Aquinas’ Fifth Way

Abstract

Aquinas’ fifth way for demonstrating the existence of God shows a confidence in its argument that is not generally shared by the contemporary reader. Natural entities that lack awareness (or perhaps self-awareness) act in an end-directed manner; this implies a fundamental relation to goals they do not consciously choose; this in turn points to the existence of a governing mind. The article tries to make sense of this argument against the background of philosophical naturalism that is assumed by much contemporary thought. It addresses the question of what kinds of end-directed activity fall under the scope of the argument, and why Aquinas thinks that these imply a governing mind. The article notes the unusual structure of the argument, suggesting that it should be understood (as should all of the five ways) as an attempt to wake the reader up to something so fundamental that it is usually overlooked. The main difficulty for the contemporary reader is to recover an Aristotelian sense of what it means for something to exist. If readers can achieve this, they may come to appreciate why an argument that seems enormously controversial in the present day seemed reasonably obvious to Aquinas.

I. Introduction

Thomas Aquinas thinks we live on the edge of something that is of a different sort from anything else we know. While it is present to us at every moment, it is not a kind of thing in the way that other things are, and does not fall under a larger common genus.¹ We have to work a bit to become aware of it, but we don't have to work that much; a few lines of reflection will do. This is what Aquinas seems to have thought about the existence of God. The fifth way is the shortest of the five ways for demonstrating God's existence. Aquinas calls it the way from “governance.” Things lacking awareness (or maybe “understanding”) act for an end. This is shown by the fact that they almost always do what is best for them. We should conclude that they are governed by something with “awareness and understanding,” in the way that we conclude that if an arrow hits the target almost every time, it is shot by an intelligent archer. Aquinas concludes that the behaviour of natural things points to the existence of a heavenly governing intelligence.²

The brevity of the proof is surprising. Aquinas obviously thinks it clear enough, and offers a mere summary of the argument, as he does with all of the five ways. Presumably he relies on familiarity with material he has dealt with elsewhere.³ But to a contemporary reader, the proof can seem not just brief, but positively careless. It begins with agents that lack some of the elements needed for acting in an end-directed manner, although they clearly act in such a manner. It concludes to the existence of something that supplies the missing elements. Logic seems to require that the elements that are lacking are the same ones that are then supplied by the additional agent. But while Aquinas has cognitio for the former, he includes intelligentia with the latter, which seems to make the conclusion of the proof stronger than the premises warrant. End-directed activity by something that lacks awareness might well imply the existence of something that possesses awareness. But it does not seem to imply any more than this. As we know, a stick can be converted to end-directed purposes by a chimpanzee; there is no reason to conclude that the animal therefore possesses intelligent understanding.⁴ Furthermore, the “natural bodies” referred to in the first part of the demonstration seem limited to plants and natural elements. But the end of the demonstration refers to omnes res naturales “all natural things,” as though everything requires divine governance in the way specified by the proof. It is as though the detail does not matter much, given that it is just a sketch of something that everyone understands, and with whose conclusion everyone is more or less in agreement.⁵

In this article, I want to make sense of what the demonstration is trying to say, and then to look at the difference in worldview between Aquinas and ourselves that leads him to regard the demonstration as reasonably obvious, while this does not seem to hold for ourselves.

II. The Question of Nature

The argument maintains that God relates to the processes of the world as an archer relates to an arrow speeding towards a target. If I am a bystander looking at a wall with a target painted on it, and arrows are flying through the air from somewhere behind me, and all or most are hitting the target, I conclude that they are being fired by someone with knowledge and intelligence. They are not just being blown by the wind or whatever. Something is in play that believes that hitting the target is better than not hitting it, and which is obviously trying to achieve the best outcome, and to avoid worse ones. Aquinas thinks that a good number of the entities of the world—perhaps all of them in the end—are like the arrows. Plants are the most obvious example. Always, or for the most part, they actively do what is best for them, although they lack knowledge. Aquinas thinks we should conclude that there is an intelligence in the background, directing them.

In another discussion on the question of whether nature acts in an end-directed manner, in the commentary on Aristotle's Physics, Aquinas broadens the range of examples to include animal activity, seemingly agreeing that the premise should not refer just to a lack of “awareness” or “knowledge” among the end-directed agents, which would leave animals out of the discussion, but to a lack of “intelligence,” which could therefore include animals, though not humans. He repeats standard Aristotelian plant examples of leaves shading fruit and roots pointing downwards so as to absorb nourishment, and then adds some animal examples (also found in Aristotle), a swallow building a nest and a spider spinning a web, holding that these also show natural end-directedness of the sort that is under discussion, and concluding that they must be moved by something that has understanding or intelligence.⁶

Whatever the detail of the argument, it does not sound promising. There are two main problems for a contemporary reader. The first is our lack of clarity about what exactly counts as an example of natural teleology, and whether such examples are in fact found in nature. We are aware of extraordinary results that arise simply from chance configurations of material parts, and feel that it is at least possible that everything in the world might arise like this. Mass, gravity and centrifugal force produce regular elliptical orbits;⁷ a volcano produces a beautifully patterned ash-cloud; and above all, in evolutionary biology a blind selection process produces an unbelievable range of functional organisms. Richard Dawkins, who insists that possibilities of wonder remain even for those who affirm philosophical naturalism, strongly denies that we ever need appeal to anything more than materials and properties to explain the emergence of such entities.⁸ There is therefore a need to get clear on the kinds of things or processes Aquinas refers to when he speaks of things that act regularly in an end-directed manner, so that their action is for the “best.” The second problem for a contemporary reader is that the fifth way seems to imply that certain things within the created order cannot perform some of the basic tasks that seem part of their normal functioning. So there is need to invoke a higher agent, something larger than anything we know, but which operates more or less alongside the things we know, in much the same way. This sounds like an appeal to the God of the Gaps. Both of these questions have received much attention in recent decades, focussed by the Intelligent Design controversy. Here the claim was made that natural processes are insufficient to explain some of the results of evolutionary development, implying that the divine sometimes interferes directly in the evolutionary process. Because the perspective of this controversy still tends to dominate the discussion, the ontology that underlies Aquinas’ argument needs clarifying, to see just what kind of thing he is talking about.

In the first place, Aquinas and Aristotle assume an important distinction between two kinds of action. Much of what goes on in the everyday comes down to something being interfered with by another, where the second thing alters what the first would otherwise have done. As well as this, the Aristotelian tradition recognizes a different sort of action, which is seen as more fundamental. It is not imposed on a thing at all, but rather proceeds out of its original identity. It describes for example the urge in a plant to send out roots, and generally to take the steps needed to ensure its future, so that it comes to completion. Among the many things it could do or not do, it consistently aims at the right thing and sets about realizing it. Aristotle sees actions of this sort as “natural.” These seem to be the kinds of action that Aquinas sees as requiring divine governance.

The difference between the two, an action where one thing is interfered with by another, and an action where an entity moves to complete itself, can be seen most clearly in an animal example. In one of his best-loved books, R. L. Stevenson converts a donkey called “Modestine” into a pack-animal to carry his bedding, as he travels through parts of south-eastern France. He realizes he has turned her into a sort of artificial object, and jokingly refers to her as “an appurtenance of my mattress, or self-acting bedstead of four castors.”⁹ The joke reflects the fact that the bed-carrying has been imposed on a more fundamental set of ends and actions that belong to Modestine herself. These include breathing, feeding, doing the things donkeys need to do to survive and thrive. They are not imposed on her in the way that carrying the bedding is imposed; rather they help constitute her original identity as a donkey. Stevenson exploits Modestine's natural commitment to these ends, her need for food and drink, her dislike of the goad that pricks her if she refuses to move, to turn her into a bed-carrier. She herself is not unaware of what is happening, and her behaviour shows that she realizes that carrying the pack is not among her original duties. She stages small rebellions along the way. So there are two levels, a natural level of urges and actions that constitute the identity of the thing, and an artificial level where something is imposed from outside. Aquinas maintains that God is needed to constitute the original level of natural urges and actions. We need to invoke God to explain why Modestine moves towards realizing the original goals of a donkey, feeding herself and so on, and aiming to achieve the “best” result, as the fifth way puts it.

Aristotle calls actions imposed from outside, like making the donkey carry the bedding, “violent” actions. The fifth way does not see God's governance as imposing itself in this way. Later in the Summa, Aquinas says that God's governance is applied to each entity in a way that respects its nature. Non-rational creatures are governed differently from rational creatures. Governance does not therefore mean that a single uniform action is imposed on everything from outside.¹⁰ We see that the archer example is in fact misleading, because the flight of the arrow is not a natural action, but is imposed on the arrow in the way that bed-carrying is imposed on the donkey. By contrast, when talking about autonomous agents, Aquinas emphasises that God's governance works “interiorly” in them, and does not force them from outside. This seems to set the tone for what he means by God's “governance.”¹¹ The argument is, therefore, not about the roots of a particular tree being moved artificially by a divine agent, as if God has to step in and nudge them in the right direction. Rather, the drive of the tree itself to feed and maintain itself and push its roots out, reflects divine governance. Aristotle and Aquinas hold that there is something surprising here, as if timber moved of its own accord to form a ship.¹² If something like this happens, Aquinas thinks there must be a governing mind in the background. Obviously the conclusion does not speak for itself. It is even in tension with other aspects of Aristotle's position, given that it implies a logical entailment from natural end-directedness to the presence of mind. This would suggest a stronger analogy between natural entities and artefacts than Aristotle wants to allow, with the result that nature could start to look like a subcategory of art. To confuse nature with art in this way has been described by one commentator as an Aristotelian “mortal sin.”¹³ For all that, Aquinas presses a strong analogy between the two.¹⁴ Presumably his difference with Aristotle goes back to the peculiar role of the creator in Aquinas’ thinking.

III. The Notion of Goal

Aquinas’ reasoning to a governing mind becomes clearer with close consideration of the notion of a goal. A goal has to be distinguished from a mere “outcome,” which is a result that simply follows from something else. The dust-cloud that rises after an eruption looks beautifully patterned, and the patterning is a regular occurrence. We do not, however, say that the eruption “aims” at it, as if it were a goal or good, or that it represents the completion of the mountain's processes. Similarly, if arrows are fired at random at a wall, the resulting pattern of hits is a mere outcome, and does not indicate the presence of a goal. To be identifiable as a goal, a particular outcome needs to be identifiable as an independent good, so that we know what it is before the action starts. In some sense, the target has to be already in place. If someone says “this will be a great shot” and we ask, “what are you aiming at?” and they say “I'll tell you afterwards,” we can reply that this indicates an outcome, but not a goal. If a musical performance is to count as end-directed activity, and not mere doodling on an instrument, the piece must already exist in some way, sitting on the piano in manuscript form, or waiting in the mind of the player, representing what the performance is aiming at, and providing a norm by which it can be judged. Similarly, if Modestine's natural activities are to be seen as goal-directed, the goal must be already somehow in place before the activity begins. A particular outcome has to be distinguished from all the other possible outcomes as constituting the successful future of a donkey. The argument of the fifth way holds that there are clear examples of end-directed activity in nature, where there is however no earthly mind in play that could privilege one of the sets of outcomes over others. Examples are all around us, with plants, swallows, spiders, and so on. If they are genuinely seeking a goal, the goal has to be held somewhere, in a mind, given that its prior existence is a condition for end-directed activity.

Given the difficulty of the material for a contemporary reader, it helps to look at the kind of explanation that Aquinas wants to exclude. This would see outcomes simply as following one another in the way that sets of particle-clouds or swarms succeed one another, without any privileging of a particular outcome over others. We might group particles together and give them a name, as happens when a particular swarm is conceptualized as a “donkey.”¹⁵ But in themselves entities remain collections of particles that happen to exist alongside one another. As such, they have no relation to better or worse, no internal goal or standard to which they are trying to measure up; to all futures, they just say “whatever!” Aristotle calls things like this “heaps.” He says of the archaic Greek philosopher Empedocles, who held such a position, “there is no origin (physis) of anything, but only a mingling and separation of things which have been mingled.”¹⁶ Aristotelian ontology even includes artefacts in this category. While an interested user holds the parts of an artefact together in light of a purpose, the purpose belongs to the user, and not to the artefact. A particular configuration of parts is called a “vacuum-cleaner.” But while it serves the goal of keeping the house clean, it does not itself have any interest in this, or in anything else. Only when seen in light of the interests of a house-owner do its parts cease to be a mere “heap,” and come together as a whole.

Aristotle is fascinated by the peculiar unity that characterizes natural wholes, a unity that is in his view fundamental to what it means to exist. Living things are the best examples. He recognizes that their constitution does not just come down to the location of parts alongside one another. If it did, then a newly dead human would be very close to a live human, although lacking some of its more interesting qualities. Aristotle insists it is not like this. With the departure of life, the human has not lost a quality; it has rather ceased to be. The peculiar living unity that characterizes its parts while it is alive is not a property or quality, but is the very existing of a living thing, as Aristotle says.¹⁷ Whatever happens to Modestine, she comes back and back to the same pursuit of goals, as if to say “I am a donkey, for goodness’ sake,” and when she eventually loses touch with these goals, she does not lose a set of qualities, but ceases to be, relapsing into a heap of materials. For Aristotle, the integral unity of living entities is not to be contrasted in the first place with mere elements (as tends to happen in the contemporary world), but with collections or heaps of objects. Heaps as such have no interests of their own, though their natural materials retain their original natures, along with their primitive “actions.” This applies even when the materials have been formed into artefacts.¹⁸ Because a heap has no interests, it has no real identity.

Aquinas thinks that if a natural process is end-directed, the goal must be held somewhere, in a way that keeps it independent of the status of a mere outcome. Only in this way does it become a genuine goal, something that serves as a norm for an end-directed process that is trying to get something right. Aquinas sees such norms as going back to the very life of God, which they try to represent and imitate in their own lowly way, each of them caught up in one of the ways that the divinity understands itself as imitable. Even Aristotle echoes aspects of this, seeing the attempt of living things to reproduce themselves as so many attempts to participate in the life of eternity, although Aristotle does not think that the divinity is interested in them, or that it has knowledge of them.¹⁹ By contrast, Aquinas sees God as an active creator, and holds that the different ways in which the divine pattern of life can be represented are held as ideas in the mind of God, and serve as models for the act of creation, holding the natural order together.

Hence we should say that divine wisdom holds the originals of all things, and these we have previously called the Ideas, that is the exemplar forms existing in the divine mind. And though they are many and various in the relationship of things to them, nevertheless they are not really other than the divine essence proportionably to the manifold sharing of its likeness by diverse things.²⁰

This means that the entities of the world are so many attempts to get something right, in relation to which they can measure up or fall short. They are not just heaps, even heaps that happen to have come together in a functional way. Rather they are constituted by a relation to a particular set of ends. Aquinas understands teleology as this original direction to ends that constitutes the very substance of earthly things.²¹

IV. A Clash of Ontologies

We may get a glimpse of what the argument is trying to show; we probably feel it is a long way from anything we can take for granted in the contemporary world. The current default ontology of the West sees things precisely as heaps, configurations of loose materials that have sometimes taken on a functional shape that is the outcome of a long evolutionary history. For scientific naturalism, a plant puts its roots down and not up because an ancestor collection of materials survived that way, and reproduced, passing on the necessary mechanisms. While a functional arrangement manages to reproduce itself, it does not aim at anything, and in the end it has the same ontological status as any other heap. Evolutionary naturalism denies what seemed obvious to Aristotle and Aquinas, that natural things have a unity that binds them together in a way that is qualitatively different from the unity of mere collections of materials. If Aristotle and Aquinas thought it madness to propose a philosophical naturalist account, contemporary Westerners find such an account familiar, and, it has to be said, live with it comfortably enough.

To conclude, I will offer some reflection on the wider context of this extreme divergence of viewpoint, where something that Aquinas saw as reasonably obvious has become a matter of doubt and conjecture. Clearly the background assumptions for an enquiry into God's existence have changed between Aquinas’ time and ours. Contemporary thinkers tend to assume that a question about the existence of God is not too different from any other question, for example whether there is a tenth planet. In fact, however, there is a striking difference between the logic of these two questions. When we ask about the existence of a contingent thing, we begin with the thought that there might or might not be one of these, and this thought remains unchallenged even after we have concluded that the thing does exist. It is very different with the question of the existence of God. We start off with an open mind, thinking that there might or might not be a God. But if we come to a positive conclusion, we revise our view of the opening of the enquiry, realizing that we had it wrong. It was never in fact possible that there was not a God. The argument for God's existence therefore involves a striking transcendence of its own beginnings. It is as if the central move is a kind of waking up to what was always necessarily the case, though it might not have been apparent to us. Necessarily, the things of the world were held together by God's governance, though it took time for us to wake up to the fact. Aquinas seems to have been able to rely on a heightened state of wakefulness here. This might explain the strange ambiguity at the start of the fifth way, where we begin with natural bodies and plants, but conclude at the end to something that is true of all natural things. Aquinas chooses the most striking example for us, realizing that the important thing is to jog us into awareness, so that we wake up to the necessity of universal governance. In principle, he could have taken any example; the one he chose was deemed the most useful for us.

Aquinas was familiar with versions of the alternative, that there might not be a divine governing mind, and entities might just be heaps of particles. He reacts to the view in a way that might surprise us, saying that if the world is this way, there aren't really any entities at all. There are merely heaps, and heaps do not exist in any robust sense, but remain “a mingling and separation of things which have been mingled” in the Aristotelian phrase noted above.²² Aquinas says that “nothing can exist that is not referred to the divine goodness as end.”²³ Things without higher governance “would do nothing intentionally”²⁴ (the older translation perhaps has it better, saying that they “tend to nothing definite.”)²⁵ Something that is a directionless heap all the way down, so that its materials are heaps in turn, and so on forever, lacks any necessary relation to an idea, and therefore escapes from the order of divine governance. Probably we cannot in fact imagine such a thing, except as a limit case. Aquinas says bluntly, “if it were totally outside the sphere of God's government, it would be absolutely nothing.”²⁶ If it is not trying to realize some pattern that gives it a natural direction, it is reduced to something like a point in space, a locale that at best registers a succession of disparate outcomes. Something that tends radically to nothing definite like this, cannot be said to “exist” in any obvious way. The thought is powerfully expressed by C. S. Lewis in A Grief Observed, where he refers to the possibility that scientific naturalism is true, and that his deceased wife, to whom he refers by the initial “H,” was really just a combination of particles, so that she could not in any way survive her death. Lewis says this:

If H. ‘is not’, then she never was. I mistook a cloud of atoms for a person. There aren't, and never were, any people. Death only reveals the vacuity that was always there. What we call the living are simply those who have not yet been unmasked. All equally bankrupt, but some not yet declared.²⁷

Lewis speaks here from within Aquinas’ and Aristotle's world, where entities are constituted by teleological drives, which give them an identity. For Aquinas, these constituting goals have to be held in the divine mind if they are to exist. For contemporary scientific naturalism, by contrast, there are swarms of particles, some of which have developed interesting looping configurations and unusual functional properties, so that the pattern of life described in ancient and medieval philosophy as an “essence” has no place any more.²⁸ The clash between the two positions is perhaps the most interesting philosophical divide of our time.

Lewis offers an argument for the Aristotelian point of view. After describing the “vacuity” of a swarm of particles, its ultimate meaninglessness, and wondering whether this vacuity could characterize all the things in the world, he says

But this must be nonsense; vacuity revealed to whom? Bankruptcy declared to whom? To other boxes of fireworks or clouds of atoms. I will never believe—more strictly I can't believe—that one set of physical events could be, or make, a mistake about other sets.²⁹

We cannot arrive at an end-directed activity like knowledge, out of heaps adjusting themselves, where everything comes down to mere outcomes. Lewis makes the point well. But philosophical naturalism could reply that the argument begs the question. The philosopher Richard Taylor maintains that when we think a cloud of atoms incapable of end-directed activity like understanding, it is because we think of it as being in the end something like a stone, and we know that stones don't understand. But what if we said that while certain clouds of atoms, the ones that make up stones, cannot understand, everyone accepts that other clouds of atoms, for example the ones that look like human beings, can understand without any problem?³⁰

Lewis's argument comes down to a familiar point in such exchanges, that a position like that of Taylor excludes the knowing subject, and focuses resolutely on the objects known, if the account is to work. To hold that our thinking and talking come to pass in the same way as an ash cloud rises from an eruption, seems to leave out the subject doing the reflection. A philosophical tradition can simply shrug its shoulders here, refusing to let itself be worried by the fact that the reflection denies its own possibility. But the worries tend to linger. We live in a world that wants to understand itself as did the world of Democritus and Lucretius, but also wants to keep hold of notions of freedom and spirituality, concepts that once seemed important for human self-understanding. Some contemporary thinkers, like Nancey Murphy, think that in their old forms, these concepts are not so important, and that they can be adjusted. We can settle for a scientific ontology and still have a spirituality, without any problem.³¹ Others are not so sure.

These impasses show how unresolved questions of ontology sit behind discussions of the existence of God, and probably render such discussions inconclusive in the contemporary world. Aquinas argues from within an ontology that he sees as needed for the things of the world to be properly articulated. From this position, it is relatively easy to show that we are committed to a belief in God. The fifth way summarizes a passage of thought which makes this commitment explicit. But when the ontology no longer speaks for itself, as is the case in the twentieth and twenty-first centuries, a hasty focus on details of a proof for God's existence might misplace the needed emphasis. The real question is back with the ontologies that are in play, raising the question of what exactly leads us to commit ourselves to one ontology rather than another, given that each in its own way includes all the possible “evidence.”

The locus of the problem is apparent in a completely different context in remarks of the philosopher Fichte, in talking about the relation of belief in God to fundamental ontological commitments. Fichte dislikes attempts to make God's existence an object of proofs. He believes that the important step is rather a kind of ontological choice between two worlds. One of these, which includes the possibility of freedom and morality, he considers to be the right choice, given it affirms the fundamental conditions for the very activity in which we are engaged as we consider such things. It is the only choice that we can reasonably make, however much we might try to convince ourselves otherwise. He thinks it is also a commitment to belief in God. In the short text “On the Ground of Our Belief in a Divine World-Governance” he says

It is therefore a misunderstanding to say that it is doubtful whether a God exists or not. It is not doubtful at all but rather the most certain thing that there is. Indeed, it is the ground of all other certainty, the single absolutely valid objective fact: that there is a moral world-order, that a determinate place in this order is assigned to every rational individual … that the destiny of each individual … is a result of this plan; that without this plan no hair falls from his head and within his sphere of activity no sparrow falls from a roof; that each truly good action succeeds and each truly evil one fails; and that for those who rightly love only the good, all things must conduce to the best.³²

For this point of view, every argument affirms a moral order, and this in turn affirms the existence of God. For Fichte, these come down more or less to the same thing, a conviction that caused him considerable trouble with the religious and civil authorities of his time. If one agrees with this, the rest is settled quite easily. End-directed natural entities point to a divine mind that holds fundamental patterns of existence in place. While the fifth way of Aquinas shows us how to proceed with such a train of thought, it assumes important questions of ontology, which are in our own time, alas, far from settled.