Descartes On The Unification of Arithmetic, Algebra and Geometry Via The Theory of

Revista Portuguesa de Filosofia
Descartes on the Unification of Arithmetic, Algebra and Geometry Via the Theory of
Proportions
Author(s): DAVIDE CRIPPA
Source: Revista Portuguesa de Filosofia, T. 73, Fasc. 3/4, Ciências Formais e Filosofia: Lógica
e Matemática / Formal Sciences and Philosophy: Logic and Mathematics (2017), pp. 1239-
1258
Published by: Revista Portuguesa de Filosofia
Stable URL: http://www.jstor.org/stable/26291336
Accessed: 31-05-2018 16:21 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/26291336?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Revista Portuguesa de Filosofia is collaborating with JSTOR to digitize, preserve and extend
access to Revista Portuguesa de Filosofia
This content downloaded from 163.178.101.95 on Thu, 31 May 2018 16:21:01 UTC
All use subject to http://about.jstor.org/terms
Revista Portuguesa de Filosofia, 2017, Vol. 73 (3-4): 1239-1258.
© 2017 by Revista Portuguesa de Filosofia. All rights reserved.
DOI 10.17990/RPF/2017_73_3_1239
Descartes on the Unification of Arithmetic, Algebra and

Geometry Via the Theory of Proportions *
DAVIDE CRIPPA **
Abstract
In this paper, we explore the role of the theory of proportions in the constitution of Cartesian
geometry. Particularly, we intend to show that Descartes used it in an essential way to
achieve a unification between geometry and arithmetic. Such a unification occurred mainly
by redefining the operation of multiplication in order to include both operations among
segments and among numbers. Finally, we question about the significance of Descartes’
algebraic thought. Although the goal of Descartes’ Géométrie is to solve geometric problems,
his first readers emphasized the role of algebra as a study of relations.
Keywords: algebra, Descartes, Euclid, geometry, multiplication, proportion theory, structure.
1. Introduction
T
he relationship between algebra and geometry in Descartes’ math-
ematics has often been the source of dilemmas for scholars:1 on
one hand Descartes attempted to bring geometry and algebra to
unity by providing a method to represent curves (geometrical objects) via
equations (algebraic ones), on the other he clearly maintained the logical
and epistemological priority of geometry over algebra, as it shines through
his practice in solving problems (construction of equations) or through
his argument to justify the exactness of curves and their acceptability in
geometry.
However, I think that this idea of a tension between geometry and
algebra in Descartes is the fruit of a misconception, largely due to an
* This paper was originally presented for a workshop on the history of the theory of proportions at
the APMP meeting (Association for the Philosophy of Mathematical Practice) held in Salvador de
Bahia, from 23rd to 27th October 2017. I would like to thank especially Abel Lassalle Casanave,
Eduardo Giovannini and Jesper Lützen for their insightful comments and suggestions.
** The Czech Academy of Sciences, Institute of philosophy.
 davide.crippa@gmail.com
1. For instance, Cathay Liu, ‘Re-Examining Descartes’ Algebra and Geometry: An
Account Based on the Regulae’, Analytic philosophy, 50-1 (2017): 29-57. DOI: 10.1111/
phib.12093, 29.
1239-1258
1240 Davide Crippa
anachronistic understanding of the meaning and role of algebra in the

Géométrie (1637), Descartes’ most accomplished mathematical work. As I
would like to show in this article, the essential contribution put forward
in the Géométrie consists of a “geometrical calculus” i.e. an algebra which
deals with segments and whose range of applicability includes any kind of
homogeneous quantities, magnitudes, numbers, times, velocities, which
can be treated by Euclid’s theory of proportions. Algebra was the key to
unify the domains of geometry and arithmetic, precisely because of its
structure: it is constituted by Descartes by specifying a set of fundamental
geometrical operations between segments analogous to the five funda-
mental operations of arithmetic.
While this achievement can be read as a stepping stone into modern
science, because it opens the possibility of a mathematical physics, it can
be also understood in the background of ancient mathematics, since it
responds to a question which loomed large in Greek mathematics: how
can geometry, or the science of continuous quantity, and arithmetic, or the
science of number, be reconciled? In what follows we shall discuss how
Descartes articulated a convincing answer to this question.
2. Ancient attempts to unify arithmetic and geometry
2.1 Euclid’s theory of proportions
The question about the unity between arithmetic and geometry,

central to Descartes’ mathematical thinking from his youth to the publi-
cation of the Géométrie in 1637,2 had a long tradition going back to Greek
mathematics and one of its most sophisticated achievements: the theory
of proportions of Eudoxus, that we know mainly through its elaboration
in Euclid’s book V of the Elements.
One of the original goals of proportion theory was to overcome the
impossibility, which stems from the discovery of incommensurability, of
expressing through numbers (i.e. multiplicities of units) certain lengths,
such as the diagonal of a square whose side has length 1. The solution
proposed by Euclid (and presumably going back to Eudoxus) consisted in
2. The problem of the relations between arithmetic and geometry is raised by Descartes
for the first time in a well-known letter to Beeckman from 1619.
Vol. 73
Fasc. 3-4 RPF 2017
Descartes on the Unification of Arithmetic, Algebra and Geometry Via the Theory of Proportions 1241
defining a notion of ratio sufficiently general to hold for any pair of homo-
geneous magnitudes.
The definition of ratio given in the Elements is interesting but math-
ematically useless:
A ratio is a sort of relation in respect of size between two magnitudes of

the same kind.3
This definition tells us that: i) ratio is a relation; ii) this relation can
occur only between homogeneous quantities, such as two segments, two
areas, two volumes, but cannot occur between a segment and an area,
for instance, or any two objects which have different dimensions. The
following definition 5 is fundamental. Euclid specifies in it what it means
“to be in the same ratio”:
Magnitudes are said to be in the same ratio, the first to the second and
the third to the fourth, when, if any equimultiples whatever be taken of
the first and third, and any equimultiples whatever of the second and
fourth, the former equimultiples alike exceed, are alike equal to, or alike
fall short of, the latter equimultiples respectively taken in corresponding
order.
This is an important move towards abstraction since it establishes,

without presupposing any definition of ratio (as a matter of fact we could,
under the assumption of having pairs of homogeneous magnitudes,
simply forget the previous definition of ratio), what it means for two pairs
of magnitudes, either commensurable or incommensurable, to be in the
same ratio.
Perhaps because of this abstract treatment of ratios, Euclid’s theory
of proportions was identified since antiquity as a theory which included
both magnitudes and numbers, namely quantities in general, as its subject
matter.4
3. El. V, df. 4. Unless otherwise specified, all references to the text of Euclid’s Elements
are from: Euclid, The Thirteen Books of Euclid’s Elements Translated from the Text of
Heiberg With introduction and Commentary, trans. Thomas L. Heath (Cambridge:
Cambridge University Press, 1908. New York: Dover Publications, 1956).
4. One of the most interesting but also controversial suggestions is made by Aristotle,
in a passage which has raised several problems to its interpreters (Aristotle, Posterior
Analytics, trans. Johnathan Barnes (New York: Oxford University Press, 1994), 8).
This view is also supported by a scholium to the Elements quoted in David Rabouin,
Mathesis universalis. L’idée de “mathématique universelle” d’Aristote à Descartes (Paris:
Vol. 73
Fasc. 3-4 RPF 2017
1242 Davide Crippa
However, there are sound objections against this interpretation of

Euclid’s theory as a “metatheory” dealing with such an abstract notion of
quantity. Firstly, if Euclid’s proportion theory were about quantities, either
magnitudes or numbers, certain choices made in the Elements appear
utterly redundant. A peculiar feature of Euclid’s treatment of proportions
is that certain concepts such as that of “proportionality” are defined twice,
once for the case of geometry and once for arithmetic.
Likewise, certain propositions are proven twice, by two distinct
proof, such as the following pair:
If four magnitudes are proportional, then they are also proportional

alternately.5
If four numbers are proportional, then they are also proportional alter-
nately.6
The practice of reproving a theorem betrays evident kinship with the

Aristotelian epistemological principle, according to which sciences are
to be distinguished according to their kinds, and proofs cannot be trans-
ferred from one kind to another:
We cannot in demonstrating pass from one genus to another. We cannot,

for instance, prove geometrical truths by arithmetic (…) where the kinds
are different, as with arithmetic and geometry, you cannot attach arith-
metical demonstrations to what is incidental to magnitudes – unless
magnitudes are numbers.7
Therefore, if we assume that the theory of proportions expounded

in Book V of the Elements belongs to geometry, it would be illegitimate to
transfer the proof obtained for proportions among magnitudes to propor-
tions among numbers. The same proposition should be reproved, as in the
cases of El., V 16 and VII, 13.
The idea that the Aristotelian epistemology may have influenced the
structure of the Elements poses further problems, notably the presence of
propositions which overtly involve numbers and magnitudes, as in Book
X:
PUF, 2009), 93.

5. El., V, 16.
6. El. VII, 13.
7. An. Post., I, 7, 75a 38.
Vol. 73
Fasc. 3-4 RPF 2017
Commensurable magnitudes have to one another the ratio which a

number has to a number.8
If two magnitudes have to one another the ratio which a number has to
a number, the magnitudes will be commensurable.9
These examples show well how Euclid’s criterion to decide whether

two magnitudes are commensurable involves the comparison between
magnitudes and numbers, an apparent violation of Aristotle’s standpoint.
How to account for this apparent inconsistency in the Elements?
Let us recall that from Aristotle’s point of view the possibility of
crossing kind is to be conceded when a science is subordinate to another,
as if with optics and geometry, or harmony and arithmetic. We could
always assume that arithmetic and geometry stand in a relation of subor-
dination, and either that geometry is prior to arithmetic, and includes the
latter as its subdiscipline, or that arithmetic is prior to geometry.
This hypothesis, even if it has been advanced during history, is not
prima facie obvious on the basis of Euclid’s text. In fact, there are few
significant counterexamples, propositions that hold for proportions about
numbers but not about magnitudes, or viceversa. Consider, for instance, El.
VII, 19 which states a property true in the framework of Euclid’s Elements,
only for numbers:
If four numbers be proportional, the number produced from the first

and fourth will be equal to the number produced from the second and
third; and, if the number produced from the first and fourth be equal
to that produced from the second and third, the four numbers will be
proportional.10
We can find a similar proposition concerning magnitudes in proposition

VI, 16:
If four straight lines are proportional, then the rectangle contained by

the extremes equals the rectangle contained by the means; and, if the
rectangle contained by the extremes equals the rectangle contained by
the means, then the four straight lines are proportional.11
8. El., X, 5.
9. El., X, 6.
10. El., VII. 19.
11. El., VI, 16.
Vol. 73
Fasc. 3-4 RPF 2017
1244 Davide Crippa
The similarity is made more precise by the fact that the product of
two segments can be interpreted as the rectangle having such segments for
sides (El., II, df.1). But this analogy holds only if the proportional terms
are segments, whereas it does not hold for arbitrary magnitudes. In other
words, there is no way to define an operation of multiplication between
two surfaces or volumes within the framework of the Elements, since the
geometrical equivalent of the operation of multiplication behaves as an
external law of composition.
Conversely, we can also find propositions on proportions which
hold for geometric magnitudes but not for numbers. Since a “number”
is, for Euclid, a multiplicity of unities, we cannot take an arbitrary part
of it, while we certainly can take arbitrary submultiples of a magnitude.
It results that any proposition whose proof depends on the operation of
taking an arbitrary submultiple would hold for magnitudes but not for
numbers (for instance, El. V, 5).
This couple of examples is sufficient to show that Euclid did not
possess a unified concept of “magnitude”, but that geometrical magnitudes
(segments, surfaces, volumes and angles) and numbers were conceived as
distinct objects, and distinct and non-communicating were the sciences
which investigated them. At the same time, this reading leaves open the
question about the epistemological correctness of Book X, which conflates
considerations about numbers and magnitudes such as in propositions 5
and 6.
Against scholars such as Mueller explicitly considered the lack of a
unified treatment of proportions as a “failure” of Euclid’s theory of propor-
tions,12 there is room to argue that Euclid’s theory of proportions can be
considered a general theory which encompassed both geometrical magni-
tudes and numbers, and at the same time does not violate the Aristotelian
principle against crossing kinds.
One argument to save the generality of Euclid’s proportion theory has
been advanced by Rabouin.13 It can be thus summarized: Euclid’s theory of
12. According to Mueller, in fact: “Euclid’s failure to establish a correlation between his
two treatments of proportionality before developing the material in book X is probably
the greatest foundational flaw in the Elements.”, In Ian Mueller, Philosophy of math-
ematics and deductive structure in Euclid’s Elements (New York: Dover Publications,
2006), 139.
13. David Rabouin, ‘The problem of a “general” theory in mathematics: Aristotle and
Euclid’, in The handbook of generality in mathematics and the sciences, eds. Karine
Chemla, Renaud Chorlay, David Rabouin (Oxford: Oxford University Press, 2016),
113-135.
Vol. 73
Fasc. 3-4 RPF 2017
proportions is a general theory thanks to a logic of analogy, which allows

one to circumvent the prohibition of transferring proofs from one kind
to the other while maintaining the Aristotelian distinction into kinds. In
other words, in virtue of analogy the proof that certain properties hold for
a certain subject matter, i.e. geometry, can extend its validity and hold also
within a different subject matter, for instance arithmetic.
Analogy has a technical meaning in Aristotle’s theory of science,14 and
is aptly illustrated by examples deriving from geometry (An. Post. 76a):
Of the items used in the demonstrative sciences, some are proper to each
science and others common-but common by analogy (…) that if equals
are removed from equals, the remainders are equal.15
The example cited by Aristotle is about one of Euclid’s “common

notions”: while it is valid for geometric magnitudes, it is also valid for
arithmetical ones. To take another simple example of general statements
insofar as they are valid across kinds, the number 2 and the segment AB
belong to different kinds, however they can entertain the same relations
with, respectively, the number 4 and a segment CD, double of AB.
The theorems about proportions proven in Book V, although they
predicate properties valid for magnitudes, can hold for numbers by
the same logic of analogy, since Euclid’s proportion theory deals with
properties and operations of magnitudes which precede the distinction
commensurable/incommensurable.16 In fact, the properties treated in Book
V regard the relations between multitudes and parts of given magnitudes
or, as Vitrac points out in his commentary to the Elements: “reasoning on
multitudes, which constitutes a proper domain preceding the distinction
arithmetic/geometry, concerning certain combinatorial properties of
finite sets”.17 This “combinatorial” mathematics, as rudimentary as it may
appear, enabled Euclid to reach outstanding results and, with respect to
the theme of our talk, results that are valid by analogy for both magnitudes
and numbers. For instance, Euclid managed to provide: (ii) an algorithm
to calculate the greatest common measure, which holds for numbers but
14. Mary Hesse, ‘Aristotle’s logic of analogy’, The Philosophical Quarterly 15-61 (1965):
328-340. DOI: 10.2307/2218258, 331.
15. Aristotle, Posterior Analytics, 15.
16. Rabouin, Mathesis universalis, 92-93.
17. Euclid, Euclide Les Eléments. Vol. 2. Trans. by Bernard Vitrac (Collection Bibliothèque
d’histoire des sciences. Paris: P.U.F, 1994), 127.
Vol. 73
Fasc. 3-4 RPF 2017
1246 Davide Crippa
can also extend its domain of validity to mutually commensurable magni-

tudes and (iii) a criterion to decide whether two magnitudes are mutually
commensurable or incommensurable (Elements, X, 5, 6) and considering
these examples, we can say that the unification between geometry and
arithmetic is achieved within Euclid’s proportion theory on a level that
we can call “analogical”: the same set of fundamental operations hold for
objects belonging to different kinds, for instance of geometrical magni-
tudes commensurable one to another and of numbers, without implying
the existence of a science dealing with a common kind such as a general
and abstract notion of magnitude.18
2.2 Geometric algebra
A different approach to the unification of geometry and arithmetic

consists in looking for an analogy between arithmetical operations on
numbers, namely addition, product and square root extractions, and
geometric constructions using straight lines and circle. This approach
came to the fore especially with the emergence of the algebraic thinking
among Arab and Renaissance mathematicians, and can be carried out
without necessarily employing the theory of proportions. An example of
geometrical construction modelling arithmetical operations is the so-called
“application of areas” illustrated Book II and VI of the Elements. Although
these books are devoted to proving theorems and solving problems in
plane geometry, these results have received an arithmetic or an algebraic
interpretation.19 For Arabic and Renaissance algebraists, Euclid’s Elements
had also a foundational role because, using theorems from plane and solid
Euclidean geometry (to be found in Euclid or provable from the Elements)
mathematicians were able to offer a geometric justification for algorithms
to solve linear, quadratic and cubic equations.20
18. Rabouin, The problem of a general theory, 128-129.

19. Bartel Leendert van der Waerden, ‘Defence of a “shocking” point of view’. Archive for
the history of exact sciences, 15-3 (1976): 199-210.https://doi.org/10.1007/BF00412256;
Leo Corry, ‘Geometry and arithmetic in the medieval traditions of Euclid’s Elements:
a view from Book II’, Archive for the history of exact sciences, 67 (2013): 637-705. DOI
https://doi.org/10.1007/s00407-013-0121-5; Marco Panza, ‘What more there is in ear-
ly modern algebra than its literal Formalism’, in Philosophical Aspects of Symbolic
Reasoning in Early Modern Mathematics, eds. Albrecht Heeffer, Marteen Van Dyck
(London: College Publications, 2010), 193-229.
20. Paolo Freguglia, La geometria tra tradizione e innovazione (Torino: Bollati
Boringhieri, 1999).
Vol. 73
Fasc. 3-4 RPF 2017
The role of the second book of the Elements as a foundation of algebra

was explicitly discussed by Clavius in his commentary to the Elements,
a text also known by Descartes since his youth. Clavius noted that the
language of Euclidean geometry was more general than that of arithmetic,
and therefore more apt to justify the generality of algebraic procedures. In
fact it could be used to express the result of operations among surd quan-
tities (irrational) in a way that avoids the complications of dealing with
quantities expressible only through infinite approximations.21
However, an algebra of geometric magnitudes as the one just
described was limited in its applications by the structure of geometrical
space, and by dimensionality and homogeneity. The impossibility of
reasoning with geometrical objects in more than three dimensions was
perceived by ancient geometers as hindering the possibility of finding
general solutions to problems. Let us consider, for instance, the original
account of the so-called “problem of Pappus” which Descartes would use
as a test-case to prove the strength and novelty of his own geometry. The
problem asks to determine a locus (a curve) in the plane, given certain
requirements all its points had to satisfy, or to prove that any point satis-
fying such and such requirements lie on a certain locus. Greek geometers,
Pappus acknowledged, have only solved a limited number of instances of
this problem, like the one above, but entered unsurpassable difficulties as
soon as the number of given line in the problem grows:
Now if (they are drawn) onto only two (lines), the locus has been proved
to be plane, but if onto more than four, the point will touch loci that are
as yet unknown, but just called ‘curves’, and whose origins and prop-
erties are not yet (known). They have given a synthesis of not one, not
even the first and seemingly the most obvious of them, or shown it to be
useful … If onto more than six, one can no longer say “the ratio is given
of the something contained by four to that by the rest”, since there is
nothing contained by more than three dimensions.22
21. “Nam ex nonnullis harum propositionum demonstrantur regulae illae admirabiles

Algebrae, quibus vix credo in disciplinis humanis praestantius aliquid reperiri, quip-
pe cum miracula quadam numerorum (ut ita dicam) eruant tam abstrusa, ac recon-
dita, ut facultas illa omnem captum humanum superare videatur … ex aliis deinde
propositionibus huius lib. eliciuntur demonstrationes, quibus inter se adduntur, sub-
strahuntur, multiplicantur, atque dividuntur numeri surdi, (quos dicunt) hoc est, qui
nullo modo exprimi possunt”, Christophorus Clavius, Euclidis Elementorum Libri XV
(Roma: apud Bartholomaeum Grassium, 1589), 253.
22. Pappus, The book 7th of the Collection, trans. Alexander Jones (Dordrecht Heidelberg
New York London: Springer, 1986), 123.
Vol. 73
Fasc. 3-4 RPF 2017
1248 Davide Crippa
Pappus proposed a way to circumvent the problem using the notion of

“compound ratio”, but did not develop this idea thoroughly and generally.
In taking over the challenge posed by Pappus’ problem, Descartes made
in the Géométrie a radically new use of algebra as a bridge between arith-
metic and geometry.
3. Descartes’ geometrical calculus as an attempt to unify

arithmetic and geometry
Descartes pursued the task of unifying arithmetic and geometry by

reforming geometric algebra through the redefinition of certain funda-
mental operations as internal laws of composition. In this way, he overcame
the traditional constraints of geometric algebra, namely homogeneity and
dimensionality. As we shall see, the role of Euclid’s theory of proportions
was fundamental to attain this goal.
The opening paragraph of the Géométrie is introduced by the inter-
esting subtitle: “how the calculations of arithmetic are related to the
operations of geometry”.23 In the following lines, Descartes grounded a
veritable “geometrical calculus”, namely an algebra of segments or an
“algebra speciosa”, in the words of his early commentator Florimond de
Beaune: “science to investigate and discover [the proof or solution of]
problems and theorems”.24 This science, De Beaune insists, includes the
“geometrical analysis of the ancients” (“veterum analysis geometrica”) as
well as the numerical algebra and extends its application to any problems
reducible to proportions. De Beaune’s words here are clearly reminiscent
of Descartes’ criticism of both the ancient and modern analysis sketched
in the Discourse.
To show how arithmetical calculus and geometrical operations relate,
let us start from the case of addition and subtraction. Although these oper-
ations are never properly defined in Euclid’s geometry, Descartes could
interpret them intuitively, as the placing a segment next to a given one or
the cutting off a segment from a given one (Elements, 1.2, 1.3).
23. René Descartes, The Geometry of René Descartes, Trans. David Smith and Marcia
Latham (New York: Dover, 1952), vi.
24. “Scientiam investigandis inveniendisque Theorematis et Problematis inserviens”.
René Descartes, Renati Descartes Geometria. Editio Secunda. Multis accessionibus ex-
ornata, et plus altera sua parte adaucta, trans. Frans Van Schooten (Amsterdam: Apud
Ludovicum et Danielem Elzevirios, 1659-1661), 107.
Vol. 73
Fasc. 3-4 RPF 2017
On the other hand, Euclid’s theory of proportions enters crucially

into the definitions of product, quotient and square root of segments. Let
us consider specifically the case of multiplication, that Descartes defines
as follows:
… taking a line which I shall call unity to relate it as closely as possible

to numbers, and which can in general be chosen arbitrarily, and having
given two other lines to find a fourth line which shall be to one of the
given lines as the other is to unity.25
Fig. 2. In the Géométrie, the product of two segments is a segment, Descartes, 1952, 298.
We immediately remark that the notion of product between segments

here introduced differs from the classical interpretation in term of the
rectangle construction seen above. According to Descartes, the product of
segments BD x BC (fig. 2) is the segment BE which satisfies, in virtue of the
construction shown in fig. 2, the following proportion: AB:BD ∷ BC:BE. In
this proportion, the segment AB takes the role of a unit segment (“l’unité”).
Defined as the search for the fourth proportional, multiplication becomes
an internal operation which, thanks to the introduction of a geometrical
unit, does not violate homogeneity: the product BD x BC will be in fact a
segment, and not a rectangle as it was customary in the earlier geometric
algebra. In a similar way, division and square-root extractions are defined
in terms of the search for a fourth proportional (division) or a mean
25. Descartes, Geometry, 2. The passage in the original French goes as follows: “Ou bien
en ayant une, que je nommeray l’unité pour la rapporter d’autant mieux aux nombres,
et qui peut ordinairement estre prise a discretion, puis en ayant encore deux autres, en
trouver une quatriesme, qui soit a l’une de ces deux, comme l’autre est a l’unité, ce qui
est le mesme que la Multiplication” (ibid., 3).
Vol. 73
Fasc. 3-4 RPF 2017
1250 Davide Crippa
proportional (square root) given three segments. Moreover as Descartes

explains:
It is not, however, the same thing when unity is determined, because

unity can always be understood, even when there are too many or too
few dimensions; thus, if it be required to extract the cube root of a2b2 – b,
we must conside the quantity a2b2 divided once by unity, and the quantity
b multiplied twice by unity.26
The definition of multiplication reveals that the role of a unit segment

is crucial for the constitution of a geometrical calculus analogous to arith-
metic. This idea is stressed by Frans Van Schooten in the commentary to
the Géométrie:
… by unity, understand a determinate line, which has to any of the

remaining lines the same relation that the unit has to a certain number.27
But in which sense the segment chosen as a unit bears the same
relation to other segments as the number 1 to other numbers? It seems in
fact that the role of the geometrical unity in the Géométrie is not that of
expressing the measure of a certain segment.
To clarify this point, it is useful to analyze the definition of multipli-
cation in terms of the construction of a fourth proportional, which is the
one finally adopted by Descartes. Let us recall that, in the arithmetical
books of the Elements, multiplication is the only operation defined among
numbers, more precisely:
A number is said to multiply a number when that which is multiplied

is added to itself as many times as there are units in the other, and thus
some number is produced (VII, df. 16)
This definition is framed on the Euclidean notion of number

as “a multitude composed of units” (df. 2). Hence, to multiply the
26. Ibid., 6. The original follows here : “mais que ce n’est pas de mesme lorsque l’unité est
déterminée, à cause qu’elle peut estre sousentendue par tout où il y a trop ou trop peu
de dimensions: comme s’il faut tirer la racine cubique de , il faut penser que la quan-
tité aabb est divisée une fois par l’unité, et que l’autre quantité b est multipliée deux fois
par la mesme.” (ibid., 7).
27. “Per unitatem intellige lineam quandam determinatam, quae ad quamvis reliqua-
rum linearum talem relationem habeat, quales unitas ad certum aliquem numerum”
(Descartes, Geometria, 147).
Vol. 73
Fasc. 3-4 RPF 2017
number m by the number n means to add n multitude composed of m

units: m1 + m2 + ... mn . On the other hand, multiplication can be related to
the theory of proportions via propostion 19 from El. VII:
If four numbers be proportional, the number produced from the first

and fourth will be equal to the number produced from the second and
third; and, if the number produced from the first and fourth be equal
to that produced from the second and third, the four numbers will be
proportional.
The equality of two products becomes a criterion to decide whether four

numbers are in a proportion.
If we admit 1 as a number, the above theorem can easily turn into a
definition of product. We shall have in fact that:
a x b =df the number x, such that 1:a ∷ b:x.
This definition raises problems in the context of the Elements for at

least two reasons. Firstly, it presupposes the existence of a fourth propor-
tional. This existential claim is never made explicit by Euclid, and anyway
does not generally hold in arithmetic. Secondly, it presupposes that 1 is a
number, which is not the case according to Euclid’s definition.
However, this definition of arithmetical product in terms of propor-
tions is to be found in several treatises from the Renaissance and early-
modern period (cf. fig. 3). A possible explanation may be that in the context
of practical computations, in the middle ages and the Renaissance, it
became clear that the arithmetical unity was part of the current systems of
computations and was therefore considered to behave just like the other
(whole) numbers. It is not surprising, therefore, to read in a treatise on
methods of computations from the beginning of the XVIth century the
following definition of multiplication (fig.3):
Multiplication is the creation of a number, which is proportionally to the

multiplicand as the multiplicans is to the unit.28
28. ‘Multiplicatio est numeri procreatio proportionaliter se habentis ad multiplicandum

sicut multiplicans ad unitatem se habet’, Johannes Huswirt, Enchiridion Algorismi
(Koln: In officina felicis memorie honesti viri Henrici Quentell, Smith 1501). See
also Johnathan Crabtree, ‘A new model of multiplication via Euclid’, Vinculum, 53, 2
(2016): 16-21.
Vol. 73
Fasc. 3-4 RPF 2017
1252 Davide Crippa
Fig. 3: “Multiplication is the creation of a number, which is proportionally to the multiplicand as

the multiplicans is to the unit”.
On the ground of similar examples, which abounded in the arith-

metical tradition, Descartes might have taken the very property expressed
in El. VII, 19 as a definition of the product between numbers, and applied
the same proportional conception of multiplication to geometry (this is how
we may call it to distinguish it from multiplication conceived as repeated
addition). To have the same definition for arithmetic and geometry, a
segment having the same role of the arithmetical unit had to be intro-
duced. Echoes of the proportional conception of arithmetical operations
can be found in Van Schooten’s commentary.29
In this way, a unique definition of multiplication can hold for both
numbers and magnitudes, specifically segments. In the latter case, the
operation of multiplication also acquires a definite constructive meaning,
because the fourth proportional from three given segments can be
constructed by ruler and compass on the basis of El., VI, 12. In a similar
way, Descartes was also able to define other operations of elementary
arithmetic (table below).
Interpretation in
Arithmetic Geometrical
terms of proportions interpretation
Algorithmic construction
operations
(between numbers)
1:a ∷ b:x (“e” El., VI, 12 “to three given
axb=x 1:a ∷ b:x is called unity, straight lines to find a
chosen at will) fourth proportional”.
a
⁄b x:1 ∷ a:b x:e ∷ a:b The same.
El., VI, 13 (“to two given

√a a:x ∷ x:1 a:x ∷ x:e straight lines to find a
mean proportional”)
Table 1
29. Descartes Geometria, vol. 1, 147-149.
Vol. 73
Fasc. 3-4 RPF 2017
The table above can be extended by the operations of taking the n-th
power of a number or a segment (which can be conceived of as a repeated
multiplication) or extracting n-th root of a number or a segment. The
extraction of higher-order roots corresponds to the problem(s) of inserting
2 or more mean proportionals, which are expressible via the theory of
proportions but not constructible within Euclidean geometry, as Descartes
stated and attempted to prove.30 Therefore, to extend his geometrical
calculus and be able to solve algebraic equations in any degree, Descartes
tackled in the second Book of the Géométrie the methodological problem
of extending the constructions admissible in geometry beyond the limits
of Euclid’s postulates, and included into geometry all curves constructible
by one, continuous motion.31
Eventually, Descartes attained a unity between arithmetic and
geometry in a way that recalls how generality is attained in book V of
the Elements, at least according to the explanation given in the previous
section. In both cases, an essential use of analogy is made. On one hand,
Euclid’s proportion theory is general because it holds for geometrical oper-
ations which presuppose the distinction between commensurability and
incommensurability, and for numbers. On the other, Descartes’ geomet-
rical calculus is general because its fundamental operations are defined
in the same way both for segments and numbers alike. There is no need
to presuppose that numbers are used in Cartesian geometry, for instance
to measure lengths, hence there is no violation of purity. Arithmetic and
geometry are dealt with as separate disciplines, as for their respective
objects, but they are unified on the level of operations.
One may be tempted to go a step further and argue that, once the
operation of multiplication is defined as Descartes does, segment arith-
metic and ordinary arithmetic of numbers are endowed with the same
structure, namely that of a field. The unity-segment of Descartes’ calculus
can be thus conceived as a neutral element with respect to multiplication:
multiplying or dividing any segment by that segment always yields the
original segment as a result. On this basis, scholars such as Mahoney 32
30. Jesper Lützen, ‘The Algebra of Geometric Impossibility: Descartes and Montucla
on the Impossibility of the Duplication of the Cube and the Trisection of the Angle’,
Centaurus, 52-1 (2010): 4-37. DOI: 10.1111/j.1600-0498.2009.00160.x.
31. Henk Bos, Redefining geometrical exactness (Dordrecht Heidelberg New York
London: Springer, 2001); Marco Panza, ‘Rethinking geometrical exactness’. Historia
mathematica,38-1 (2011): 42-95. https://doi.org/10.1016/j.hm.2010.09.001.
32. Michael Mahoney, ‘The beginnings of algebraic thought in the seventeenth century’,
Vol. 73
Fasc. 3-4 RPF 2017
1254 Davide Crippa
suggest, albeit cautiously, that Cartesian geometry may contain an outline

of a field theoretic approach to geometry. This would account for a deep
unity between geometry and arithmetic on the level of their mathematical
structure, while magnitudes and numbers would continue to be treated as
distinct objects.33
It is not easy to ascertain whether Descartes and his early-modern
readers had this view in mind, even if implicitly. As we have argued above,
they were aware that the unity-segment entertains the same relations with
other segments just as the number 1 has with the other whole numbers.
According to a more cautious interpretation, these relations are those
involved in the definitions of operations such as multiplication, division
and root extraction.
Did Descartes have a deeper understanding of the unity between
geometry and arithmetic on the level of algebraic structures? Before
venturing an answer, it is important to stress an important difference,
which arises we inquire about the similarities and differences between the
unity as a neutral element in our modern field theoretic interpretations
of geometry and the unity in Cartesian geometry. In the former, exem-
plified for instance by the arithmetic of segments presented in Hilbert’s
Grundlagen der Geometrie, a fixed, standard unit segment is assumed.
As Hartshorne explains, upon commenting Hilbert’s achievement in the
creation of an arithmetic of segments, one must: “choose arbitrarily, and
then fix once for all, a segment class we call unit segment and denote it by
1”.34 In Cartesian geometry, on the contrary, it is never question of choosing
a fixed unity element once for all, because the unity varies according to the
problem one intends to solve. We can choose as unit a segment which can
offer the easiest solution to our problem, depending on a given config-
uration, but we could also not choose one, if the equation to which a
problem is reduced contains only homogeneous terms. In the latter case,
in fact, any choice of the unity segment would leave the relations between
known and unknown magnitudes unaltered, as well as the solution of the
original geometric problem, so that it is simply unnecessary. This leads us
to conclude that, in Cartesian geometry, a fundamental presupposition to
in, Descartes: Philosophy, Mathematics and Physics, ed. Stephen Gaukroger (Barnes &
Noble, 1980), 145.
33. This conclusion is not in Mahoney’s article, but it can be taken as a logical conse-
quence of his suggestion.
34. Robin Hartshorne, Geometry: Euclid and beyond (Dordrecht Heidelberg New York
London: Springer, 2000), 169.
Vol. 73
Fasc. 3-4 RPF 2017
construe an arithmetic of segments à la Hilbert is missing. On this basis,

it seems more plausible to think that the generality in Cartesian theory
is achieved by means of analogy, in a similar manner to Euclid’s theory
of proportions, rather than through the (tentative) construction of one
abstract structure (namely a field) modelling both domains of segments
and numbers.
4. The geometrical calculus as a model of algebraic thinking
The previous remarks lead us to inquire whether and to what extent

we can talk about algebraic thinking in Descartes. In his paper on The
beginnings of algebraic thought in the seventeenth century, Mahoney char-
acterized algebraic thinking by the following three features:
1. operational symbolism;
2. The preoccupation with mathematical relations rather than with
mathematical objects,
3. Freedom from any ontological questions and commitments and,
connected with this, abstractness rather than intuitiveness.
While the first aspect is an undeniable feature of Descartes’ algebra,

we may question the latter two. After all, there is good evidence to claim
that the geometrical calculus presented by Descartes in the Géométrie
is grounded in geometry. Firstly, it applies to geometrical problems,
secondly the algebraic symbolism is ontologically “committed” since it
refers to segments or numbers, thirdly equations themselves are compact
and simple ways to code proportions between segments, and algorithms
to construct them. In Descartes’ programme of the Géométrie, algebra is
an instrument to solve geometrical problems. The geometrical root of
Cartesian algebra is even clearer considering the differences between our
modern understanding about the study and solutions of equations and
the Cartesian understanding of it. For us, solving an algebraic equation
means to find a formula in order to express the unknown in terms of the
coefficients; in Descartes’ geometry it meant to construct a segment corre-
sponding to the unknown.35 Descartes’ algebra, considered as a part of his
35. Bos, Redefining exactness, 363-375.
Vol. 73
Fasc. 3-4 RPF 2017
1256 Davide Crippa
broad programme in geometry, appears to occupy a subordinate position

as a means to provide a systematic solution to problems in plane geometry.
The question of the relations between geometrical and algebraic
thinking has raised numerous discussions among scholars.36 It seems to
me that the mathematicians who firstly studied and diffused Cartesian
geometry, such as Van Schooten, started the process of “degeometrization
of analysis”, which became evident at least one century later, by shifting
the emphasis from the application of algebra to geometry to a study of
algebra qua an independent subject matter. It is worth noting, for example,
that Frans Van Schooten’s introductory lesson in Cartesian geometry or
Principia Matheseos Universalis (1651) edited and published by his student
Erasmus Bartholin, contains a formal exposition of the rules of Cartesian
algebra or “mathesis universalis”, i.e. independently from any reference
to a geometrical or arithmetical content. It is stressed in this work how
algebra was able to study, in an abstract and universal way, those rela-
tions, expressible via proportions, among quantities belonging to different
mathematical disciplines.37
These considerations indicate that a decisive step towards a view of
algebra as a science of relations (Mahoney’s requirement i) independent
from ontological commitments (Mahoney’s requirement ii) had been
taken, which would not be without consequence for the later history of
algebra.38
Two elements have an important function in this development.
The first is Euclid’s theory of proportions, which anchors the abstract
dimension of algebraic reasoning on a geometrical basis (equations code
proportions among line segments) and determines the scope and limits of
algebraic thinking; and a symbolic language, which endows algebra with
universality and abstractness.
36. Paolo Mancosu, Philosophy of mathematics and mathematical practice in the 17th
Century (Oxford: Oxford University Press), 1996.
37. “Unde cum in universa illarum scientiarum constitutione, licet diversa objecta respi-
ciant, non nisi relations sive propositiones quaedam, quae in iis reperiunturm con-
siderentur.”, Erasmus Bartholin, ‘Principia matheseos universalis’, in Renati Descartes
Geometria. Editio Secunda. Multis accessionibus exornata, et plus altera sua parte
adaucta, ed. Frans Van Schooten, vol. 2 (Amsterdam: Apud Ludovicum et Danielem
Elzevirios, 1659-1661), 1.
38. Rabouin, The problem of a general theory.
Vol. 73
Fasc. 3-4 RPF 2017
References
Aristotle. Posterior Analytics, translated by Johnathan Barnes, New York: Oxford University Press. 1994.
Bartholin, Erasmus. ‘Principia matheseos universalis’, in Renati Descartes Geometria. Editio Secunda. Multis
accessionibus exornata, et plus altera sua parte adaucta, edited Frans Van Schooten, vol. 2. Amsterdam:
Apud Ludovicum et Danielem Elzevirios, 1659-1661.
Bos, Henk. Redefining geometrical exactness, Dordrecht Heidelberg New York London: Springer, 2001.
Clavius, Christophorus. Euclidis Elementorum Libri XV, Roma: apud Bartholomaeum Grassium, 1589.
Corry, Leo. ‘Geometry and arithmetic in the medieval traditions of Euclid’s Elements: a view from Book II’,
Archive for the history of exact sciences, 67 (2013): 637-705. DOI: 10.1007/s00407-013-0121-5.
Crabtree, Johnathan. ‘A new model of multiplication via Euclid’, Vinculum, 53,2 (2016): 16-21.
Descartes, René. Renati Descartes Geometria. Editio Secunda. Multis accessionibus exornata, et plus altera sua
parte adaucta, edited and translated by Frans Van Schooten. Amsterdam: Apud Ludovicum et Danielem
Elzevirios, 1659-1661.
Descartes, René. Oeuvres de Descartes, 12 vols. Edited by Adam C., Tannery P. Paris: Cerf, 1897 – 1913.
Descartes, René. The Geometry of René Descartes. Translated by David Smith and Marcia Latham. New York:
Dover, 1952.
Euclid. Euclide Les Eléments. Vol. 2. Translated by Bernard Vitrac. Collection Bibliothèque d’his toire des
sciences. Paris: P.U.F, 1994.
Freguglia, Paolo. La geometria tra tradizione e innovazione. Torino: Bollati Boringhieri, 1999.
Hartshorne, Robin. Geometry: Euclid and beyond. Dordrecht Heidelberg New York London: Springer, 2000.
Heath, Thomas L. The Thirteen Books of Euclid’s Elements Translated from the Text of Heiberg With introduction
and Commentary. University Press, Cambridge, 1908. Reprint: New York, Diver Publications, 1956.
Hesse, Mary. ‘Aristotle’s logic of analogy’, The Philosophical Quarterly 15-61 (1965): 328-340. DOI:
10.2307/2218258.
Huswirt, Johannes. Enchiridion Algorismi, Cologne: In officina felicis memorie honesti viri Henrici Quentell,
Smith 1501.
Jullien, Vincent. Descartes. La « Geometrie » de 1637. Available at: http://caphi.univ – nantes.fr/IMG/pdf/
Descartes-Geometrie.pdf.
Liu, Cathay. ‘Re-Examining Descartes’ Algebra and Geometry: An Account Based on the Regulae’, Analytic
philosophy, 50-1 (2017): 29-57. DOI: 10.1111/phib.12093
Lützen, Jesper. ‘The Algebra of Geometric Impossibility: Descartes and Montucla on the Impossibility
of the Duplication of the Cube and the Trisection of the Angle’, Centaurus, 52 – 1 (2010): 4-37. DOI:
10.1111/j.1600-0498.2009.00160.x
Mahoney, Michael. ‘The beginnings of algebraic thought in the seventeenth century’, in, Descartes: Philosophy,
Mathematics and Physics, edited by Stephen Gaukroger, Barnes & Noble, 1980. http://www.princeton.
edu/~hos/Mahoney/articles/beginnings/begin nings.htm).
Mancosu, Paolo. Philosophy of mathematics and mathematical practice in the 17th Century. Oxford: Oxford
University Press, 1996.
Mueller, Ian. Philosophy of mathematics and deductive structure in Euclid’s Elements. New York: Dover
Publications, 2006.
Panza, Marco. ‘What more there is in early modern algebra than its literal Formalism’, in Philosophical Aspects
of Symbolic Reasoning in Early Modern Mathematics, edited by Albrecht Heeffer and Marteen Van
Dyck), 193-229. London: College Publications, 2010.
Panza, Marco. ‘Rethinking geometrical exactness’. Historia mathematica,38-1 (2011): 42-95. https://doi.
org/10.1016/j.hm.2010.09.001.
Pappus. The book 7th of the Collection, edited by Alexander Jones. Dordrecht Heidelberg New York London:
Springer, 1986.
Rabouin, David. Mathesis universalis. L’idée de “mathématique universelle” d’Aristote à Descartes. Paris: PUF,
2009.
Rabouin, David. ‘The problem of a “general” theory in mathematics: Aristotle and Euclid’, in The handbook of
generality in mathematics and the sciences, edited by Karine Chemla, Renaud Chorlay, David Rabouin,
113-135. Oxford: Oxford University Press, 2016.
Van der Waerden, Bartel Leendert. ‘Defence of a “shocking” point of view’. Archive for the history of exact
sciences, 15-3 (1976): 199-210. https://doi.org/10.1007/BF00412256 .
Vol. 73
Fasc. 3-4 RPF 2017

Descartes On The Unification of Arithmetic, Algebra and Geometry Via The Theory of

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Descartes On The Unification of Arithmetic, Algebra and Geometry Via The Theory of

Uploaded by

Copyright:

Available Formats

Revista Portuguesa de Filosofia

Descartes on the Unification of Arithmetic, Algebra and

anachronistic understanding of the meaning and role of algebra in the

2. Ancient attempts to unify arithmetic and geometry

2.1 Euclid’s theory of proportions

The question about the unity between arithmetic and geometry,

A ratio is a sort of relation in respect of size between two magnitudes of

This is an important move towards abstraction since it establishes,

However, there are sound objections against this interpretation of

If four magnitudes are proportional, then they are also proportional

The practice of reproving a theorem betrays evident kinship with the

We cannot in demonstrating pass from one genus to another. We cannot,

Therefore, if we assume that the theory of proportions expounded

PUF, 2009), 93.

Commensurable magnitudes have to one another the ratio which a

These examples show well how Euclid’s criterion to decide whether

If four numbers be proportional, the number produced from the first

We can find a similar proposition concerning magnitudes in proposition

If four straight lines are proportional, then the rectangle contained by

proportions is a general theory thanks to a logic of analogy, which allows

The example cited by Aristotle is about one of Euclid’s “common

can also extend its domain of validity to mutually commensurable magni-

2.2 Geometric algebra

A different approach to the unification of geometry and arithmetic

18. Rabouin, The problem of a general theory, 128-129.

The role of the second book of the Elements as a foundation of algebra

21. “Nam ex nonnullis harum propositionum demonstrantur regulae illae admirabiles

Pappus proposed a way to circumvent the problem using the notion of

3. Descartes’ geometrical calculus as an attempt to unify

Descartes pursued the task of unifying arithmetic and geometry by

On the other hand, Euclid’s theory of proportions enters crucially

… taking a line which I shall call unity to relate it as closely as possible

We immediately remark that the notion of product between segments

proportional (square root) given three segments. Moreover as Descartes

It is not, however, the same thing when unity is determined, because

The definition of multiplication reveals that the role of a unit segment

… by unity, understand a determinate line, which has to any of the

A number is said to multiply a number when that which is multiplied

This definition is framed on the Euclidean notion of number

number m by the number n means to add n multitude composed of m

If four numbers be proportional, the number produced from the first

The equality of two products becomes a criterion to decide whether four

a x b =df the number x, such that 1:a ∷ b:x.

This definition raises problems in the context of the Elements for at

Multiplication is the creation of a number, which is proportionally to the

28. ‘Multiplicatio est numeri procreatio proportionaliter se habentis ad multiplicandum

Fig. 3: “Multiplication is the creation of a number, which is proportionally to the multiplicand as

On the ground of similar examples, which abounded in the arith-

El., VI, 13 (“to two given

29. Descartes Geometria, vol. 1, 147-149.

suggest, albeit cautiously, that Cartesian geometry may contain an outline

construe an arithmetic of segments à la Hilbert is missing. On this basis,

4. The geometrical calculus as a model of algebraic thinking

The previous remarks lead us to inquire whether and to what extent

While the first aspect is an undeniable feature of Descartes’ algebra,

35. Bos, Redefining exactness, 363-375.

broad programme in geometry, appears to occupy a subordinate position

You might also like