dans le systme newtonien, in Oeuvres compltes (ref. 65), i, 43750, p. 450.
DAlembert, Essai sur les lmens de philosophie, in Oeuvres compltes (ref. 65), i, 115348,
p. 345.
Ibid., 341.
B. le Bovier de Fontenelle, Oeuvres (Paris, 1996), vii, Prface de lditeur, 37782, p. 377.
Ibid., 378.
Baron de Marivetz et Goussier, Physique du monde (Paris, 1780), v, 57, cited by Bachelard, La
formation de lesprit scientique (ref. 30), 231.
DAlembert, op. cit. (ref. 127), i, 341. This views recalls Max Plancks when he wrote: A
new scientic truth does not triumph by convincing its opponents and making them see the
light, but rather because its opponents eventually die, and a new generation grows up that
is familiar with it; Plancks own scientic autobiography and other papers, transl. by F.
Gaynor (New York, 1949), 3334.
Ibid., 346.
Diderot, op. cit. (ref. 61), 216.
Jean-Baptiste Biot, Trait de physique exprimentale et mathmatique (Paris 1816), p. xii.
On Biot, see Frankel, J. B. Biot and the mathematization of experimental physics in
Napoleonic France (ref. 86).
Biot, Trait (ref. 135), p. xiv.
Ibid., p. xxiii.
Ibid., p. xv.
Harman (ed.), op. cit. (ref. 90), i, 671.
Ibid., 671.
Ibid., 672.
Fish suggests that, for Whewell, the formal system employed should fully cash out empirically,
op. cit. (ref. 89), 45.
University College London, Fleming Coll. Ms Add 122/37. My thanks to Sungook Hong for giving
me access to his notes taken from this manuscript.
P. M. Harman, Metaphysics and natural philosophy: The problem of substance in classical
physics (Brighton, 1982), 1456.
Cited by Harman, ibid., 146.
Cited in ibid., 132. On the meaning of the use of Lagrangian formulations, see Martin J. Klein,
Mechanical explanations at the end of the nineteenth century, Centaurus, xvii (1972),
5882; Mario Bunge, Lagrangian formulation and mechanical explanation, American journal
of physics, xxv (1957), 21118.
E. Cassirer, Substance and function (New York, 1953), provides a philosophical analysis
of this process.
A. Comte, Cours de philosophie positive (Paris, 1968), ii, 340.
416
YVES GINGRAS
149. H. A. Lorentz, Considerations on gravity, Amsterdam Koninklijke Akaemie Physica, ii, (1900),
55974. On Le Sage, see Samuel Aronson, The gravitational theory of Georges-Louis Le
Sage, The natural philosopher, iii (1964), 5374.
150. For the period from 1700 to 1900, very few proposed explanations of gravity were published in
journals controlled by the physics community. I will later publish a more detailed history of
the various attempts at a mechanical explanation of gravity. For a brief survey, see F. H. van
Lunteren, Gravitation and nineteenth-century physical worldviews, in P. B. Scheurer and G.
Debrock (eds), Newtons scientic and philosophical legacy (Dordrecht, 1988), 16173. Outside
the eld of physics, attempts at providing a mechanical explanation still goes on; for relatively
recent examples, see Dewey B. Larson, Beyond Newton: An explanation of gravity (Portland,
1964); Ren de Puymorin, Lorigine de la gravitation (Paris, 1975).
151. H. A. Lorentz, The theory of the electron (New York, 1953), 43; cited by M. Jammer, Concepts of
mass in contemporary physics and philosophy (Princeton, 2000), 36.
152. For the conceptual evolution of the concept of mass, see Max Jammer, Concepts of mass in
classical and modern physics (Cambridge, Mass., 1961) and his more recent Concepts of mass
in contemporary physics and philosophy (ref. 151).
153. For a survey of these developments, see G. N. Cantor and M. J. S. Hodge (eds), Conceptions of
ether: Studies in the history of ether theories, 17401900 (Cambridge, 1981).
154. E. P. Wigner, The unreasonable effectiveness of mathematics in the natural sciences,
Communications on pure and applied mathematics, xiii (1960), 114. For more examples
of such preoccupations by scientists and for a philosophical approach to the question, see
Mark Steiner, The application of mathematics as a philosophical problem (Cambridge, Mass.,
1998). Husserl also provided a philosophical analysis of the meaning of the mathematization of
nature in Edmund Husserl, The crisis of European sciences and transcendental phenomenology
(Evanston, 1970), 2159.
155. Arthur I. Miller, Redefining anschaulichkeit, in Abner Shimony and Herman Feshbach
(eds), Physics and natural philosophy (Cambridge, Mass., 1982), 376411; Daniel Serwer,
Unmechanischer zwang: Pauli, Heisenberg, and the rejection of the mechanical atom,
19231925, Historical studies in the physical sciences, viii (1977), 189256.
156. David Bohm and F. David Peat, Science, order, and creativity (New York, 1987), 7, 9.
157. Bohms view are now undergoing a revival; see Peter R. Holland, The quantum theory of motion
(Cambridge, 1993), and Russell Olwell, Physical isolation and marginalization in physics:
David Bohms cold war exile, Isis, xc (1999), 73856.
158. For a recent critique of the lack of physical explanations in the modern mathematical approach
to physics, see Daniel Athearn, Scientic nihilism: On the loss and recovery of physical
explanation (Albany, 1994).
159. For very recent examples, see Nature, cccciv, issue of 2 March 2000, 2829; Science, cclxxxvii,
issue of 7 January 2000, 4950.
160. See, for example, Karen Hunger Parshall, Chemistry through invariant theory? James Joseph
Sylvesters mathematization of the atomic theory, in Paul H. Therman and Karen Hunger
Parshall (eds), Experiencing nature (Dordrecht, 1997), 81111; Ana Simoes and Kostas
Gavroglu, Quantum chemistry qua applied mathematics: The contributions of Charles Alfred
Coulson (19101974), Historical studies in the physical and biological sciences, xxix
(1999), 363406, and idem, Quantum chemistry in Great Britain: Developing a mathematical
framework for quantum chemistry, Studies in history and philosophy of modern physics,
xxxi (2000), 51148; Giorgio Israel, The emergence of biomathematics and the case of
population dynamics: A revival of mechanical reductionism and darwinism, Science in
context, vi (1993), 469509.
161. Yves Gingras, La substance vanescente de la physique (ref. 16).