Four approaches to teleology

In his fine book Aristotle, Christopher Shields usefully distinguishes Aristotle’s approach to teleology from two others. Teleological eliminativism is Shields’ label for the view that there is no genuine final causality to be found in the natural order; the atomists Democritus and Leucippus would be two ancient advocates of this view. Teleological intentionalism is his label for the view that there is teleology to be found in the natural order, but only insofar as a divine intelligence has put it there; Anaxagoras would be an ancient representative of this view. Aristotle, as Shields notes, takes an intermediate position: there are final causes inherent in the natural order, but they do not require explanation in terms of a divine ordering intelligence. They’re just there in the nature of things. (This is so, in Aristotle’s view, even though we must affirm the existence of an Unmoved Mover who moves the world as its final cause; for what leads us to the Unmoved Mover is the need to explain motion or change, not the need to explain the existence of final causality per se.)

Andre Ariew, in his article “Teleology” in The Cambridge Companion to the Philosophy of Biology, draws a similar set of distinctions. The materialist, who corresponds to Shields’ teleological eliminativist, denies teleology. The Platonic teleologist, like the teleological intentionalist, affirms teleology but regards it as imposed by a divine intelligence from outside (e.g. by the demiurge of Plato’s Timaeus). The Aristotelian teleologist also affirms teleology but regards it as immanent to the natural order rather than imposed from outside. As Ariew notes, William Paley of “design argument” fame and contemporary Intelligent Design theorists are essentially Platonic teleologists.

These distinctions are very helpful, but they do not exhaust all the possibilities. For as I have noted, Aquinas rejects both Aristotle’s position and that of the Platonic or intentionalist teleologist. His position might be seen as a middle position between theirs. Like Aristotle, and unlike Paley and ID theorists, Aquinas regards final causality as immanent to the natural order. For Paley and ID theory, it is at least possible that natural objects have no end, goal, or purpose; they just think this is improbable. The reason is that they accept an essentially mechanistic conception of nature, viz. one which denies Aristotelian formal and final causes and models the world on the analogy of a machine. The bits of metal that make up a watch have no inherent tendency toward functioning as a timepiece; it is at least theoretically possible, even if improbable, that a watch-like arrangement might come about by chance. And natural objects are like this too: There is nothing inherent in any natural object or system – no essences, natures, substantial forms or anything else corresponding to such Aristotelian-Scholastic categories – by which we might read off final cause or teleology. The world might be like a collection of bits of metal that have by sheer accident come together in the form of something resembling a watch. It’s just that this is so highly improbable – so the argument goes – that the “best explanation” is that some intelligence arranged the bits that make up the world into their present purposive configuration, much as a watchmaker arranges bits of otherwise purposeless bits of metal into a watch.

Like Aristotle, Aquinas will have nothing to do with this picture of nature. For them, the world is not comparable to a machine – that is to say, it is not a complex arrangement of parts which have no intrinsic tendency toward the ends they actually happen to serve, and thus must be forced to do so “from outside” as it were. Rather, all natural substances have essences, natures, or substantial forms, and their final causes are therefore inherent or built-in. For Paley, it at least makes sense to think that the eye (say) might not actually be for seeing. It’s theoretically possible that the fact that eyes tend to result in seeing is an amazing accident. It’s just that, when we weigh the various explanatory alternatives, we find this one so highly improbable that we can rule it out. But for Aquinas, probability has nothing to do with it. The very idea that eyes might not really be for seeing makes no sense and is not even theoretically possible. If eyes are typically associated with seeing, then that can only be because it is in their nature to see. And that in turns entails that their natural end or final cause is seeing. We can know this just by considering eyes themselves, without getting into the question of their origin, divine or otherwise.

Unlike Aristotle, who leaves it at that, Aquinas thinks that the existence of final causes nevertheless requires an explanation, for the reasons sketched in my previous post on teleology; and he also thinks that this explanation must lie – not might lie, not probably lies, but necessarily lies – in the existence of a divine intellect which conserves the order of final causes in being from instant to instant. Here too, Aquinas’s God is not a “watchmaker” god who might have died off in the time that has passed since he finished his work, leaving his finely crafted timepiece to carry on without him. Without the divine ordering intellect, the order of final causes could not be sustained even for an instant.

As usual, the details can be found in The Last Superstition and Aquinas. The point is just that Aquinas’s position cannot be assimilated to that of either Aristotle or Paley. It is, again, a middle ground between them: call it Thomistic teleologism or Scholastic teleologism.

As the latter suggested label perhaps indicates, the situation here might usefully be compared to the debate over the problem of universals. Nominalism and conceptualism are essentially anti-realist positions, regarding universals as artifacts of language or the workmanship of the human understanding rather than having any objective basis. Platonic realism takes universals to exist entirely independently of either the natural world or of any mind. Aristotelian realism takes universals to exist only in the particular things that instantiate them and in intellects which abstract them from these particulars. Scholastic realism – the position of Augustine and Aquinas – takes what is in effect a middle ground position between Platonic and Aristotelian realism. Like the Aristotelian realist, the Scholastic realist affirms that universals can exist only in either their concrete instantiations or in an intellect. But like Plato, he also affirms that they nevertheless have a kind of existence beyond those instantiations and beyond finite, human intellects. For universals pre-exist both the material world and all finite intellects qua ideas in the infinite, divine intellect, as the patterns according to which God creates the world.

Nominalism and conceptualism, in their anti-realism, are comparable to teleological eliminativism, denying the objective existence of the phenomena in question. Aristotelian realism corresponds to Aristotelian teleologism: just as the former affirms universals but regards them as immanent, existing in their instantiations, so too does the latter affirm a kind of teleology, but only one existing immanent to the things that manifest it. Platonic realism corresponds roughly to teleological intentionalism: like Platonic realism, which regards universals as existing entirely separated from the material world, Platonic teleologism affirms teleology, but only as something which strictly speaking exists apart from the world rather than immanent to it. The patterns we observe in the natural world might lead us to postulate the existence of the Platonic Forms, but the Forms are entirely outside the world. Similarly, the order of the natural world might lead us to postulate the existence of Platonic teleology, but strictly speaking such teleology is – for Paley and ID theory, unlike Aristotle – not in the natural world itself but only outside it, in the mind of a designer.

Scholastic realism, then, corresponds roughly to Scholastic teleologism: Universals are immanent to the natural world, and therefore the natures of things can, at least to some extent and in principle, be known and studied without reference to their creator; and this remains true even though any ultimate explanation of universals and the things that instantiate them must make reference to God. Similarly, the final causes of things are immanent to the natural world and can, at least to some extent and in principle, be known and studied without reference to God – despite the fact that their explanation too must ultimately be referred to God.
Insisting on the immanence of universals and final causes is arguably crucial to avoiding the extremes of occasionalism and deism. For if we deny that universals and final causes – and thus natures and causal powers – are inherent to natural objects, then we are likely to conclude either (a) that there are no secondary causes, and God as first cause is really the only true cause of everything that happens, or (b) that since the world can continue to operate without inherent causal powers, it must also be capable of continuing in operation without the continued operation of the first cause. The first, occasionalist option tends in turn to approach pantheism, while the second, deistic option tends to lead to atheism.

I leave for homework the question of whether these inferences are strictly unavoidable – obviously they would need in any event to be fleshed out. Suffice it to say that there are theological as well as philosophical grounds to prefer what I have called the Scholastic position. As I have emphasized in TLS and elsewhere, there are scientific advantages too. For the further we go in the deistic-cum-mechanistic direction, the more we invite Humean skepticism about causality and thus threaten to make science unintelligible. While the further we go in the occasionalist direction, the less the world seems to reflect impersonal and predictable forces and the more it comes to seem in every detail comparable to the unpredictable behavior of a conscious subject – thus making natural science impossible. (As Alain Besançon has argued, a tendency toward an occasionalist conception of divine causality is part of what distinguishes Islam from Christianity – and this is no doubt one reason why natural science progressed in the West and stagnated within the Islamic world.)