On his radio show yesterday, Dennis Prager acknowledged that one reason he believes in God – though not the only one – is that he wantsit to be the case that God exists. The thought that there is no compensation in the hereafter for suffering endured in this life, nor any reunion with departed loved ones, is one he finds just too depressing. Prager did not present this as an argument for the existence of God or for life after death, but just the expression of a motivation for believing in God and the afterlife. But there have, historically, been attempts to develop this idea into an actual argument. This is known as the argument from desire, and its proponents include Aquinas and C. S. Lewis.
An obvious objection to any such argument would be that it is manifestly delusional to suppose that something is real simply because we want it to be. After all, the frustration of desire happens all the time – unrequited love, failed careers, empty stomachs, and so forth. Yet you might suspect that, precisely because this objection is soobvious, it must be missing something – that proponents of the argument from desire are not in fact reasoning in so crude and manifestly fallacious a way. And if so, you’d be right.
Hence, consider Aquinas’s argument from desire for the immortality of the soul. One version can be found in Summa TheologiaeI.75.6:
[E]verything naturally aspires to existence after its own manner. Now, in things that have knowledge, desire ensues upon knowledge. The senses indeed do not know existence, except under the conditions of "here" and "now," whereas the intellect apprehends existence absolutely, and for all time; so that everything that has an intellect naturally desires always to exist. But a natural desire cannot be in vain. Therefore every intellectual substance is incorruptible.
Another, longer version can be found in Summa Contra GentilesII.55.13.
Now, notice first that Aquinas does not say that just any old desire is bound to be satisfied. What he says is that “a natural desire cannot be in vain.” Let’s consider both of the italicized parts of this statement.
First, by a “natural” desire, Aquinas means a tendency toward some end that a thing has just by virtue of being the kind of thing it is – that is to say, by virtue of its essence. (To put the point in more technical terms, he is talking about immanent final causes grounded in substantial forms.) So, consider a tree’s tendency to sink roots into the ground so as to take in water. That would be an example of the sort of thing Aquinas has in mind, and as that example indicates, a “desire” of the sort he is talking about needn’t be conscious. For a tree is not conscious, but it still “desires” water in the sense that by virtue of its nature it will send out roots so as to acquire it.
Human beings, of course, do consciously desire things, and sometimes what we consciously desire is some end toward which our nature moves us. Water would be an example here too. Given our animal nature, we need water and will seek it out. Some of our desires, however, are not “natural” in the relevant sense. For example, suppose I desire a copy of Captain America Comics #1. That would not be an unnatural desire, to be sure, but neither is there anything in my nature that directs me toward that particular end. Simply qua human being, I will not be directed toward the end of acquiring a copy of that comic book, in the way that I am qua human being directed toward the end of acquiring water.
However, some desires can also be positively unnatural. A strong desire not to drink water would be an example -- illustrated by the woman in the movie Ed Wood who claims to be allergic to it. A real-life and only slightly less bizarre example would be the woman with a compulsion to eat household cleanser. Less extreme examples of strange compulsions, habituated vices, etc. are familiar from everyday life. As such cases illustrate, that a desire is very strong and deep-rooted does not suffice to make it “natural” in Aquinas’s sense.
What does Aquinas mean when he says that a natural desire cannot be “in vain”? He can’t mean that such desires are always in fact satisfied, because he is as aware as his readers are that they are very often not satisfied. For example, trees, people, and other plants and animals die of thirst all the time. What he means is that a thing couldn’t naturally be directed toward some end unless that end were real. Such desires can be fulfilled at least in principle even if they are not always fulfilled in fact. Hence if trees, human beings, and other plants and animals are naturally directed toward seeking out water, there must really be water out there for them to seek (even if they don’t always find it). If there were not, the desire for water would be in vain.
So, to refute the key premise that Aquinas’s argument rests on, it will not suffice to find examples of unfulfilled desires, or even unfulfilled natural desires – which are, of course, so easy to come by that it should be obvious even to the hostile reader that that is not what Aquinas had in mind. What one would need is an example of a desire that is both “natural” in Aquinas’s sense and also “in vain” in the sense of being unfulfillable in principle because its object is unreal. And examples of that sort of thing are far from obvious.
Indeed, even people who would not think of themselves as sympathetic to Aristotelianism or Thomism will essentially apply Aquinas’s principle without realizing they are doing so. For example, if a paleontologist digs up the remains of a heretofore unknown species of animal and finds that it has many long, sharp teeth, he will suppose that there must have been living in this animal’s environment other kinds of animal that it preyed upon, even if the paleontologist has no independent evidence of such other animals. For otherwise, the animal’s possession of such carnivorous teeth would be in vain. (And even if it were finally judged that the carnivorous teeth are vestigial, this doesn’t affect the basic point, because even in that case they would have originally existed for meat-eating purposes.)
Now, what Aquinas does in his argument from desire is to add to this premise about natural desires a further consideration about human nature, specifically. Other animals have only sensory knowledge, which is directed toward the here and now. Human beings, by contrast, are rational animals, and their sort of knowledge – intellectual knowledge, which involves the grasp of universal concepts and universal truths – is directed beyond the here and now, indeed toward “all time.” But whatever is like this, Aquinas says, “naturally desires always to exist.” In the Summa Contra Gentiles passage, he says that creatures with intellects “know and apprehend perpetual being [and] desire it with natural desire” (emphasis added). Hence we must have perpetual being, or our desire for it would be in vain.
Needless to say, even given the qualifications I’ve made, this argument needs further spelling out if it is to be convincing. And an objection raised by John Duns Scotus might seem at first glance to torpedo it. He writes:
As for the proof that man has a natural desire for immortality because he naturally shuns death, it can be said that this proof applies to the brute animal as well as to man. (Philosophical Writings, p. 159)
Scotus’s point is that if such a proof would fail in the case of brute animals (which it certainly would in Aquinas’s view, since he denies that such animals have immortal souls), then it must fail in the case of human beings as well.
Now, a non-human animal will, no less than a human being, be inclined as long as it exists to try to keep itself in existence. Scotus is right about that. But it doesn’t follow that brute animals have the same desire that Aquinas is attributing to human beings. To see why not, you need to get your Scotus on and draw a distinction. Consider the two claims:
(1) X always has a desire to preserve itself.
(2) X has a desire to preserve itself always.
(1) is true of brute animals, but (1) does not entail (2) and (2) is what Aquinas says is true of human beings but not true of brute animals. It is only if we blur the distinction between (1) and (2) that brute animals will seem to have the same desire Aquinas attributes to us.
But Scotus also raises a more challenging objection. In order to know that a thing has a natural desire to preserve itself always, we first have to establish that it has a natural capacity for perpetual existence. And if we knew that, we would already know that it is immortal, which would make an argument from desire redundant (Philosophical Writings, pp. 158-9).
Scotus is, I think, correct that a natural desire D presupposes a natural capacity C. But once again – the Subtle Doctor would be pleased – we need to draw a distinction. Even if the existence of D presupposes the existence of C, it doesn’t follow that knowledge of the existence of D presupposes knowledge of the existence of C. That is to say, the presupposition in question is metaphysical but need not be epistemological.
Consider, once again, the paleontology example. The existence of carnivorous teeth presupposes, in a metaphysical sense, the existence of prey who might be eaten. The former would not exist unless the latter did, whereas the latter could exist whether or not the former did. But it doesn’t follow that our paleontologist would first have independently to establish the existence of the relevant sort of prey in the environmental niche in question before judging that the teeth serve a carnivorous end. His general knowledge of the kinds of teeth there are suffices for that. Hence, if all he knows at first is that a certain environmental niche was populated by a kind of animal having carnivorous teeth, he can go on to conclude that there must also have been prey of the relevant sort living in that niche. He does not have to remain agnostic on that question pending direct evidence. The presupposition in question is not an epistemological one.
In the same way, Aquinas can argue that there is a natural desire for perpetual existence – and he does so on the basis of the nature of intellectual (as opposed to merely sensory) knowledge – and then go on to conclude that there must be a natural capacity for such existence. Knowledge of the desire can be prior to knowledge of the capacity, even if the capacity itself is metaphysically prior to the desire.
So, does Aquinas’s argument work? I think that when all the relevant metaphysical background theses and the subsidiary arguments are spelled out thoroughly – and I haven’t addressed all of that here – it plausibly does work. However, anyone who is convinced of the soundness of that larger body of philosophical claims is also likely already to be convinced of the immortality of the soul by other and more straightforward Thomistic arguments. So, while Aquinas’s argument from desire is a useful and illuminating part of the overall Thomistic view of things, it isn’t the most effective standaloneargument for immortality.
Compare the situation with Aquinas’s Fourth Way of arguing for God’s existence. I defend the Fourth Way, along with the rest of the Five Ways, in my book Aquinas. But one has to do so much general metaphysical stage-setting in order properly to understand how the argument works that, for purposes of establishing God’s existence, it is much more efficient to use a relatively more streamlined argument like one of the other Ways. Once one is independently convinced of the overall Thomistic system, the Fourth Way provides a very important and illuminating part of the story. But it’s not a good way to break into the system.
Or at least this is the case given the situation that happens to exist in contemporary philosophy. In a context where broadly classical (Platonic or Aristotelian) metaphysical presuppositions were widely accepted, arguments like the argument from desire or the Fourth Way would be much more plausible standalone arguments. But when those larger background presuppositions are not taken for granted and are even treated with some hostility, it is generally more effective to make use of other arguments.