Kant’s critique of the leibnizian theory of organisms



Kant’s critique of the Leibnizian theory of organisms:

An unnoticed cornerstone for criticism?

Philippe Huneman (Rehseis, CNRS, Paris)

In the Antinomies of reason, there is a short text whose meaning can be grasped only by determining what is exactly Kant’s target in the critique that he elaborates there (A525/B553-A528/B556). This target is the Leibnizian conception of organisms as “infinitely articulated machines”. What is at stake in this critique is thus the presuppositions of such a conception. Then, reading Kant’s critic will may allow us to understand the extent to which the theory of organised beings in Leibniz is embedded in his own metaphysical commitments. Thus, reciprocally, such an interpretation will grasp the precise link between questions about organisms and metaphysical problems, so that we will understand why the critical project is in itself concerned by the question of defining organisms. In fact, this critique of Leibniz is important since the leibnizian conception shares many presuppositions and schemes of the common metaphysical framework of what Kant called dogmatism, and which he critically addressed in his works after 1764. Whereas While Leibniz addressed the question of organised beings in order to avoid Cartesian mechanism and vitalism (mainly in the form of Stahl’s animism), Kant’s critique points out that this project could not be fulfilled within Leibniz’s guidelines. The present paper reconstruct Kant’s critique within this background, and questions its significance for criticism in general.

After having briefly reminded the Leibnizian conception, I shall expose present the concept of those Antinomies of reason, so toin order to expose the second Antinomy; I shall will justify that at this point, the idea of organised beings raises a specific problem, so that, in the course of the second Antinomy, such issue should be addressed and one has to define some requisites for any theory of organisms. Then I shallI end up by stressing the consequences of such an analysis upon the meaning of the divorce between Kant and Leibniz (and, more generally, classical metaphysics), and its consequences concerning further Kantian conceptions.

1. Leibniz and organisms

The Cartesian thinking initiated the metaphor of the animal-machine; this meant two often conflated theses: a. the origin and the functioning of living beings are plainly understandable by laws of mechanics; b. machines are useful and necessary models to conceive organised phenomena. Leibniz kept the requisite of an absolute lawlikeness of nature, such that organisms are submitted to mechanism. As he wrote against Stahl, responding to his Theoria Medica vera (1712): “nothing happens in the bodies that could not be explained by mechanical, hence intelligible, reasons.”[1] The only limit to this explanation is our present state of understanding; what is then, not mechanically intelligible, is not bodily phenomena, but the principles of this mechanics itself, since they require an appeal to the reasons why God created this world (with those natural laws) rather than another possible world. “All phenomena could be mechanically explained, if we were to understand them enough; but the very principles of mechanics could not be geometrically explained, since they hang on more sublime principles, which attest its author’s wisdom concerning order and perfection in the work.”[2]

However, he thought that, contrary to the machines made by us, or “artificial machines”, living creatures are “natural machines”, which means “infinitely organised machines”. Leibniz formulates it in those words : the machines from nature “e.g, living bodies (…) have a really infinite number of organs, and are so well designed and in beware of any accident, that it is not possible to destroy them. A natural machine is still a machine in its tiniest part, and, moreover, it stays always this same machine as it already was, being only transformed by the folding it undergoes, and being sometimes extended, sometimes compressed, and like concentrated when one believes it is lost”, whereas “a machine made by the art of man is not a machine in any of its part. For example: the tooth of a wheel has parts or fragments that are no more something artificial [= designed], and do not wear any more the trace of the machine regarding the use it was intended to.”[3] Hence, the difference between machines and organisms is the difference between finite and infinite organisation, e. g. divine and human techniques. The whole conception then relies on this thesis expressed in the Theodicy’s preface, according to which there is an infinite difference of degree (out of nature) between divine and human understanding[4]. One has to notice, here, that the technical paradigm through which the living being has to be apprehended, remained constant from Descartes to Leibniz, albeit the latter criticizes the identification between organisms and machines. For Descartes, mechanism meant that those natures that we find in animals, have no special causal powers to account for specific features of organisms, they are the same kinds and causal powers than in plain nature and in our machines : for example, the fire in the mammal’s hearts is not another fire than the one in machines and fireplaces[5]. But Leibniz avoids this question of natures : his mechanism means only that the laws according to which organisms has to be understood are the very same universal laws of nature, so even if hearts were other substances than fireplaces, they nevertheless obey the same laws and should be understood in the same ways. To a certain extent then, Leibnizian mechanism is an epistemological one, and Descartes’ mechanism is rather ontological. However, any metaphysical critique of their conceptions will have to challenge this technical paradigm common to epistemological and ontological mechanism, as the case of Kant will confirm it.

The point of Leibniz is that this infiniteness of natural machines accounts for the feature aforementioned, which has always been taken as constitutive of life, namely: the faculty of self repairing. Because, being infinite, an organism is build in such a way that if you alter or withdraw one part, it is still infinitely organised, hence the lesion is not so crucial since the organism does not lose its identity – and this self-conservation is the basis of the capacity of self repairing, which is all the more intensive than the creature is less complex, as we know. And if the alteration is not effective upon a living being, this means that its unity is an internal one, opposed to the external unity of a technical machine - that is Leibniz’s claim.

The argument in the beginning supposes Leibniz’s criterion for real substances: they should be defined not by extension (pace Descartes) but by a force, since the essence of the substance has to account for its accidents[6]. (Else, we would go against the logical principle of truth, which for Leibniz says that predicates are inherent to the substance.) But pure extension can’t account for the accidents happening to the substance. So something akin to soul, a kind of force, is the essence of substance[7]. In this context and concerning our question, Leibniz distinguishes three stages of unity within matter, and contrasts them with living beings. “through soul or form, there is a real unity which corresponds to what we call me in us ; which could not take place neither in the machines of art, nor in matter, no matter how much organised it can be; (matter) that one can only consider, either as an army, or as a herd, or as a pond full of fishes, or as a watch composed of springs and wheels.”[8] Here, three kinds of multiplicity occurring to matter are opposed to living beings: two of them (herd, pond) are in natural matter, two of them are “matter of art” (army, wheel). In any case, the unity is external, which means both that what is substantial comes from the substance of the individuals within it, and that the unity of the multiplicity comes from the outside (the pond, or the shepherd or the army’s commander, or finally the watchmaker). So, “no matter how matter is organised”, this kind of unity will never be changed. “Any machine supposes some substance in the pieces from which it is made”: e.g., notwithstanding the complication and sophistication of a machine, it will share the status of armies or herds in the sense that the substantiality comes from the substantiality of the pieces; hence, the unity of the whole will still be provided by some instance outside the whole.

But if an external unity means that some external entity – for example, a designer – built the machine by putting the parts together, created its unity, then in this case, he is therefore entitled to destroy this machine by withdrawing parts, and the organisation of such a machine is thereby destroyable by any external action. The case appears quite different when it comes to natural machines, because infiniteness of organisation prevents the machine from being externally altered: if you take away a part, the organization will by definition still be infinite. Thus, its internal unity, as correlative to its infiniteness of organisation (given that no external designer can put together an infinite organisation) becomes a source of self-preservation. That is the reason why, being so constituted in the internal unity of an infinite organisation, living creatures can be told to have a soul as a self preserving principle, which was denoted by the “me” in the above quotation. The three concepts of (internal) unity, (infinite) organisation and soul are strictly correlative and collectively required to define living beings[9]. Hence, the salient character of organisms, which is the conservation of form through the indefinite change of matter, is explained through this correlation between internal unity and infinite organization: since, in fact, this unity governs an infinity of parts, any change in the matter of the parts is not likely to modify the organization of the whole.

This ontological articulation of a principle of unity (soul)[10] and a kind of organisation (infinite organisation) allows Leibniz to epistemologically equate knowledge of organisation of natural machines, and discourse on their souls. There could be no gap between talking of the purposes of animal souls and describing the causes of their organic motions according to mechanical laws. Leibniz will later express this harmony in the framework of his Monadology: « Souls act according to the laws of final causes through appetitions, purposes and means. Bodies act according to the laws of efficient causes or motions. And the two realms, the one of the efficient causes and the one of final causes, are harmonic between them.”[11]

Then, corresponding to the difference between technique and infinite technique, that yields the ontological difference between the two kinds of “machines”, there is the difference between external and internal unity. Those two couples of concepts allow Leibniz to specify the peculiar ontological character of living beings without giving up the requisite of the general lawlikeness of nature. It should here be noted that Leibniz can not allow organisms to be excepted from lawlikeness of nature, since this would destroy any science of nature. In effect, since organisms do present the highest form of unity and substantiality, Leibniz somehow appears committed to the radical thesis according to which the very reality is the organisms. “I restraint corporal or composed substance to the mere livings or organic machines in nature; what’s left is for me only aggregates of substance, that I call substantiates; the aggregate is merely an accidental being” [12] Hence, if organisms were not under a mechanical lawlikeness, there would be no science of nature. This motivates him to counter the Stahlian notion of organisms, as opposed to mechanical laws, while challenging the Theoria Medica vera. And this lawlikeness should be somehow accorded in advance to divine purposes in nature: since the Discourse of metaphysics, Leibniz then recognized that mechanical intelligibility of whole nature (with no exceptions) went hand in hand with the purposive character of everything[13], since the metaphysical grounding of mechanisms is the maxims of convenience that rule God’s choice of this world with its particular laws, maxims entailing purposiveness. Hence, at the level of the plain metaphysics of nature Stahl is misleading since he opposes as “mechanism” and “organism” two instances of the same thing; but at the level of the epistemology of organisms, his mistake has to be demonstrated, and a concept of living beings has to be formulated which accounts for the lawlikeness of living nature.

In one of the strangest passages of the Critique of pure reason, Kant undertakes a refutation of this conception. In fact, the Leibnizian idea appeals to the infinite divisibility of space. For this reason, this conception meets the philosophical problems raised by such a contention. Kant deals precisely with those problems in the Antithetic of reason. In this framework, he will get the arguments for rejecting the leibnizian theory of organization. After situating the context, we will investigate the meaning of this rebuttal, and, above all, the reason why Kant had to set a room for such very detailed critics in the general argument of the Transcendental Dialectics.

2. The second Antinomy and the organized beings

a. The context

While the understanding was the “power (Vermögen) of rules”, reason is now the “power of principles” (A299B356). In other words, the understanding produces the rules which make experience possible, those a priori synthetic principles that the Transcendental analytics explicitated. Reason on its side brings towards unity those rules used by the understanding in order to create unity within the manifold of phenomena (A302). Kant names “principle” the possibility of such a unity. The principle which says “with a given conditioned rule, all the conditioning rules are given” is the only principle capable to unify the rules of the understanding. Reason demands the unconditioned, “all the conditions” being ipso facto Unconditioned.

In the reasoning aiming at the conditions of a conditioned statement – a “regressive reasoning” -, one can reach knowledge only whenever we “suppose that at least all the members of the series on the side of the conditions are given” (A332), which becomes the principle of reason. The principle of reason hence entails the following tension: in the same time, the possibility of the reasoning (Vernunftschlüsse) requires presupposing as given the totality of the conditions in the three perspectives (e.g. categorical, hypothetical and disjunctive conditions); and this totality can not be given to the intuition (since intuition can’t access to totalities but only to particulars). The problem is that the Transcendantal Analytics showed that the conditions of objectivity implied the possibility of being represented in the intuition. For this reason, the nature of the “givenness” postulated by such a presupposition of reason raises a question: subjectively necessary[14], is it objectively valid ?

Now, the cosmology – or metaphysical discourse on this totality denoted by the word “world” - stems from the necessity that “the totality (of the real things), as they fill space and time, must be represented under the concept of a world” (Fortschritt…, Ak.XX, 287). Within cosmology, the transcendental dialectics takes the appearance of the antinomies, as it follows from the structure of the Unconditioned when it comes to this idea of world[15]. The basic consideration is that the relationship between the conditioned and its condition is, in this case, a regress in a series : series of moments in time that precede one each other, series of parts in space that fill one another, causal series, series of the dependency between the contingent item and its conceptual conditions. So, in the cosmological field, there will be four senses of the Unconditioned, corresponding to those four series, series defined by the four kinds of categories[16]. In those cases, the Unconditioned is likely to be conceived in two different ways: either the whole series is Unconditioned, hence the regress is an infinite one, like in the causal series conceived by the Spinozists; or the Unconditioned is a member of the series (the last one), which appears then, unlike all other members of this series, to be without conditions (A416)[17]. This latter possibility involves a finite world, composed of simple substances, leaving room for free actions and related to a God as its necessary condition; the former option involves an infinite world of infinitely divisible phenomena, embedded in infinite causal series, and without any necessary condition[18]. Now, the two options concerning the nature of the Unconditioned give birth to one thesis and one antithesis related to each of the four questions stated above, concerning the integrity of the conditions in a regress.

The basic problem is that “all those questions deal with an object which is not likely to be given outside our thought, namely: the absolutely Unconditioned of the synthesis of the phenomena” (A481). Such an idea of “world” can be conceived (as it bears no contradiction) but this does not entail that it can be known, because it has no relationship with any possible object. But this status of the idea remains hidden until that one has investigated the principle of pure reason (e.g. “when a conditioned is given, the whole series of conditions is given”) and specified whether it concerns the things as merely conceived, or the things given insofar as that they are experienced, namely, the phenomena. But, the phenomena, being in the apprehension nothing more than an “empirical synthesis (in space and time)” according to criticism, are given only within the apprehension. Hence, no conditioned phenomenon brings with itself the series of its conditions, but such a series takes place only within the regress from a condition to a condition, which, by effectuating the synthesis, constitutes those phenomena that are, one after the other, conditions (A499)[19].

b. Bodies in general and their constitution

In any interpretation, concerning the phenomena (e.g. everything that is in space and time, hence everything that falls under the concept of “world”), the series of conditions is not given but proposed together with the conditioned – and the task of the reason is therefore to determine it. (The mistake that yields the in general Antinomies conflates “given” and “proposed”.) Hence, the series of conditions happens in a regress which is a successive synthesis, therefore the “in the same time” stated in the principle of pure reason does not obtain any more. In this case, the principle of pure reason becomes an axiom allowing the regress in the series of conditions of a given conditioned. It is not a “constitutive principle” of reason, e.g. a principle which would constitute purely rational objects (=wholes) beyond the sensible, but a so-called “regulative” principle. By this word, Kant means that this rule “postulates what should be effected by us in the regress, but does not anticipate what would be given in the object prior to any regress” (A509). This rule enables us to go from the conditioned to the Unconditioned through all the conditions subordinated to one another[20]. Therefore, to solve the antinomies means to explain how the principle of pure reason can provide a rule for the four cosmological syntheses which gave birth to the antinomies, in other words, to explicate “the empirical use of the regulative principle of reason with regard to all the cosmological ideas” (ib.).[21].

In each of the cosmological Antinomies, the regulative principle indicates that “in the empirical regress, there can be encountered no experience of an absolute boundary, and hence no experience of a condition as one that is absolutely unconditioned empirically.” (A517) Now, concerning the cosmological Idea of the “totality of the division of a given whole » - which is the Idea of elementary substances composing an entity -, what does this mean? Here, notice that what is at stake is any composition, be it or not essentially in space[22] It means that an entity can not be composed by simple parts that a division would ultimately reach, because this would imply that those parts pre-exist to the empirical regress which finds out the parts, so that we would “attribute a peculiar existence, prior to the experience, to a simple phenomenon which can exist only within experience” (Prolegomena…, §52). So the Thesis, “every composite substance in the world consists of simple parts” (A434), is ruled out by the correct understanding of the regulative principle. But this does not entail that we subscribe to the other position, the Antithesis, which says that the bodies are composed of an infinite number of parts. Some commentators (e.g. Bennett) have noticed that the Antithesis seems closer to Kant’s critical position and its regulative principle. However, as Grier (2001, 209) emphasizes, the Antithesis takes the spatial character of the composite, which implies its infinite divisibility, as an absolute property rather than as a condition of our intuition, and hence misses the critical point. If we keep on considering this text, this difference is explained by Kant through considering the status of the division itself.

Concerning the regress in parts of space and time, Kant distinguishes between the regress in indefinitum and the regress in infinitum. The former applies to the first Antinomy (limits of the world in space and time), because the condition is always outside the conditioned, since any part is so to speak environed by a bigger part, and any slice of time has prior to itself the past periods. Hence, the conditioned is never given within the condition, so that each time, indefinitely, I must repeat the operation. On the contrary, in the second antinomy, if I see a body, all its parts are contained within itself: all the conditions are given within the same limits as the conditioned (A524). The wholes seem composed of parts, because they are from the beginning perceived or conceived with parts in it. Nevertheless in this case a whole is “divisible to infinity” but not “consist of infinitely many parts” (ibid), because whereas the parts are included in the same limits than the whole, the division – which determines the parts – is not previously included in the first intuition of the whole, but takes place within this regress from part to parts which “which first makes the series actual” (ibid). So, the parts are since the beginning contained in an aggregate which forms the whole, but the series of the parts, within which those are coordinated, is never given, only achieved in the regress: it is “infinite successively and never is as a whole”. Malzkorn comments by saying that the totality of the parts is never given as such (1999, 274). But more precisely, what is never given is the way this totality of parts is intrinsically connected; since precisely the determinate connection of two parts in a division follows from the determination of the boundaries through the act of dividing. Hence, the regress can not present an “infinite multiplicity or the taking together (Zusammennehmung derselben) into one whole” (A524), because the determinate connexion of the parts taken together (which distinguishes the series from the aggregate) does not pre-exist to the regress. The evidence for that is the simple fact that, within a same whole, the division could be effectuated in various manners, which would not be the case if the parts were existing prior to the division[23].

Finally, the body is thereby infinitely dividable, but not constituted of an infinite number of parts. “The latter (= extended whole) is thus divisible to infinity, without, however, consisting of infinitely many parts.”(A525/B553)[24] For this reason, the antithesis, e.g. the claim that matter is constituted by an infinity of infinitely small and infinitely divisible parts, or infinitesimal parts, does not obtain either. Regress, which is the act of synthesis ruled by the regulative principle of reason that compels us not to stop somewhere in the division, does not tell us anything about the composition of matter in itself, independently and prior to this synthesis.

This distinction here rests on the German opposition between teilen, “divide”, and ausmachen, “constitute”; the German carries the idea that the result of a division is always a part, hence dividing is decomposing the whole into its parts. This connotation is lost in English. That’s why for Kant division goes with “decomposition”, because a division picks out parts so dismantle the composites, but decomposition has to be distinguished from the constitution or the composition by constitutive parts. The English word “decomposition” translates Dekomposition, which is quite synonymous with Teilung, but is not the opposite of a composition into constitutive parts; “constitutive parts” mean that the parts are determined, hence the identities of the parts and then their boundaries are somehow fixed, hence the decomposition is no more possible. Guyer and Wood translate Zusammensetzung, which means “all the parts taken together”, as “composition” (A525)[25], but this “composition” does not ipso facto give constitutive parts, because it only concerns only the totality of the regions of space. It does not mean composition as assembly of constitutive parts. The critical point is precisely that the decomposition that goes along the division gives not the constitutive parts. Hence in the Antithesis, where we deal with matter in space, this composition would be a real, concrete one, it concerns the thing as existing, whereas the division and the parts into which the division decomposes the composite, concerns only the intuition of the whole, or the mathematical aspect.

c. The organism as infinitely articulated.

It was, in fact, the infinite divisibility of space, which allowed Kant to state his conclusions, thus those conclusions were about continuous quantities, this continuity being “bound to the fact of filling a space”. Here, we must state a discrepancy in the second Antinomy, concerning precisely the spatial character of the composition in both thesis and antithesis. As Grier (2001) has noted, the meaning of the composition is not exactly the same in the thesis and the antithesis – and this is the precise point of the antinomy. Whereas the thesis is concerned by composition in general, assuming that matter is prior to the forms of intuition, the antithesis contends that matter is submitted to these forms, hence to space and time – but a space thought uncritically as ontologically absolute. So, matter in the thesis is not the empirical concept of matter, but a noumenal concept of matter[26]; on the other hand, the antithesis proves infinite divisibility and composition of matter as spatial matter. In effect, if space exists in itself, as the form of the things in itself, division and composition will be the same thing – this difference cannot be made unless we assume the transcendental ideality of space. According to Grier, the common ground of thesis and antithesis is the fact that a space conceived as absolute thing is thought as this compared to what it has to be decided whether matter is prior or not. This ground could be expressed by saying that both thesis and antithesis conceive divisibility as identical with composition; they just differ about whether composition stops on simple parts of not. In effect, in the thesis, the entity – whose matter is supposed prior to space – is a composite, and the division naturally divides it into the elements of the composite, hence division and composition are the same. And in the antithesis, the result of division is precisely those spatial parts that, in turn, constitute the composite supposed to be essentially in space. Saying that division occurs through regress – which is the critical position - , hence that constituting parts are not given, means that division and composition are different.

But now, even if the Thesis could be true of things in itself (indeed, this is not testable[27]), this does not matter when it comes to what Kant is concerned with now, e. g. the appearances of those things, which should occupy space, hence be continuous. This consequence is clearly noticed in Metaphysik L1 : “Leibniz thus says : all substances are monads or simple parts that have power of representation, and appear among all phenomena. But (…) all appearance is continuous, and no part of the appearance is simple, thus bodies do not consist of simple parts or monads. However if they are thought through the understanding the substantial composites consist of simple parts.” (Ak. 28, 208) Hence in any case – whether it’s about noumenal matter as in the Thesis or spatial matter as in the Antithesis – the constraints of space imply the Kantian solution: divisibility without infinite composition.

As it is clear from the quotation of the lectures on Metaphysics, Kant’s concern in the second antinomy seems rather Leibniz’s Monadology, than the composition of space and the truth of atomism. But more precisely, Kant’s interrogation and critique of Leibniz’s Monadology inevitably relies on considerations of space and its divisibility. So, in the end, atomism and Monadology are the target of Kant’s Dialectics, here. This is somehow legitimate, since monads are a kind of metaphysical atomism, as Leibniz says himself: "there are only atoms of substance, namely units and absolutely deprived of parts, that are the sources of actions, and the first absolute principles of the composition of things and like the last elements of substantial analysis. We could call them metaphysical points : they have something vital and some kind of perception, and mathematical points are their points of view to express the universe. »[28] Kant’s point is only that this Monadology says nothing about the empirical world.

Now, if Monads are ruled out, and if all appearances in space, since they are continuous, are infinitely dividable, what about entities composed of discrete parts? I think that this last question addressed by Kant in the solution of the second antinomy, which will concern Leibniz’s doctrine of organisms, arises because of the following reasons. First, even if the antithesis is dismissed, so even if we can’t say that any whole is composed of an infinite number of parts, is it possible that some composites are composed of an infinite numbers of parts ? In effect, composition and division are not related in the same way in non continuous wholes; but continuity was precisely the reason of such solution. Second, even if spatial composition is as such continuous, some entities are composed of discrete parts, e.g. machines; those parts can in their turn be either continuous or composed of discrete parts too: those entities could then be candidates for being these infinite composed entities that are still possible. Third, the conception of simple part, as used by Kant throughout the Antinomy, oscillates between the logical meaning of an individual in a set of individuals, and the mereological meaning of a part in a whole, in a way which is not always explicit to Kant himself[29]: now, since the latter meaning has been used in the demonstration of the antinomy, the former meaning is left. And since discrete entities are set of parts, may be precisely the second meaning of “parts”, here required, will lead us to another result concerning division.

So, what happens when we try to extend the regulative rule of division to “cover the multiplicity of parts already detached with certainty in a given whole, constituting a quantum discretum” (A526). Given the importance of continuity of space in the applicability of this rule of judgement, one can not easily extend the rule of division to a whole which is already divided, namely, an “articulated (organised) whole” (ibid). Kant argues that this is indeed transcendentally impossible, by the following reasoning.

The solution of the antinomy was based on the regress determining the parts, in other words, before the regress there is not any partition of the phenomenon, “the whole is not in itself already divided (eingeteilt)”, since what we have is only “a multiplicity of parts absolutely undetermined in itself”. But what is peculiar to an articulated whole such as any machine or any organism, is « to be represented in this concept as already divided. » (A527). An articulated whole is effectively organised, because the articulation, Gliederung, entails that the division is not arbitrarily drawn: it follows a design, it is thus determined by a concept. The Kantian notion of articulation is made clear in the Methodology of the Critique of pure reason, in the context of the characterisation of a systematic science and scientific development. Kant argues that parts of the sciences are articulated, which means that the boundaries between them are somehow conceptually and a priori determined. “I understand by a system, however, the unity of manifold cognitions under one idea. This is the rational concept of the form of he whole, insofar as through this the domain of the manifold as well as the positions of the parts with respect to each other is determined a priori. The scientific rational concept, thus, contains the end and the form of the whole which is congruent with it. The unity of the end, to which all parts are related and in the idea of which they are also related, allows the absence of any part to be noticed in our knowledge of the rest, and there can be no contingent addition or undetermined magnitude of perfection that does not have its boundaries determined a priori. The whole is therefore articulated (articulation) and not heaped together (coacervatio); it can, to be sure, grow internally (per intus susceptionem) but not externally (per appositionem), like an animal body, whose growth does not add a limb but rather makes each limb stronger and fitter for its end without any alteration of proportion.” (A832/B860). So, articulation implies a fixed a priori determination of boundaries; the reference to animal growth is essential, and in this context means than if an articulated whole is to grow, this growing will have a specific character opposed to mere aggregation from the outside. This is Kant’s reformulation of Leibniz’s internal unity. But, articulation seems the generic concept for both things likely to grow, like animals, and not likely to grow, like technical artefacts. Consequently, each part in it has an a priori a determined position, and for this reason, when the articulated whole is called “articulated” or “organised”, “organised” means any arrangement of parts according to a concept. To this extent, a watch (with its wheels) as well as a beaver (with its organs) are organised wholes. Now, as we saw it, the Leibnizian theory of organisms says that the latter of those wholes possesses an infinite number of parts, each of those being in its turn infinitely organised, e. g. each being made by parts articulated to one another according to some form (and the same holds for its component parts, and so on). The position that Kant will criticize is exactly the Leibnizian thesis: “through the decomposition into parts one always discovers new organs (Kunstteile)” (A526) [30]. Kant will show that this position is not allowed by the solution of the Antinomy, albeit the regulative principle of reason supports infinite divisions of spatial entities. “To assume that in every whole that is articulated into members (organized), every part is once again articulated, and that in such a way, b dismantling the parts to infinity one always encounters new complex parts – in a word, to assume that the whole is articulated to infinity – this is something that can not be thought at all, even though the pars of the matter, reached by its decomposition to infinity, could be articulated.” (A526/B554, my emphasis)

Indeed, this position is a contradictory one. In effect, if the regress can always divide the body one more time, this presupposes that the body is not already divided. “For the infinity of the division of a given appearance in space is grounded solely on the fact that through this infinity merely its divisibility, i.e. a multiplicity of parts, which is in itself absolutely indeterminate, is given, but the parts themselves are given and determined only through the subdivision – in short, on the fact that the whole is not in itself already divided up.” (ibid) So, the Leibnizian position denies what is the condition of infinite divisibility through regress, whereas it claims an infinite division of the given organic body. According to this thesis, one would consider the division as already achieved albeit infinite, since it is represented in a concept (this point defines precisely what is articulation, Gliederung); and in the same time one would conceive it as a never-fulfilled series, since one applies here the rule of regress which means producing an infinite division through the regress itself, the form of this division being undetermined prior to the regress. So, the leibnizian concept implies simultaneously, that the whole is already divided, as an organised being, and that the regress divides it indefinitely, as infinitely organised, because the infiniteness of the division is produced by the regress alone (according to the regulative principle of pure reason). Hence the verdict of contradiction: “one contradicts oneself, since this infinite development is regarded as a series that is never to be completed (as infinite) and yet as one that is completed when it is taken together (In einer Zusammennehmung als vollendet) ” (A527/B555) Here, the completeness of the division is a consequence of its a priori character, defining the concept, as it is clear from the text of the methodology : no supplementary part can be added, because it would contradict the a priori characterization of the parts. This means that no supplementary division through the regress can divide the parts provided by the initial division as articulation. The idea of new parts always to be discovered within the parts through division, essential in Leibniz’s conception of organisms, proves therefore contradictory: that’s why Kant have said that this “cannot be thought at all”.(A526/B554) It’s not like an Idea of reason, a conception that has no reference or testability in experience, but is nevertheless conceivable and could be used as principle of reason : it is a plain contradiction.

In the end, the mistake involved in the concept of an infinitely organised being relies on a confusion in the conception of space. What in fact allows me to apply the principle of regress, is that the phenomenon is continuous, hence, as it fills space, infinitely divisible exactly like space. “The infinite division indicates only the appearance as quantum continuum, and is inseparable from the filling of that space.” (A527/B555) In this sense, any appearance, as such, is spatial and hence infinitely divisible; the articulated beings, be they organisms or machines, are in the same case, so as appearances they are infinitely dividable. But Kant argues that the division in parts which is determined by a previous idea is not the same as this division through the regress. As spatial beings, they are not divided in the same way than they are divided as organisms. So in the second sort of division I am not compelled to divide in an infinite way. This subtlety accounts for Leibniz’s mistake, because he conflates the first division, necessarily infinite, with the second, necessarily limited. In effect, concerning the division of the organised being as such, Kant then writes : “as soon as something is assumed as a quantum discretum, the multiplicity of units in it is determined; hence it is always equal to a number” (A527). This number may be huge, but it will in any case be determined, else we would still have a continuous spatial thing and no more. Therefore, what is represented, whether I mean it or not, in the concept of an articulated (organised) whole, is always a determinate number of parts. Since the parts have to be determined, the number of parts has to be finite. The organisation can extend very deeply: each visible part may be dividable in a great number of distinct parts, and “only the experience can settle how far the organization in an articulated body may go” (ibid). The articulation which is known to be a priori determined, is a posteriori discovered: this is why we could mistake this organisation as infinite, by thinking that there could be infinitely smaller parts left within what we have already discovered. But there can be no infinite composition of parts under the concept of the organized whole. Suppose that the last result of the division according to the prior concept is found: now, either this part can be divided in two parts, that are constituting parts, but then the whole comprehends parts that are not determined by the concept (since the division according to the concept has ended), then the whole is not an organized whole since indeed there is no concept ruling the division; or this part can not be divided then the division is finite.

So the number of parts is an empirical question, together with what we would now call the number of levels of organisation[31]. However, one can know a priori that, in such a body, there will be some parts – no matter how small they are – which will not be organised in themselves. Yet, sometimes, we have no knowledge of the composing parts of an organized being, and the latest elements reached by the empirical decomposition seem themselves organized[32]. Kant thus writes: “even if it (experience) were certain to attain to no inorganic parts, such parts must nevertheless ay least lie within possible experience.” (ibid) The contingent reach and extension of experience – relying on our present cognitive and scientific abilities – is not the criterion of the end of the decomposition of organized beings, but this end is prescribed in their concept, as concept of organized beings in space and time, hence as a special kind of appearances in our possible experience. Hence, even if the current experience does not contain those limits, they are nonetheless contained in possible experience. The incompleteness of actual experience, the fact that organic parts are out of our reach, does not prove any infinite composition of organized beings, but is only a contingent fact.

Thus the case of organized beings is complicated because they are submitted to two kinds of divisions: as organized beings, there is this composition in organised parts which in the end should stop by inorganic parts, in a way prescribed by the concept of the articulation. As appearances in space, they are subject to the transcendental regress from wholes to parts, which is not concerned by the empirical fact of the limits of division. Hence: “how far the transcendental division of an appearance in general may reach is not a matter of experience at all, but it is rather a principle of reason never to take the empirical regress in the composition of what is extended, in conformity with the nature of this appearance, to be absolutely complete.” (ibid).

So there are two kinds of incompleteness of division, and Kant argues that we must not conflate them. The first one is an empirical one: we have divided an organised being, and the parts that we found are organic ones, so we think that it could be infinitely divided further. But this is false, since the limits of the division, although not experimented yet, reside in possible experience – they should lie within it, due to the concept of the organised being. And there is the incompleteness of the division of this being considered as an appearance in space. Here, it is a transcendental rule that division should be indefinitely pursued, and to take empirical regress as radically incomplete, since reason prescribes us the rule of always continuing the regress from whole to parts concerning extended matter. Conflating those two incompleteness of division leads precisely to a Leibnizian mistake: organised beings would be infinitely organised since we cannot but empirically divide them into already organised parts.

The Kantian position on this issue of organisation is a major piece in his debate with Leibniz. The general leibnizian mistake consists in treating the phenomena as mere objects of our understanding, thus forgetting that as phenomena they must be objects of a possible experience, and hence fall under the conditions of time and space. By misunderstanding that organized beings are appearances as far as they are composed of spatial parts, he forgot that they were hanging on the structure of space. The opposition of continuous and discrete prevents one to consider as a continuous quantum an articulated whole; but speaking of infinite divisibility entails continuity within matter, so the idea of “infinite organisation” illegitimately conflates continuousness and discreteness. Surely, Leibniz thought that substances do logically precede space, which is nothing more that the “order of their co-existence”, hence his statements concern substances as such. However, what Kant has established, is that those statements get their meaning- since Leibniz did speak of infinite divisibility – only insofar as they are about an object in space and time (e.g. a phenomenon), else one could not even talk about such a divisibility. This was precisely the refutation of the Thesis, in the Antinomy, showing that, when you think matter as prior to forms of intuition, your statements are conceivable but not objectively meaningful.

In other words, when one deals with the inability of the Leibnizian thinking to conceive the original character of the organised beings, one is led, in the end, to the lack of a difference between substance as a mere conceived entity, and substance as we meet it in experience, namely as a manifold in time and space. And in this case such a manifold has to be thought under the regulative rule formulated in the second Antinomy; a consequence of this rule, is that this empirically accessible manifold is either infinitely dividable but not composed of an infinite number of parts (it is a continuous phenomenon), or articulated in a determinate manner (it is a discrete phenomenon), albeit infinitely dividable if considered only as a spatial entity (and not as an organised entity). In this second case, logical relationships between determinate parts (the wheels of a watch, the cells of a beaver) are superimposed to mereological relations of spatial parts. However, Leibniz did conflate those two dimensions, and conceived the logical determinate parts as result of the infinite division (rather than the mereological parts).

Briefly said, the Leibnizian doctrine does not teach us anything concerning the organisms as we meet them in the experience, because it forges only an empty concept of them. As soon as this concept is applied to experience, all the distinctions just indicated here have to be made, and the concept is no more valid. Leibniz’s concept of organisms tells what perhaps could be an organised being for the divine understanding, but not what is to be labelled organism by us.

Throughout this analysis, we see that the phenomenon of life indicates a problem which put into question the frames of our experience, namely space and time, since providing an exact account of organisms implies presuppositions concerning those dimensions. In the section that we commented, the Leibnizian conception of life appears as a test of his metaphysics of space and time. For this reason, the very idea of criticism, which implies an investigation of the nature of this framework of any experience, is crucially concerned by problems about how organisms are to be conceived. The first moment of the ordinary notion of life, as it is at stake here, is the organization, since our first experience of life confronts us to a whole whose parts are related in a constant - even heritable - way, allowing those parts to fulfil some tasks. Those parts thereby are called “organs”, Kunstteile, which meant instruments in Greek. And Kant’s rebuttal proves now that Leibniz did not give an account of even this first characteristic of living things, w.

But our result here is the following: to the extent that the problem of organisms is a test for the refutation of Leibnizism as well as the critical conceptions of space and time, the negative characterisation of the Leibnizian account of life is not enough, because one would expect something like a solution of the problem, namely, an account of the specific status of organised beings. This is not to be found in the first Critique. So, the idea of the third Critique, which will explicitly address this task, is explicitly prescribed in the agenda of the criticism since this early critique of Leibniz.

3. Life, organisms and criticism.

Kant’s paragraph on the Leibnizian concept of life then concludes: one can not understand a phenomenon by conceiving it as infinitely organised. This implies that we can not formulate an ontology which would divide the world between, on the one hand what we call machines – finitely organised – and on the other hand what we call organisms – infinitely organised. And this result holds for the first Critique itself. Truly, the concept of “organised being” in the Critique of pure reason is not capable of accounting for life, because it provides no criterion of the difference between machine and organisms – since the only available criterion provided by a view of life as organisation is the leibnizian one, which has been dismissed. All the organised beings are articulated, hence finally organised in a determinate, finite manner, so in this regard technical machines and organisms are homogenous. In the Dreams of a spiritseers, Kant stated the idea that Stahl, opposing mechanisms and organisms, was somehow right against the more scientific conceptions by Haller or Boerhaave, committed to a kind of mechanisms (Ak.II, 331). But this sort of feeling was not able to be conceptually justified: in the first Critique, it’s still not the case.

Now, this idea of “organised beings” is pervasive in Kantian thinking. But - and this proves essential – when the §65 of the Critique of judgement is entitled “the natural purposes are the organised beings”, this supposes a shift within the very concept of “organised beings”, indicated by this sentence: “(a product of nature) as organised and self-organising may be called a natural purpose.” This means that, in the third Critique, the idea of organisation is divorced with the idea of articulation. However, this presupposes that previously a criterion has been given which allows us to distinguish between articulated machine and vital organisation – a criterion which can not be the infiniteness of organisation. Only then, will one be able to link “organisation” as articulation of organs and “organisation” as organism in the phrase “Organisierte Wesens”’. And this paragraph of the Critique of judgment, precisely, will provide such a criterion – which is the reciprocal causation of parts and parts, and wholes and parts[33].

Retrospectively considered, and situated within the macroevolutionary course of ideas, this evolution from the Second Antinomy to the third Critique concerning the issue of organisms, which proves essential for the testing of criticism (and more precisely, transcendental aesthetics), corresponds to the re-shaping of the life sciences, and particularly physiology and embryology, in the eighteenth century. Whereas Leibniz, albeit rejecting Cartesian mechanism, subscribed to a machine-view of organisms (even if it was only an epistemological stance), later scientists such as Haller in Germany, Bordeu and the vitalists in France (on the physiological side), or Caspar Wolff in embryology in Germany[34], vindicated a more “immanentistic” view of organism that sought its intelligibility in the identification of its own properties or forces and their expression (instead of considering the infinite extension of a machine-mechanistic model)[35]. To some extent, Kant’s critique of Leibniz in this short text, and evolution towards a concept of “organism as self-organising” fits quite well this evolution of concepts and models within life sciences[36]. The diagnostic on Leibniz’s conception of organisms picks out the negative moment of this conceptual history.

In such a way, Kant’s position in the debate matches Leibniz’s controversy against vitalism, and particularly Stahl’s vitalism. Whereas Leibniz emphasizes the preestablished harmony between efficient causes and final causes, hence body and souls, in order to explain away the stahlian postulation of a particular soul proper to organisms, Kant emphasizes the fact that no essential kind of matter – characterized by its infinite organisation – is to be found in organisms. That’s why he insists on the necessary presence of inorganic parts of organisms in possible experience (if not in current experience). But where Leibniz challenges animism in the name of natural science and universality of lawlikeness in nature, Kant fights another threat to natural science, the threat that matter would be by itself alive, and then contradicts the principle of inertia, which is a necessary principle of physics. He calls this hylozoism, and will try to eliminate its possibility from our concepts of nature in the Metaphysical foundations[37]. However, even if hylozoism is a conceptual possibility present in the various theories we have mentioned, that focused on the peculiar productivity of living beings, Leibniz himself paved the way to hylozoism, because if substance is a force, and the monads a representative power, then everything which exists in nature is somehow alive, as Leibniz himself had recognized[38]. But, since we have shown that those two conceptions, the internal unity of organisms as a soul, and the infinite organisation, are correlative in Leibniz’s conception, and since Kant has refuted the latter, then, from the critical point of view, the souls in substances are also refuted, and hylozoism is made impossible. According to the first critique, in nature (as appearance, of course) there is only inert matter, or some articulated arrangements of matter, nothing more. Leibniz’s monads, for Kant, are a firm ladder to hylozoism[39], and should therefore be dismissed.

Now, what is philosophically at stake here could be termed in one word: “technique”. Leibniz’s thesis, even if it is epistemologically understood, e.g. in the weak sense of a statement concerning how to investigate living things, contends that schemes provided by “technique” are the proper way of understanding the functioning of organisms. The Kantian rebuttal of this thesis means that technique is not sufficient to account for what is proper to organisms, even on an epistemological point of view. But in the text that we commented here, nothing is still proposed in order to fill this insufficiency. However, one of the main features of living beings left unexplained by technical models is the embryological phenomena, exemplified by the fact of generation. Nevertheless, from Buffon’s Histoire naturelle to Wolff (Theoria generationis, 1759) and Blumenbach (Uber Bildungstrieb, 1781), precisely at the times Kant is moving from pre-critical to critical philosophy, those features are beginning to become the constituted object of a new scientific field, descriptive embryology, which has now tools and argument to vindicate for a position, epigeneticism, previously remained confined in the philosophical and a priori debates on preformism versus epigeneticism[40]. This position, in fact, bears precisely some consequences that are urging for a positive conceptual theorizing of the epistemological resources that it requires, since they emphasize on the ability of organisms to build themselves, a capacity which is opaque to any technical-modelling account of organisms. And Kant however, from 1777 to 1789, in three opuscules (Von den Verschiedenen Menschenrassen, 1777, Uber die Bestimmung…, and Gebrauch der teleologischer Prinzipien, 1789) will continuously elaborate on those problems[41]. He will then write the third Critique in order to philosophically account for those (conceptual) works in natural sciences, and then he will be about to positively fill the blank left by the rebuttal of technical model instantiated in the critique of the Leibnizian doctrine in our Second antinomy.

From this perspective, the task of the third Critique concerning organisms is understandable. There is no distinction between articulated machines and organised beings, this is the result of the second antinomy; in those two cases, the articulation following an a priori concept superimposes on an extended appearance a new kind of divisibility, which is an a priori finite composition. So, when it comes to the organised beings, if it’s not the kind of composition which distinguishes machines from organised beings (infinite vs. finite: that was Leibniz’s thesis), then one should look to the status of this concept which determines the articulation, Elucidating the peculiar epistemological status of this concept, and its embodiment in matter, in the case of organised beings, will be exactly Kant’s task in the critique of teleology, because teleological judgement is precisely a judgement involving an a priori concept at the basis of an organisation.

Conclusion.

Finally, the Leibnizian doctrine of organisms – no matter how precious was the advances it made upon Cartesian mechanism – reflected the impossibility, within classical metaphysics, to make room for a concept of self-organisation. Kant, in this part of the Second Antinomy, thereby established the necessity of a critique of metaphysics in order to provide a status for vitality and life as such, epistemologically as well as ontologically. The failure of the Leibnizian doctrine of organisms reveals that there is no room for life in the classical metaphysics, which means that in the structure of knowledge that corresponded to such a metaphysics, namely a knowledge whose model is the mathesis universalis, and which is composed of a philosophia naturalis supposed accessible through mathematical methods exemplified by geometry, no life science as such can emerge. This explains that, historically, the grounding of a unified science of life in the beginnings of the XIX century, as it is usually recognized by various scholars[42], occurred simultaneously with the end of the classical metaphysics through Kantianism, and the giving up of an ideal of a mathesis universalis.

Within the structure of knowledge bound to classical metaphysics, the technical scheme, infinitely expanded by the Leibnizian conception, is in fact only able to grasp life through physical and mechanical perspectives, added with theological assumptions in order to elucidate the origins of those infinitely organised machines; but there is no proper epistemology for the understanding of organisms as immanently arranged. This is mainly due to the externality of the concept of technique (which entails an external arranger) – reflected in the concept of Gliederung here criticized by Kant, since in the Gliederung, the determining concept is external to the composite so that the immanent arrangement of matter in organisms is not accountable for. Of course, this is what is reached by the insights on the difficulties of any philosophical account of life sciences in his time that Kant provided in our paragraphs of the Transcendental dialectics, and it is only negative. No positive condition, positively accounting for what is in the same time emerging in the empirical investigation of life, is given by the philosopher until the Critique of judgement.

References

Erich Adickes, Kant als Naturforscher, Berlin, De Gruyter, 1923.

Sadik Al-Azm, The origin of Kant’s argument in the antinomies, Oxford, Clarendon Press, 1972

Karl Ameriks, “The critique of metaphysics : Kant and traditional ontology ”, Cambridge companion to Kant, P.Guyer ed., Cambridge University Press, 1992, 249-280

Giulio Barsanti, « La naissance de la biologie. Observations, théories, métaphysiques en France, 1740-1810 ”, Nature, Histoire, Société, R.Rey, C.Blanckaert, J.L.Fischer éd., Paris, Klincksieck, 1995, 196-228.

Jonathan Bennett, Kant’s dialectic, Cambridge University Press, 1974

Johann Blumenbach, Über der Bildungstrieb und das Zeugungsgeschäfte, Göttingen, 1781

Théophile de Bordeu, Recherches sur les glandes, Montpelllier, 1751.

Peter J. Bowler, “Preformation and pre-existence in the seventeenth century: a brief analysis”, Journal of the history of biology, 4, 2, 1971, 221-244.

Gerd Buchdahl, Metaphysics and the philosophy of science, MIT, 1969

Stephen Cross, “John Hunter, animal physiology and late 18th century physiological discourse ”, Studies in history of biology, 5, 1981, pp.1-110

Brigitte Falkenburg, Kants Kosmologie. Die wissenscheftliche Revolution der Naturphilosophie I, 18.Jahrhundert. Frankfurt, Klostermann,2000

Michel Foucault, Les mots et les choses, Paris, Gallimard, 1967

Michael Friedman, Kant and the exact sciences, Cambridge, Cambridge university Press, 1992.

Hannah Ginsborg, “Kant on understanding organisms as natural purposes”, Kant and the sciences, Eric Watkins (ed.), Oxford, Oxford University Press, 2001, 231-259

Michelle Grier, Kant’s doctrine of transcendental illusion, Cambridge, Cambridge University Press, 2001

Paul Guyer, “Organisms and the unity of science”, Kant and the sciences, Eric Watkins (ed.), Oxford, Oxford University Press, 2001, 259-282.

Ian Hacking, “Individual substance”, Leibniz. Critical essays, H. Frankfurt ed., Notre Dame Press, 1976, 137-154

Albrecht Von Haller, Prima lineae physiologae, Lausanne, 1747

Elementa Physiologae, Lausanne, 1755

Martin Heidegger, Kant und the Problem der Metaphysik, Freiburg,1929.

Philippe Huneman, Métaphysique et biologie, Lille, Presses du Septentrion, 2001

“Wolffian embryology and reflexive judgement : Kant’s shift between the first and the third Critique”, Naturalizing purpose, Collected essays on Kant and biology, Huneman P. (ed.), (forthcoming a)

Métaphysique et biologie, Kant et la constitution du concept d’organisme, Paris, Vrin (forthcoming b)

H. Ingensiep, “Die biologischen Analogien und die erkenntnistheoretischen Alternativen in Kants Kritik der reinen Vernunft B §27 ”, Kant-Studien, 85, 381-393

Norman Kemp-Smith, A commentary to Kant’s « Critique of pure reason”, London, 1930

Philip Kitcher, “Projecting the order of nature”, Kant’s philosophy of physical science, R.E.Butts (ed.), Dordrecht, Reidel, 1986, 201-235.

Hans-Ludwig Koch, Materie und Organismus bei Leibniz, Halle, 1908

Gérard Lebrun, Kant et la fin de la métaphysique, Paris, 1970

Timothy Lenoir, “ Kant, Blumenbach and vital materialism in German biology ”, Isis, 71, 1980, 77-108

Béatrice Longuenesse, Kant et le pouvoir de juger, Paris, 1993.

Wolfgang Malzkorn, Kants Kosmologie-Kritik, Berlin, De Gruyter, 1999

Anselm Model, Metaphysik und reflektierende Urteilskraft, Untersuchungen zur Transformierung des leibnizschen Mondenbegriffs in der KU, Francfort, Athenaüm, 987.

Michel Puech, Kant et la causalité, Paris, Vrin, 1990

Hans-Peter Reill, “Analogy, comparison and active living forces : late enlightenment responses to the critiques of causal analysis”, The sceptical tradition around 1800, K. van der Zande, R. Popkins (eds.), Dordrecht, Kluwer, 1998 : 203-211

“”Vitalizing nature and Naturalizing the humanities in the Late Eighteenth Century”, Studies in Eighteenth Century Culture, 28, J.C. Hayes, T. Erwin (Eds.), Baltimore, John Hopkins University Press, 1999, 361-381.

Robert J. Richards, “Kant and Blumenbach on the Bildungstrieb: a historical misunderstanding”, Studies in History and philosophy of biology and biomedical science, 31, 1, 2000, 11-32

Shirley Roe, Matter, life and generation. Eighteenth century embryolgy and the Haller-Wolff debate, Cambridge UP, 1980.

Jacques Roger, Les sciences de la vie dans la pensée française au XVIIIème siècle, Paris, (Colin 1963), Albin Michel 1993

Phillip Sloan, “Preforming the categories: Eighteenth-Century Generation Theory and the Biological Roots of Kant’s A Priori ”, Journal of the history of philosophy, 40, 2, 2002, 229-253

“Buffon, German biology and the historical interpretation of biological species”. British journal for the history of science, 12, 41, 1979, pp.109-153

P.F. Strawson, The bounds of sense. An essay on Kant’s Critique of pure reason, Londres, Routledge, 1995

Jules Vuillemin, Physique et métaphysique kantiennes, Paris, 1987²

Nécessité et contingence, Paris, 1980

W.H.Walsh, Kant’s criticism of metaphysics, Edimburg UP, 1975.

Kenneth Westphal, Kant’s transcendental proof of realism, Oxford University Press, 2005

Victoria Wike, Kant’s antinomies of reason, Their origin and their resolution, Washington, University press of America, 1982

Caspar Friedrich Wolff , Theorie von der Generation in zwei Azbahndlungen erklärt und bewiesen, Berlin, 1764 (repr. Hildesheim, Olms, 1966)

Theoria generationis Halle, 1759

John Zammito, “”This inscrutable principle of an original organization” : epigenesis and “looseness of fit” in Kant’s philosophy of science”, Studies in history and philosophy of science, 34, 2003, 73-109.

The Genesis of Kant’s Critique of judgement, Chicago University Press, Chicago, 1993

Clark Zumbach, The transcendant science. Kant’s conception of biological methodology, La Haye, 1984

Translations of the Critique of pure reason are from the Cambridge University Press edition, tr. Guyer and Wood.

Translation of Leibniz’s texts are mine.

The author warmly acknowledges a referee for his precious comments and suggestions.

-----------------------

[1] In Stahl, Opera omnia, vol.III, « Negotium Otiosum », Leibniz’s remarks, §3

[2] Tentemen anagogicum*

[3] Système nouveau de la nature et de la communication des substances, §10, ed. Erdmann.

[4] « Since this portion of reason, that we possess, is a gift of God, and consists in the natural light that remained withn the whole corruption ; this portion conforms to the whole, an differs from which is in God only as a drop of water differs from the ocean, or rather, as finite differs from the infinite.” (“Discourse on the conformity between faith and reason”, §61)

[5] Traité de l’homme, AT, XI, 202.

[6] « The nature of an individual substance or of a complete being consists in having such an achieved notion that it is sufficient to understand and to have deduced all the predicates of the subject to which this notion is ascribed. » (Discours de métaphysique, §8)

[7] “The very nature of the body does not consist only in extension, namely in quantity, figure and motion, but one must necessarily admit something that has a relation to souls, and that is commonly called substantial form. » (Discours de métaphysique, §12)

[8] Système nouveau de la nature et de la grâce, p.71*

[9] Concerning the relationship between soul and organised parts see H.L. Koch (1908), §13, « Leibnizens Auffassung des Organismus »

[10] On the internal unity that is essential to substance as such and allows Leibniz to dismiss any conception of things as bundles of properties, see Hacking, 1976, pp.146-148

[11]Monadologie, §79.

[12] GP II, 520. The concept of substantiate will receive further elaboration in the late letters to Des Bosses, nevertheless this work has not been widely known at the times of Kant.

[13] « I find that even several effects of nature can be demonstrated in two ways, namely by the consideration of the efficient cause, and by the consideration of the final cause, for example by using God’s decree to always produce its effect through the easiest and the most determined ways, as I have shown elsewhere by accounting for the rules of dioptrics and catoptricts (..) It is wise to notice this in order to conciliate those who expect to explain mechanically the formation of the first texture of an animal, and all the machinery of the parts, with those who account for this same structure through the final causes. The one and the other as good, the one and the other are useful not only to admire the art of the great worker, but also t discover something useful in physics and in medicine.” (§§21-22)

[14] « The transcendental reality (subjective) of the concepts of the pure reason is at least grounded on the fact that we are led to such ideas by a necessary syllogism of reason » (A339, my emphasis).

[15] Concerning the antinomies of reason, Victoria Wike’s book treats the topic extensively and gives a clear account of the differences between those antinomies as well as between antinomies in the first and the third Critique. Malzkorn (1999) provides an analysis of the antinomies through a formal reconstruction of Kant’s arguments, and then a critique of them; Falkenburg (1995) uses the formal apparatus but embeds Kant’s arguments within a reconstruction of the cosmological stance of its times.

[16] Which are, concerning Quantity : integrity of the collection of all phenomena (that is the question of the beginning and the limit of the world); Quality : absolute integrity of dividing a given whole (that is the question of the simple entities); Relation : absolute integrity of the origin of a phenomenon (that is the question of the spontaneity, of the free act as an Unconditioned beginning of a new series of causes); Modality: absolute integrity of “the dependence of the existence of what changes within the phenomenon” (that is the question of a necessary being) (A415).

[17]For a general commentary one can usefully refer to Kemp-Smith (1948), 480-525. He concludes on a critical evaluation of the Antinomies which rightly insists on the fact that this chapter can not be self-sufficient, since it is a natural invitation to the third part of the Dialectics, devoted to the Idea of God (“Ideal of pure reason”), hence to rational theology. Its ambiguities are mostly (and tentatively?) resolved in this chapter. A historical survey is available in S.Al Azm (1972), which vindicates the radical thesis that the antinomies are the philosophical formulation, in a critical framework, of the Leibniz-Clarke debates concerning the nature of matter and time – debate which, in the end, opposed Leibnizism and Newtonism (see pp.46-85 on the second Antinomy). However, one can not ignore that this rivalry between Newtonism and Leibnizism concerning the correct philosophical account of the new physical sciences is the main debate that the critical project was supposed to overcome. This illuminates the strategic situation of the antinomies within the critical system. Puech (1990) argues convincingly that the project of the first Critique was a philosophical conciliation between the Leibnizian framework of the Schuhlmetaphysik, and the Newtonian advancements in the natural sciences, which yielded new requirements concerning ontology. Grier (2001) gives a clear account of what is an illusion in the Antinomies.

Among the secondary literature, Kemp-Smith gives a straightforward commentary, while Bennett (1975) tries to find out and eventually improve the Kantian arguments; one of the main question he addresses, also raised by Walsh (1974), is the meaning and the validity of Kant’s reconstruction of the antithesis. The point is that to evaluate Kant's solution of the antinomies, one has to assume that the problem is well formulated in this two-sides debates, both historically and conceptually, which is not self-evident. But we can here leave aside those questions concerning the general significance of the antinomies.

[18] This alternative opposes, if not precisely Leibniz and Spinoza, but at least the “dogmatic“ metaphysics, and a kind of sceptical empiricism. In fact, while the Leibnizian character of the “dogmatic” side is easy to identify, the other side of the alternative has no definite name as a philosophical position, but draws upon elements from Spinozists metaphysics, Newtonian philosophy of nature and epistemology of the natural sciences as previously vindicated by Gassendi, by Hobbes or by Boyle. Even Leibniz’s conception of infinite divisibility of space is requested in the antithesis of the Second antiomy. Anyway, Grier (2001) is right to emphasize the fact that, especially concerning the second antinomy, since all those positions share certain presuppositions concerning transcendental realism, their repartition is not likely to match exactly the lines that Kant draws in his text.

[19] By emphasizing the fact that the series is only given in the act of regress we are subscribing to what Strawson named “mix interpretation” of the transcendental idealism in the antinomies, and that he opposed to the so-called “strong interpretation” (e.g. nothing exists in space and time) and to the so-called “weak interpretation” (e.g. things do exist in space and time, but some statements such as the sentences in the Antithetic violate the principle of meaning) (Strawson (1995), 193). According to this mix interpretation, the existing series are “the series of successive perceptions that correspond to successive advances of our empirical investigations”. Thus, this interpretation provides “a foundation or a metaphysical support” for the principle of meaning, according to which a proposition gets a meaning only as related to the conditions of the empirical application of its concepts. This ground is “the metaphysical fact that, after all, nothing exists really but our representations and experiences” (Ibid, 195) However, I don’t entirely agree with Strawson, because it seems that we can not in the same time ascribe reality to the succession of perceptions and refuse it to the series of objects. More exactly, since the series of objects is constructed within the series of perceptions – and this is entailed by the fact that the series does not exist outside the regress – then, reciprocally, the solution of the antinomies can not rely ultimately on a “metaphysical fact” which would be a kind of idealism. The series of objects does exist, and exists no less than the succession of perceptions: both exist through and as the very act of regress. So, the mix interpretation according to Strawson forgets the status of the regress as an act, status which yields the whole Kantian solution. Here, the reading of Kant that Vuillemin suggests in Nécessité et contingence, reading which endorses Kantian intuitionism, seems to give a better account of the Kantian solution to the antinomies, by reminding that the intuitionist emphasis on the operations of mind may avoid the universal validity of the law of the excluded third-term.

[20] Bennett’s useful commentary here says that a regulative principle does not transcend the experience, because it extends it in a “quantitative” manner, but not in a “qualitative” way (Kant’s dialectics, §86, 272)

[21] However, it is difficult to know what exactly would mean « to apply a regulative principle in a constitutive way». We follow here Bennett, saying that the mistake here is to believe that the principle is justified by a state of facts, instead of being a kind of advice that one gives to himself (Bennett (1974), §88, 278)

[22] Al Azm (1972, 46) emphazises thus that Kant is concerned here by any kind of substance, be it material or psychical; hence the issue of the last elements of the decomposition of substance concerns atomism as the Leibnizian doctrine of monads. Anyway, since between the antithesis and the thesis the relation of matter to time is at stake, I follow Grier (2001) who argues that this basically is true about the thesis, but the antithesis is committed to a view of matter realised in space.

[23] A formal reconstruction of the antinomy is provided by Malzkorn (1999), pp.168-190. He then criticizes Kant’s solution (270-287)

[24] “Body is therefore infinitely divisible, without consisting, however, of infinitely many parts” (Kemp Smith) Kant wrote : “Dieser ist also ins Unendliche teilbar, ohne doch darum aus unendlich viel Teilen zu bestehen.” The German contrasts divisible, Teilbar, which emphasizes the potentiality of dividing, with Teile, the parts themselves that would constitute the thing independently of any disposition of the matter, disposition correlative to our disposition to divide. The darum means that there could be a deduction, or our tendency to deduce – deduction which in fact is misleading – a deduction from, precisely the divisibility, which is about the appearance as relative to our power of knowledge, to the Teile, the parts, considered in themselves. Assuming this deduction would mean conflating what we can say about appearances, and speculations about things in themselves. The Antithesis makes this deduction, and then, is on a par with the Thesis, notwithstanding the fact that it seems less metaphysical, and more respectful of the true structure of space (namely, its divisibility).

[25] The sentence refers to the proof of the Thesis : “denn man kann allenfalls wohl zugeben: dass die Dekomposition im letzteren niemals alle Zusammensetzung wegschaffen Könne, indem alsdann sogar aller Raum, der sonst nichts Selbständiges hat, aufhören würde (welches unmögliches ist)“. Guyer and Wood translate : „for one can in any case concede that the decomposition of the latter could never do away with all composition, since then, every space, having nothing else that is elf-subsistent, would cease to be (which is impossible)“. My point is only to cancel the immediate complementarity between “composition” and “decomposition” and the subsequent assimilation between constitution and composition, because this would alter Kant’s text.

[26] Arguments of the theses do not begin with the empirical concept of matter, but the pure concept of “composite” (207)

[27] This is clear also in the lectures on Metaphysics : « to speak of simple beings we must go beyond the world of the sense, but then we have no proof of the objective reality of our concept, for we can give no example.” (Ak. XXIX, 828)

[28] Système nouveau de la nature et de la grâce, p.71*

[29]Falkenburg 2000, 237

[30] The word Kunst means, as we know, art, technique, artificial. In the Einzige Beweissgrund (1763), Kant already opposed an order immanently stemming from laws of nature, and an order originating in a divine arrangement of entities with their heterogeneous laws. The latter was called “technical” order, Kunst, in opposition with the former, a necessary order of the kind of a system. The words Kunstteile here reminds this theory ; but in the Critique of pure reason, in a way, the finality as technique is rejected, as if the only legitimate sense of order was the systemic one (on this issue, see Huneman (forthcoming a)**).

[31] In the contemporary biology, we have an example of this schema, with the decomposition of an animal into organs, tissues, cells, or of the chromosomes into genes, nucleic acids, nucleotides, etc. The last levels have been empirically discovered not so long ago

[32] In my contemporary example, it is the nucleotides; the Kantian argument says that we know empirically that the unorganised parts are the nucleotides but we know a priori that there will be unorganised parts

[33] On this topic see Zumbach (1984), Zammito (2003), Ginsborg (2001), Huneman (forthcoming a).

[34] Haller (1746, 1755), Bordeu (1751), Wolff (1759, 1764), Blumenbach (1781)

[35] In several papers (1999, 1998, etc.), Peter Hans Reill has labeled those tendencies “Enlightenment vitalism”, and highlighted one of their origin in Buffon’s methodological and historical thinking; this Enlightenment vitalism for him culminated in Herder. Hence the controversy between Kant and Herder, that Zammito (2001) analyzed in details, makes salient Kant’s necessity to criticized this trend of thought, by justifying some elements and dismissing some others (e.g. hylozoism, see below)

[36] A wide demonstration of this is given in Huneman (2001, forthcoming b)

[37] For the importance of eliminating hylozoism, see Zammito, (1993), 189-213, from the perspective of the teleological judgments and the doctrine of organisms, and Westphal (2005), 164-166, 208-221.

[38] Cf. the above quotation : “I restraint corporal or composed substance to the mere livings or organic machines in nature; what’s left is for me only aggregates of substance”

[39] The other way of refuting monads is precisely the one which will be used by the third Critique, namely, the refutation of hylozoism as a kind of spontaneous generation, and the claim of epistemological discontinuity between organism and inorganic nature. Kant’s argument there relies on the indeducibility of original organisation of organisms from mechanical configurations of matter, and this assumption is also attributed by him to Blumenbach. The argument is made in the lectures of metaphysics : “Since we have no other concept of the interior of other things than what proceeds in ourselves, which are representations and what follows from them, so (Leibniz) concluded from this that all monads would have representations (the actuality of something is not also to be assumed when it is possible), and called them powers which represent the universe or living mirrors of the universe. For if all monads were in the world, one would influence the other, but since they have nothing but mere representations, each has representations of all monads in the world. But one had to assume slumbering monads (monads sopita) which, to be sure, have representations but are not conscious of them. According to him these constitute the class of non-rational animals. But there were various degrees of the consciousness of the representations – distinct (distincte) – clear (clare) – obscure (obscure). The monads went from one state to another, from the distinct t the more distinct, until God. This is the so-called continuum of forms (continuum formarum), according to the analogy of the physical continuum (continui physici), where the minerals commence the order, through the mosses, lichens, plants, zoophytes through the animal kingdom until human being. This is nothing more than a dream whose groundlessness Blumenbach has shown.” (Ak. 28.762)

[40] On those questions, the context and the preformationism debate along the century, see Roger (1963), Roe (1980), Reill (1998, 1999), Bowler (1971). On Kant’s position see Sloan (2002); Ingensiep (1990); for a contrasted account of Kant’s epigeneticism, Zammito (2003), Huneman (forth. a).

[41] On Kant’s theories of heredity and generation, among numerous valuable contributions, see Lenoir (1980); Sloan (1989); Richards (2000).

[42] Barsanti (1995) ; Foucault (1967) ; Hoffmeier (*) challenging Foucault recognize the fact but disagrees on its radically discontinuous character ; Roger (1963) ; Cross (1981)

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download