20th WCP: Leibniz's Distinction Between Natural and Artificial Machines

Introduction

Leibniz often distinguishes between organic machines of nature and the machines that we construct. This distinction might not seem to have been an original contribution on his part. Similar distinctions were drawn by many early modern philosophers, particularly Cartesians, who contrasted our machines with the much more complex (yet mechanical) products of the divine artifice. Leibniz's distinction was not this simple. For him, the difference between our machines and organic machines of nature was not simply a difference in degree: it was not simply a matter of God's machines being more structurally complex than the mechanisms that we produce. More generally, Leibniz's distinction between organic (i.e., natural) and human-made (i.e., artificial) machines cannot be understood as long as we confine our gaze to the realm of mechanical phenomena that are described by physics, for it is a deeper metaphysical distinction rooted in his views about substances.

Leibniz does occasionally draw the distinction in terms of structural complexity, claiming that natural machines, since they were built by God, are infinitely more complex than the machines that we make. This might appear to undermine my claim that Leibniz's distinction cannot (unlike similar distinctions drawn by his contemporaries) be understood simply in terms of varying degrees of structural complexity. However, I shall contend that his formulation of the distinction in terms of structural complexity presupposes a more basic difference between natural and artificial machines, a difference that can only be adequately characterized within his metaphysics.

Structural Complexity Accounts of the Distinction

Let us first consider a passage in which Leibniz formulates hisdistinction in terms of structural complexity. In section 64 of the Monadology, he claims that an organic machine is a machine in its "least parts, to infinity," while an artificial machine (such as a clock) is not. However, some of Leibniz's other claims entail that a clock is a machine in its smaller parts down to infinity. This is because Leibniz regards artificial machines as aggregates of corporeal substances. Moreover, on his view, corporeal substances are — or at least involve — organic machines of nature. Thus each clock includes the infinitely many organic machines that figure in the corporeal substances of which it is an aggregate. That is, a clock is on a par with the fish pond that Leibniz mentions in sections 67-68 of the Monadology, for it too is an aggregate of organic machines which are themselves (according to Monadology, 64) machines in their smaller parts, down to infinity.

Perhaps the difference between organic and artificial machines is given earlier in section 64 of the Monadology, where Leibniz claims that not only is each organic machine a machine in its "least" parts, but that it is also a machine in "each" of its parts. That is to say, it is susceptible of being pulverized infinitely in such a way that every part that results from this decomposition (not just the smaller parts) is a machine. However, an artificial machine, being a mere aggregate of organic machines, shares this feature; for any mere aggregate of F's can be broken down without remainder into F's. Leibniz was, of course, aware of this consequence of his views, as is evident from his claim that, "An organic body, or any other body whatsoever, can again be resolved into substances endowed with organic bodies," (italics mine). In other words, all bodies, including artificial machines, can be divided into organic machines (i.e. organic bodies), which can in turn be so divided, and so on, to infinity.

In the light of this we should interpret Leibniz as adhering to a stronger version of the claim that natural machines (unlike their artificial counterparts) remain machines in "each" of their parts. Instead of rendering this as the claim that only organic machines can be divided infinitely in such a way that all their parts are machines, we should ascribe to him the stronger view that every way of breaking up a natural machine yields only such parts; whereas, by contrast, there will always be some way of disassembling an artificial machine that yields a part (P) that is not itself a machine but is instead a mere aggregate of machines of nature. Part P, even though it contains infinitely many organic machines, is merely an arbitrary assemblage or aggregate thereof, for the natural machines that it contains are not organized in such a way as to form a whole that exhibits the design and apparent teleology that are the characteristic mark of a machine. Thus, P figures in the economy of the machine to which it belongs merely as an aggregate hunk of secondary matter that bears certain causal-mechanical relations to the machine's other parts.

While certainly an improvement on the weaker interpretation of the view that only a natural machine is a machine in each of its parts, this way of understanding Leibniz's distinction between natural and artificial machines is not wholly satisfactory. For while it may be true that most artificial machines can, as a matter of fact, be taken apart in such a way that not all of the resulting pieces are themselves machines, there is no reason to believe that this must be true of every such device. To see why, let us consider a case that might not appear, at first, to involve an artificial machine, in which we graft a branch from one tree onto a tree of a different sort. There is no way of breaking up the resulting tree that yields a mere aggregate hunk of secondary matter; regardless of how we break it up, every part will itself be an organic machine. This does not seem surprising. We are after all talking about a tree, which seems a clear case of what Leibniz would classify as a machine of nature. Note, however, that it is not, strictly speaking, a single organic machine. It is, rather, a composite of two natural machines, namely, the tree from which we took the graft and the tree onto which we grafted it. One might wonder why this should be thought to pose a problem, for in grafting the branch onto the tree we are carefully splicing together two undeniably organic machines in order to produce a new tree. Shouldn't such a carefully crafted aggregate of organic machines itself be an organic machine? The answer cannot be as simple as it might seem. This is because, on Leibniz's view, to raise this question is simply to ask whether artificial machines are not in fact machines of nature; for our splicing together of the two trees illustrates what, for him, our artifice (as opposed to God's) always amounts to: it is always a matter of fitting together several organic machines into a larger whole that is itself sufficiently well organized to qualify as a machine. Leibniz's account thus requires that when we graft the branch onto the tree we thereby fashion a new artificial machine, not an organic one. This new artificial machine is such that every way of breaking it up yields only parts that are themselves machines.

A Metaphysical Account of the Distinction

One might object that the grafted tree is not an artificial machine, on the grounds that it, unlike artificial machines, is the same type of machine (viz., a tree) in each of its parts. According to this objection, Leibniz distinguished natural from artificial machines as follows: a natural machine's parts are, in microcosm, the same kind of machine as the larger machine to which they belong; while a machine that we produce is of a sort that is quite different from the types of machine that serve as its pieces. The grafted tree, since it does not exhibit this derangement of part and whole, is not an artificial machine.

This interpretation does not do justice to what Leibniz actually meant. What I have in mind here is intimated only rather obliquely in section 64 of the Monadology, where Leibniz gives an example of something that fails to satisfy the requirements for being an organic machine. His example is the tooth of a brass wheel, whose "parts . . . no longer have any marks to indicate the machine for whose use the wheel was intended." This remark can readily be taken to mean only that each part, when considered on its own, gives no indication as to the type of machine to which it belongs. However, I believe that Leibniz had something stronger in mind. His point is really that at some point in the decomposition of an artificial machine we reach a component that does not register the identity of the larger machine to which it belongs. By contrast, the parts of an organic machine do somehow register (or "indicate") the identity of the larger organism of which they are parts. How do they accomplish this feat? Apparently, they do so simply by virtue of the fact that the larger organic machine is in each of them. Thus, an organic machine is not only the same type of machine but also the same token in each of its parts.

This interpretation is supported by what Leibniz actually says. For instance, he claims that, "A natural machine still remains a machine in its least parts, and moreover, it always remains the same machine . . ." He seems to have meant this in a very strong sense, according to which each of an organic machine's parts is somehow it (that same organic machine) in microcosm. That he did so is attested by his case for the imperishability of natural machines: since the very same individual natural machine survives in each of its infinitely minute parts, it can never be completely destroyed (except by God). Leibniz's preformationism shares a similar grounding: the existence of a natural machine in the microscopic pieces from which it is formed antedates (what seems to us to be) its conception. In Leibniz's words, the organism "already exists in miniature in the seeds before conception"; that is, "The animal itself was there."

It should now be clear why the grafted tree is an artificial machine: it is not the same individual organism in each of its parts but is, rather, two different natural machines, for it is made of pieces that belonged to two different trees. Thus, even though its parts are all of the same machine type, they are not all the same machine token in microcosm. It should also now be clear, for the same reason, why the grafted tree is not a counterexample to Leibniz's distinction between natural and artificial machines, when that distinction is properly interpreted.

We are left with the puzzling question as to how an organism can be "the same" in each of its parts. Clearly its extended matter cannot all be present in each of its pieces. Note, though, that to interpret Leibniz in this way is to assume that when he refers to an organic machine he thinks of it simply in terms of its extended matter. But this is not how he conceived of natural machines. Instead, he regarded natural machines as aspects of individual unities, that is, of corporeal substances. As such, each natural machine comes equipped with a substantial form (i.e., an "entelechy"). That Leibniz conceived natural machines in these more metaphysical terms is evident from the following summary that he presented to de Volder:

Therefore I distinguish: (1) the primitive entelechy or soul; (2) the matter, namely, the primary matter or primitive passive power; (3) the monad made up of these two things; (4) the mass [massa] or secondary matter, or the organic machine in which innumerable subordinate monads come together; and (5) the animal, that is, the corporeal substance, which the dominating monad makes into one machine.

The fourth clause in this passage introduces the term "organic machine" before the introduction (in the fifth clause) of corporeal substances. This may seem to suggest that organic machines are logically antecedent to these substances. Note, however, that, according to the fifth clause, we cannot speak of there being one organic machine until it is constituted as such by the dominating monad of a corporeal substance. Thus, Leibniz treated natural machines as belonging to corporeal substances, each of which has its own substantial form.

These elements of Leibniz's metaphysics are essential to an adequate understanding of his distinction between natural and artificial machines, for they furnish the only plausible interpretation of his claim that an organic machine (unlike an artificial machine) is the same machine in each of its parts, or that it survives in miniature in each of its minute pieces. Such claims can only make sense insofar as the organic machine is taken to belong to a corporeal substance, all of whose parts are animated by the same dominating monad. The machine is thus "the same" in each of its parts in the sense that those parts share the same substantial form (viz., the one belonging to the dominating monad).

On this interpretation, Leibniz's distinction between natural and artificial machines is thoroughly metaphysical; it cannot be drawn without the metaphysical backdrop of corporeal substances and their attendant substantial forms. Admittedly, in some passages (e.g., section 64 of the Monadology) Leibniz does present the distinction in terms of differing orders of structural complexity that are wholly discernible within the realm of mechanical phenomena. However, as I have argued, the distinction is not simply a matter of God's machines being infinitely more complex than the machines that we produce, for it has its roots in Leibniz's deeper views concerning what it means to be an individual corporeal substance.

A Connection to Leibniz's Views about Individuals

The connection to individuality here is important. Indeed, the upshot of Leibniz's distinction seems to be this: only a natural machine possesses the requisite metaphysical underpinnings for being a true unity (an unum per se), where these underpinnings involve the possession of a substantial form. In Leibniz's words,

By means of the soul or form there is a true unity corresponding to what is called the self [moy] in us. Such a unity could not occur in the machines made by a craftsman or in a simple mass of matter, however organized it may be; such a mass can only be considered as an army or a herd, or a pond full of fish, or like a watch composed of springs and wheels.

Here, it is clear that, in Leibniz's view, artificial machines are not true unities but rather mere aggregates; that is, each such machine is a mere unum per accidens and not an unum per se. We are thus brought closer to one of the guiding intuitions of Leibniz's philosophy, namely, his conviction that an adequate metaphysics must make some provision for objectively real individuals whose status as individuals is not subject to our whims and conventions. This conviction finds expression in his system as the claim that any extended thing that is a real individual (an unum per se) is a corporeal substance that derives its unity and identity conditions from its substantial form; it is the extended thing's form that is in each of its parts and that makes it what it is. In the absence of such a form, an extended thing is merely an aggregate that we, relative to our interests and conventions, count as an accidental unity (an unum per accidens). Such aggregates are not `really real' individuals. Indeed, Leibniz lumps all such aggregates together with "moral beings" (what we might now call social institutions), thereby downgrading their ontological status to that of merely fictional individuals that are useful in the light of our interests. As he wrote,

"There is as much difference between a substance and [an aggregate such as a flock of sheep or a block of marble] as there is between a man and a community, such as a people, an army, a society, or a college; these are moral beings, beings in which there is something imaginary and dependent on the fabrication [fiction] of our mind."

By contrast, organic machines are true unities, for they are corporeal substances. This was reiterated by Leibniz late in his life, in a letter to Des Bosses, in which he wrote,

But I thus restrict corporeal or composite substance to living things alone, that is, to organic machines of nature alone. Other things are, for me, mere aggregates of substances . . . ; aggregates really constitute something one only per accidens.

A Concluding Polemical Note

This passage is particularly revealing, since Des Bosses was one of the few correspondents to whom Leibniz expressed his views with the greatest of candour. One would think, then, that Leibniz's abiding commitment to the reality of corporeal substances is born out by his repeated references to them in his correspondence with Des Bosses. It is thus hard to see why some have maintained that Leibniz abandoned his commitment to corporeal substances in his later years. Donald Rutherford, for instance, discounts Leibniz's many claims about corporeal substances in the letters to Des Bosses, on the grounds that Leibniz was there merely suggesting ways of analysing Des Bosses's Roman Catholic notion of transubstantiation; as a Lutheran, Leibniz did not himself accept that doctrine and thus had no need to resort to the notion of corporeal substances (and the substantial bonds from which they might be forged) in order to accommodate it in his own philosophy. But this explanation by Rutherford does not account for Leibniz's reference to corporeal substances in the above quotation; for Leibniz there discusses the notion of corporeal substance not just in connection with transubstantiation but also with his distinction between natural and artificial machines. It is hard to see why he would thus have explicitly associated the notion of corporeal substance with such a long-standing tenet of his philosophy (which he clearly did not abandon later in life) if he did not really believe in such substances.