Fiction, possibility and impossibility: three kinds of mathematical fictions in Leibniz's work (Q2052226)

Mathematics is the reign of objects which do not exist in the physical world. For, no perfect square or circle exist in our daily experience. Therefore, every mathematical entity can be considered a fiction and the criteria according to which the property of existence is ascribed to a mathematical object have to be specified. Actually, mathematicians have always shared the opinion that, e.g., a circle or ``the number 3'' exist, whereas something like ``a round square'' does not. However, there are objects the ontological status of which was not completely clarified until the 19th century. Among them, some are fundamental in daily mathematical work, such as imaginary numbers and infinitesimal and infinite quantities. Along the history of mathematics, the term ``fiction'' has been assigned to these entities rather than to a square, a circle, or ``the number 3''. Among the scholars who developed a profound -- though not always completely clear -- speculation on the concept of mathematical fiction, Leibniz deserves a particular place because he was the inventor of several mathematical fictions and also thoroughly reflected on their nature. The literature on Leibniz's concept of mathematical fiction is vast and deep. Within this context, the paper under review is worth of a special mention for the care and refinement with which the authors have addressed this not easy subject. In the first section (the introduction), the authors specify their analysis to be based on three different concepts of possibility/impossibility. The first one is given by the principle of contradiction: the already mentioned round square cannot exist because it is self-contradictory. Thus, it is not an admissible fiction. The two other criteria of not-existence are more subtle and concern respectively: 1) irrepresentability; 2) not-conformity with architectonical principles. In the second section ``Symbolic knowledge, symbolic cognition and fictions'', the authors specify important details of Leibniz's theory of ideas. In their context the crucial concept is that of symbolic notion: it is a sensible sign whose feature is given by its function as a support for cognition. The main goal of the paper is so expressed by the authors: ``Using this framework, we propose to elucidate the notion of fiction in terms of a symbolic notion without denotation'' (p. 617). In the third section ``Fiction as a symbolic notion'', the authors, in the light of the examination they develop along the whole section, specify the thesis already anticipated in the introduction: the actually infinitesimal and infinite quantities are impossible (namely without a denotation) fictions because they are geometrically irrepresentable or because they violate Leibniz's architectonical principles (p. 620). After these three general sections, starting from the fourth one ``Leibniz on the fictionality of infinitary concepts'', the authors address a more specific analysis of the way in which Leibniz conceived and used the infinitesimal and infinite quantities. First of all, they refer to an interesting letter addressed by Leibniz to Des Bosses on 11 March 1706 where Leibniz asserted that the infinitesimal quantities are, as the imaginary roots, useful fictions of the mind. For, they short the thinking process and the discovery, without implying any mistake. He also clarifies that such fictions (interpretable as actual infinitesimal) can be replaced by potentially infinitesimal quantities. The authors provide the reader with further evidences where Leibniz confirms this instrumental and pragmatic orientation, already pointed out in an interesting paper published by Arthur in 2009 (p. 622). For the notion of fiction in Leibniz the reader can also see [the reviewer, The complex itinerary of Leibniz's planetary theory. Physical convictions, metaphysical principles and Keplerian inspiration. Cham: Birkhäuser/Springer (2016; Zbl 1354.01002), pp. 50--53]. For the concept of fiction in Newton, also compared with Leibniz's, see [\textit{R. Pisano} and the reviewer, ``The fiction of infinitesimals in Newton's works. On the metaphorical use of infinitesimals in Newton'', in: Isonomia -- Epistemologica. Vol. 9. Reasoning, Metaphors and Science. 141--160 (2017)]. However, Leibniz expressed many doubts on the ontological status of such infinitesimal quantities (p. 622). The authors refer to two examples of the way in which Leibniz used the infinite and infinitesimal quantities. It is worth referring to the first one, at least. The situation (pp. 623--625) can be summarized like this: consider a general hyperbola $y^mx^n=a$. Then consider the rectangle obtained projecting a point of the hyperbola on the two coordinate axes. Imagine the projection on the abscissa $y$ to be infinitely small and that on the ordinate $x$ infinitely long. Leibniz argued that if $m<n$ the area of the rectangle is infinite, if $m>n$, it is infinitesimal and if $m=n$, it is finite. The authors refer here to the fundamental contributions of Knobloch. The second example concerns the rules of differentiation. Thereafter, the authors connect in the fifth section ``Mathematical fictions and impossibility'' (pp. 627--629) the problem of the ontological status of the mathematical fictions with that of mathematical existence. They recall that, according to Leibniz, the criterion of existence is possibility and the lack of contradiction indicates possibility. Therefore, though in nature, to follow Leibniz's example, no perfect pentagon exists, it is a possible object and, hence, it has a mathematical existence. Leibniz establishes, thence, a rather clear criterion. However, as to ``the fictionality of infinitary objects [he] reveals a more prudent attitude'' (p. 629). Here the concepts of paradoxicality and improbability come into play. They are far more nuanced and not easily interpretable concepts. The authors claim that ``This suggests the idea that for Leibniz both the existence and the mathematical possibility or impossibility are not a matter of absolute opposition but they do to some extent admit degrees'' (p. 629). The following fundamental sixth section ``Three concepts of mathematical possibility and impossibility'' (pp. 629--637) is dedicated to clarify such assertion. In this section, the authors clarify furtherly the nature of the couple possibility/impossibility beyond the concepts whose existence is not possible because of the contradiction principle. As already stressed, a second kind of impossibility is given by the lack of representability. So, for example, ``the infinitely small abscissa of our first example can be represented only analogically by a finite abscissa'' (p. 630). The imaginary roots are also affected by this kind of impossibility. The four following pages are devoted to clarify this particular notion of impossibility: objects such as the imaginary roots of an equation are impossible because they cannot be represented geometrically. Nonetheless, they are the result of operations which are mathematically thinkable and legitimate (p. 633). Therefore, if they are useful to solve specific problems, they must be used, which is not the case with self-contradictory objects. The third kind of impossibility is the most subtle and connects Leibniz's conception of mathematics with his metaphysics. The authors reiterate that, according to Leibniz, a mathematical object is a fiction, is not really existing if its existence is not compatible with the architectonical principles of sufficient reason and of continuity. For, on more than one occasion, Leibniz claimed that infinitary objects do not exist because they are not consistent with the principle of sufficient reason (p. 635). The authors offer an interesting example in this regard, which they explain perspicuously (ibidem): Leibniz refused to consider the motion as a series of successive infinitesimal leaps because this conception would violate the principle of sufficient reason. This clarified, it is to point out that Leibniz, in this case too, recognised the utilization of these mathematical fictions to be useful and legitimate. Furthermore, they have to be regarded as not-existing unless a convincing proof of their existence be exhibited (p. 637). Therefore, his attitude was rather prudent. In the seventh section ``The reconsidered concept of mathematical fiction'' (pp. 638--642), the authors summarize their view on the notion of mathematical fiction as an object lacking of denotation: in Leibniz, we have fiction\(_1\), referred to self contradictory entities; fiction\(_2\), referred to irrepresentable entities; fiction\(_3\) referred to entities lacking of denotation because they violate architectonical principles of Leibniz's metaphysics. After that the authors speak of the way in which mathematical fictions can be expressed (pp. 638--639): 1) verbal or written common language; 2) specific symbols as those of the infinitesimal calculus; 3) analogical and semiotic diagrammatic representation, such that the geometrical one through which the infinite and infinitesimal quantities are represented in a system of coordinates as if they were finite. As a conclusive step of their research, the authors analyse what Leibniz named ``in appearance imaginary quantities''. They enter the play while solving the irreducible algebraic equations of degree three, which have three real solutions but in the resolutive formula of which imaginary numbers appear. In this regard the authors write: ``Thus, the concept of an `in appearance imaginary quantity' or of an `in appearance impossible quantity' seems to imply an objection against our interpretation, for these quantities seem ultimately to refer to real quantities, since imaginary quantities are put in equivalence with real quantities'' (p. 639). In the light of two passages by Leibniz, the authors interpret the ``in appearance imaginary quantities'' as a syntactic procedure of symbolic nature. The expression of real roots through such quantities is a new mathematical operation (p. 640). Therefore, such quantities do not conflict with the interpretation offered of Leibniz's mathematical fictions. It seems to me that this brief part on the ``in appearance imaginary quantities'' is the less convincing of the whole paper. As a matter of fact, these quantities do not need a specific treatment within the general picture described by the authors. For, it is a usual procedure that mathematical fictions are used to obtain results concerning real quantities. This is extremely common in infinitesimal calculus. Therefore, it is not completely clear to me the necessity to dedicate a specific consideration to the ``in appearance imaginary quantities''. ``Concluding remarks'' and the ``References'' follow. This paper is very valuable. It is refined from a conceptual point of view and offers new insight in a difficult question such as the nature of mathematical fictions in Leibniz. The distinction between three kinds of fictions is noteworthy and surely deserves to be furtherly discussed and clarified. Thus, the authors also offer material for further researches on this subject.