Argument from Desire

That's an interesting argument. I have a question about (2) that might clarify things for me. Here's (2) again.

(2) Most natural desires are such that there exists some object capable of satisfying them.

I wonder if you mean to say that most desires are such that there *possibly exists* something that satisfies them rather than that there (actually) exists some object that could satisfy them. Here's why I ask. It seems natural to maintain that we generally desire what we don't have. So,
a. I desire to be President of the USA.
The state of affairs in which I am President is a possible but non-actual state of affairs.
b. I desire that I get more sleep.
Or, say,
c. I desire that I win the lottery.
All seem to be about possible states of affairs, rather than actual ones. So there does not exist some object capable of satisfying these desires. Rather, the object of these desires is a possible (but nonactual) state of affairs, or object, etc. On that reading of (2), wouldn't your argument show that it is possible (rather than actual) that the transcendent exists? And wouldn't we agree--I would anyway--that the transcendent is possible?

Hi Mike,

Good question. We’d resist that weaker reading of (2). Merely possible objects cannot enter into the right kind of story for explaining how persons have these natural desires. I take it for most natural desires there are (or were) things capable of satisfying those desires. One difficulty that I think your comment gets at is how to make the distinction between kinds of desires and desires for particular things. I have a desire for a time travel machine, but there are no time travel machines (maybe there cannot be such things!). The tactic in this case, as in your cases (a) & (c), is to explain how this desire is a manifestation of a more general desire that has or has had objects capable of satisfying that desire.

Mike, thanks this is a very good question. a. and b. don't fall under the current conception of a *natural* desire, since presidencies and lotteries are conventional. And with b. we are only concerned with the fact that there is such a thing as sleep, not that I will necessarily get it. I have a natural desire for food, and there *is* food, but I might starve nevertheless.

Still, your examples provide an excellent challenge to state the principle more precisely. As I see it, the central challenge of your examples is to separate *myself* from the state of affairs so the remainder is the extra-personal target of the desire. Here's my first pass. The intuitive idea will be to treat desire as at least a two-place relation between myself and some other thing. The second relatum is the thing which gets the benefit of the doubt with respect to positive ontological status.

So b. is NaturalDesire(me,sleep). In general: If NaturalDesire(me,x), then probably x exists.

Thanks for forcing a more structural statement of the principle.

Thanks. I certainly agree that presidencies, for instance, are not natural *things*. But then is it a necessary condition on D's being a natural desire that the object of the desire is a natural object or some natural state of affairs? The transcendental (whatever it finally comes to) is not natural, either. But then, there really is no *thing* or object, natural or nonnatural, that is picked out by 'the transcendent'. I'm pretty sure you'd agree. The phrase looks like it refers to some object in the way that 'the food' does, but it doesn't. The desire for the transcendent (assuming there is a desire properly so described) is not a desire for an object or thing, certainly, it's closer to a desire for a state of being, or a desire for a certain experiential state. I suppose you can quantify over such states (if you have to) but their existence doesn't show (again, as far as I can see) that there is a thing "the transcendent" that exists somewhere beyond.

(Quick caveat - I can't see my formating in the preview comment screen, so I hope this doesn't come out as one big text blod!)

Interesting argument, and I'll be curious to see where you guys take the AFD. Let me throw a criticism your way and see what you think.

First, although you've already highlighted some of the problems with your first premise (e.g., vagueness), I think you ought to consider these problems more lethal. Apart from a well-articulated metaphysic, it isn't clear to me how you can defend the notion that humans have desires "by nature", and that one of these desires involves "a desire for the transcendent." That is, while some conversation may certainly happen at a "moderately low level of definitional precision", this premise doesn't seem one of those places to me.

Consider that a more or less Freudian psychoanalyst would be able to agree with premise 1 in the main, but that his agreement with your premise (at it's appropriate level of vagueness) would end up sabotaging the whole argument. Transcendence might be interpreted as the desire for reunion with the quasi-narcissistic subjectivity lost by separation from the mother-womb at birth. Interestingly enough, such a desire would also be impossible to fulfill (at least, a non-sublimated version). Here's one way having desire "by nature" can be interpreted tragically - that is, on this view the world we engage (indeed, to which we are erotically attached), involves a necessary frustration of human desire rather than its satisfaction.

Further, it looks like the psychoanalysts would be challenging premise 2 through this interpretation of premise 1. Even though infants really do live for a period before they come into their 'psychological birth' (using Mahler's term), their desire to return to the womb (what is equivocated here with religious desire) may not be a simple desire but a phantasy. Ergo, while the content of the phantasy might be constructed from lived experience, the phantasy itself might be for an object that isn't part of the world at all (here phantasy provides and organization and level of intelligiblity to desire). Psychoanalysts will have various stories concerning how this can come about (e.g., the desire appears at a level of psychic organization not possessed prior to the separation and loss of wholeness). The point relevant to premise 2 of the argument, though, is that nothing will have ever existed or will exist that will satisfy the infant's desire. The Transcendant (Other) is "by nature" frustrating.

(The above interpretation of religious desire can be found in Freud's Civilization and Its Discontents. It also receives some elaboration in Lacanian psychoanalysis).

Hopefully I haven't been too vague myself and have understood your argument properly. I apologize ahead of time for any horrible misunderstandings. I do think your argument has promise, I just don't think it can be persuasive apart from (dare I say it?) a framework capable of articulating the being that is Dasein (ahh, that felt good). Heidegger jokes aside, I suspect the AFD's plausibility rests on a peculiar interpretation of its own premises. Psychoanalysts comprise one group that won't find the premises compelling in the way you've intended.

Mike, thanks for pressing on this.  "Natural" modifies the desire, not the object.  The reason I pointed out that that the *objects* were conventional was that, in that case at least, conventionality of object entailed conventionality of desire (I suspect this is universal).  No one is born with a desire for being president in the same way they are born wanting to nurse (though raw political ambition can make it seem that way!).  However, insofar as you can "build" the desire to be President out of natural desires like, say, the desire for prestige, the desire for social dominance, etc., then I'd say like all natural desires, they are satisfiable.  Different types of desire will have different types of satisfaction conditions, some will aim at concrete objects, some at more etherial states of affairs.  In any case "natural" will always modify the desire not the satisfier.  Talk of an "object" of desire was merely functional.  The "object" could be as ethereal being nurtured.  This is a state of affairs when entails a nurturer, so I think some of the states of affairs which are the objects of desire will entail objects properly so called.  The desire for endless and perfect love not limited by space or time plausibly involves an "object" like God, although it is interesting to note in this regard that the AFD is sometimes said to be an argument for Heaven rather than God directly.  Peter Kreeft uses the argument this way in his Heaven: The Heart's Deepest Longing.  So your point is taken to heart and I am working on a clean formal principle which accommodates, as it must, the concerns you ugre.

Josh, thanks for the thoughtful and challenging comments.  I am very sympathetic with your concerns about the vagueness of the terms involved and a few years ago I might have thought they were lethal.  However, I think we have plenty of resources for dealing with vagueness one of which is the strategy I used: definition by ostension: natural desires are ones like these: A,B,C,...  I think Socrates was a little unreasonable about always wanting necessary and sufficient conditions.  Some of the main points of analogy are originality, universality, and spontaneity: humans have always been religious, they are religious everywhere they are, and though, like the language instinct, human religiosity needs to be structured by culture as it develops, it seems to develop of his own accord.  I think this gives (1) enough content to be evaluated for worthiness of belief or degree of truth. 

With respect to Premise 2, it seems you attempt to give a counter-example: an unsatisfiable natural desire.  However, I think this fails both materially and formally.  First, the desire to return to the womb must be either taken literally or figuratively.  For the very very young it could perhaps be taken literally.  In this case, though the desire is highly unlikely to be satisfied, it is not unsatisfiable.  The womb *does* exist.  The baby *could* be put back in (and in the case of certain prenatal surgeries sometimes basically are).  However, in most cases the desire is really just a trope for desiring warmth and safety, things that clearly do exist and can be had under the right circumstances.  So I don't think the desire in question is unsatisfiable at all.  Secondly, even if this was a c-e, one counter-example doesn't affect a defeasible argument.  It is understood from the beginning that there may well be exceptions, that's why it's an *inductive* not a *deductive* argument.  What matters for a default rule in a defeasible inference is that the object *typically* has the target property.

Thanks Trent. I wish you luck with this argument; given my commitments, I certainly hope you're right. Here's a very quick word. You mentioned having a "Bayesian" formulation. As I see it, the argument you offer is going to stand or fall depending on the probability of alternative hypotheses and their explanation of the same phenomenon. Your argument is therefore going to resemble in structure Draper's well-known best-explanation-for-pain argument. As you know, this is where the real argument begins. Here it is not enough to offer a reasonable story (and honestly you guys are so far offering a not unreasonable story about the source of the relevant desire). You've got to show that you've got a better story than is offered by other hypotheses; indeed you've got to show that it is a better story than is offered by other hypotheses whose prior probability is much higher! The prior probability of an evolutionary explanation of the desire for the transcendent is going to be much higher than the God-implanted-the-desire story or the longing-for-home story or however you decide to explain the fact that we have such a desire. I frankly don't find Draper's argument especially convincing--and he had the probabilities on his side--so I don't expect to be confirmed in my convictions in this much more diffcult task you and Ted are undertaking. Still please send me your Bayesian version, when you're prepared to circulate it. Thanks again.

Mike, lucid comments as usual.  We do indeed hope to present a solid bayesian version of the argument.  Our current gambit is to just through out our fleece and attempt to rebut alternatives as they come in.  I do think we are helped on the priors issue by the way we frame the argument as an argument against naturalism.  So we get the benefit of the sum of all the ways naturalism could be false.  We see the role of the AFD in a total apologetic as primarily promoting further inquiry.  The conclusion is very general: there Something More than Nature.  In this we are co-belligerents with all kinds of religious views with which we have finer-grained disagreements.  We think those issues can be sorted out upon further investigation, but a first step is moving beyond naturalism.  The generality of this first step helps the priors problem at least a smidge and possibly a substantive amount.  And I'm not so sure the evolutionary explanation has so much going for it.  It will have to be a pretty particular explanation it seems to me and that will hurt its priors.  Part of what drives my intuitions here is my agreement with Swinburne that theism is a remarkably simple hypothesis, so that explanations which advert to it don't take as hard a hit to the priors as one might think.  Furthermore, that evolution has high probability on background evidence does not entail that it will have that much of an advantage in any given particular explanatory battle (and for that matter, if we are going to include background evidence for evolution as such (which we certainly don't dispute, we assume evolution is true here) then it only seems fair to do so for theism as well.  The dynamics of updating get very hard to track very fast.  Maybe Tim McGrew will help us out here (hint, hint, Tim), he's discussed similar issues with clarity in his interaction with Plantinga.  We're still following his work there and probably won't publicize the bayesian version until the dust settles there and we have a chance to really take it in).  Also, I tend to have broad sympathy with Draper's pain argument.  Given our over-arching apologetic strategy, one way to read our effort here is as fighting for a place for the AFD at the table of the cumulative case for theism.  We think it's harder than one might first think to martial naturalistic resources to show why there is some special reason why this particular natural desire is not like the others in being satisfiable.  Our plan is simply to consider any particular versions of a naturalistic explanation and come to grips with them honestly.  We have to see those alternative explanations first though.

I am fascinated by this argument, but let me play devil's advocate.

When we talk about natural desires, I presume we are talking about desires that all healthy human beings have a tendency to feel. Of course, the manifestation of any such natural desire in a particular case may be a desire that is particular to an individual. It is natural to me, qua human, to desire a long-term partner who will provide companionship and children. In my particular case, this may manifest itself as a desire to marry N, but that is not to say that most human beings desire to marry N. Furthermore, the pertinent fact is that the natural desire for long-term partner, companionship and children is one that can be satisfied in the natural run of things. This does not mean that I should expect that N should want to marry me. If she rejects my courtship, that would not, I think, constitute an unsatisfiable natural desire in the relevant sense. The counter-example would have to be something that is never satisfied in the natural course of events, rather than something that is not satisfied in many individual cases.

Evolutionary psychology can tell us a lot about why we have desires that can be satisfied. If we were satisfied without food, we would starve to death, and if we were satisfied without sex, we would not reproduce and leave descendents. Successful reproducers will have been unsatisfied until they completed many activities, therefore there is evolutionary value in these activities being desirable, and in being unsatisfied until they have been completed. Being easily satisfied is bad.

However, the awareness that we have of time can create an awareness that many of our desires cannot be satisfied in the natural order of things. I want to be loved today, and tomorrow, and the day after, that is natural. It is not good that I should ever want the prospect of a day when I will not be loved. However, awareness that I, and the people who love me, are mortal would seem to make that inevitable. It is natural that I should want a shelter, and that I want my shelter to be durable. If I am satisfied with a shelter that I know will fall down, then I am too easily satisfied. However, my awareness of time and history leads me to realise that
The cloud-capp'd towers, the gorgeous palaces,
The solemn temples, the great globe itself,
Yea, all which it inherit, shall dissolve
And, like this insubstantial pageant faded,
Leave not a rack behind.
The awareness of time that we have is good from an evolutionary perspective: it enables long term projects. The desire for durability is good: it stops us from being satisfied with shoddy workmanship. Put these two traits together and you have a desire for a building that will last forever. That seems like a transcendent desire. Or, even if we don't class this as a transcendent desire, it certainly seems like a desire that cannot be satisfied in the natural order, and our awareness of time could give any natural desire this quality: I want not merely to be loved, but to be loved always, not merely to leave behind offspring but to leave behind an eternal monument and so on.

I do not assert with confidence that the desires so described are the transcendent desire of which Lewis wrote: I do not myself like the thought that the speech from The Tempest I quoted above is merely Shakespeare's way of saying that he would like to build a more durable shelter. But, and this is what counts against the AFD, one can, in this way, form many desires that cannot be satisfied in the natural order of things, so that by continually striving to achieve more than is possible, we manage to achieve the very best that can be done.

It is appropriate to close by saying that I find the discussion so far, including my own comments, unsatisfactory. I say that as a positive, hopeful statement.

Trent, thanks for responding to my first criticisms. To be sure, I missed the important quantifier at the beginning of premise 2 ("most") such that adducing a counter-example, even a legitimate one, doesn't nullify the AFD as presented. I also think you made an important point - demanding a fully worked out metaphysic in order to make the argument might be a little too much. My goodness, we might never get around to making any arguments that way! Defining the key terms through ostension may prove adequate.

However, the real core of my intended objection lies in a contention that the AFD as so far presented lacks sufficiently rich premises to establish its conclusion (namely, that there either is something transcendent, or that it is likely that there is something transcendent). I understand that the conclusion is meant to be light - you're not trying to prove theism or any such position via the AFD, you're merely clearing the way, so to speak, for a non-naturalistic account of the world. The dialogue between you and Mike led me to appreciate the fact that this argument would probably be part of a holistic apologetic rather than standing alone (i.e., "So, here's another set of reasons why a non-naturalistic account of the world would better fit human experience."). Nevertheless, I can envision plenty of naturalists agreeing with your premises as stated yet reaching naturalistic conclusions. The problem isn't just that the premises are vague, or defined by ostension, but that in their present form they aren't tight enough to secure an interpretation of them that leads to a non-naturalistic conclusion.

For example, an evolutionary psychologist might understand the desire to transcend as an attachment mechanism that's been selected for by the evolutionary process. While the mechanism's subjective dimension feels vague and "transcendental", it's really just a desire to be with other people which in turn promotes my survivability. Here, our make-believe evolutionary psychologist might believe both premises and yet stay safely within a naturalistic scheme.

I don't think it'd be impossible to enrich these premises to secure your conclusions - far from it. But the value of the argument cannot remain in its lack of commitment to firmer conceptualizations. Provided my assessment is correct, it can't stay at its current "moderately low level of definitional precision." It's premises at least need to eliminate possible naturalistic interpretations. One way to do this would be to elaborate what natural desires are, how humans came to have them (at least insofar as a genealogy of natural desire might rule out certain objections), and a firmer definition of transcendence.

As a side note, I'm not sure the first premise will hold out under empirical scrutiny even though you've mentioned "academic data plausibly taken to be in support of it." The ubiquity of religion, for instance, doesn't seem to me to necessitate a natural desire for transcendence - other psychic or sociological mechanisms could well be responsible for its universal development.

Good luck - I can't wait to read future installments,

Josh ;)

Ben, could you clarify the structure of your argument for me?  It seems that you are proposing a kind of naturalistic explanation of the desire for the transcendent.  Here is my top hypothesis as to how the explanation is supposed to go.

(N1) The desire for the transcendent is just an instance of the desire for the durable.
(2) The desire for the durable is explained without remainder by the adaptiveness of the desire.
(C1) The desire for the transcendent is explained without remainder by it's adaptiveness.

Is that how it's supposed to go?

One last, quick comment just so my example of an evolutionary psychologist, grabbed from Ben, isn't misunderstood. The example is possible, I think, because the premises are loose at the moment. If the terms 'natural' and 'desire' remain open, it's possible to effect a disjunction between a desire's object, its somatic or structural origin (if it has one) and it's subjective sense and/or intentionality. Per the example, the desire to transcend has as its object the community of other people (this is what satisfies it and how it achieves its aim) even while subjectively construed the agent feels a mystic union with the community. Again, this isn't an example of a counter-example defeating the argument, but rather an attempt to persuade you that the premises admit of legitimate interpretations that then thwart the desired conclusion.

Josh, thanks for the further comments.  It really seems to me you are running two separate issues together: 1. Clarity of terms in the premises, and 2. Truth of the premises.  You start by talking about Type-1 issues saying that the premises aren't "tight" enough.  If this doesn't mean "clear" then I'm not sure what you're getting at.  If it *does* mean clear, then I think I already addressed that concern.  Then you move on to Type-2 issues, offering what appears to me to be a naturalistic explanation of the desire.  You seem to present these as one point, but offering an alternative explanation of the data assumes a sufficient understanding of the data.  Believing the premises and denying the conclusion is perfectly consistent if you have an alternative explanation.  The way to proceed in explanatory arguments is not to provide *definitions* which "eliminate" alternative explanations, but simply to give *better* explanations of the data.  The way to eliminate an explanation is to provide a better one.  That there is a transcendent reality for which we were made is the supernaturalist's explanation.  It will eliminate all and only explanations it is better than and be eliminated by all and only explanations better than it.  I'll just consider the alternative explanation itself.

The naturalistic explanation you propose seems to be an instance of the same gambit I think Ben was suggesting.

(N2) The desire for the transcendent is just an instance of the desire to be with other people.
(2) The desire to be with other people is explained without remainder by its adaptiveness. 
(C1) The desire for the transcendent is explained without remainder by its adaptiveness.

I must say that (N2) doesn't sound very plausible to me.  The desire to transcend oneself is the desire for something of another *kind* not just a greater *degree* of the same old thing.  The desire adverted to in (1) of the AFD is manifestly not the desire to "hang out" for countably many days.  Perhaps I didn't emphasize enough the idea that having this desire makes one feel like one doesn't belong to this world alone.  I wonder too if the data set shouldn't be enlarged to include *belief* in the supernatural.  I've come to think belief and desire are not so easy to separate anyway.  I think it would and should strike Freud what a bright future this "illusion" had after the writing of "The Future of an Illusion." 

So perhaps the data set need to include belief as well as desire, however, that point hasn't come just yet for why think (C1) is evidence against (3)?  I mean, evolutionary theory doesn't strike most people, especially Catholics, as incompatible with providence, so why should it strike us as incompatible with implantation of desire for Him, Augustine's "resquiescat"?  Why should thinking evolutionary theory is incompatible with (C1) seem any less quaint than thinking evolutionary theory is incompatible with creation?

Trent, my amateurish attempt at evolutionary psychology is intended to show that there is a plausible naturalistic explanation of desires that cannot be satisfied in the natural order. If this is accepted, it raises a dilemma.

Either such desires are treated as separate from transcendent desires. In that case, they count against the claim that most natural desires can be satisfied. Of course, it might still be the case that more than half of our natural desires cannot be satisfied, but at least it helps the naturalist case and weakens the argument if examples of non-naturally satisfiable natural desires can be found.

Alternatively, all such cases could be treated as transcendent: the desire for an eternal home is a desire for a supernatural home, which is ultimately an expression of our deeper yearning for the beatific vision. In this way, we could maintain that all of our desires may be satisfied naturally apart from one desire, the transcendent desire. However, in identifying the non-satisfiable desire for housing with the transcendent desire, we have admitted that there is a naturalistic explanation for our having this one unsatisfiable desire - the transcendent desire simply becomes the sum of all of our unsatisfiable desires.

This is a difficult question to decide, I think, because we are approaching desires from two perspectives. On the one hand, we have a theory that predicts what kind of desire a human being is likely to feel. On the other hand, we have the desires that we actually feel with all of their complexity and richness. I'm sympathetic to your suggestion that we have desires for a different kind of thing, not merely more of the same thing. Certainly, if I had offered Shakespeare a theatre that would never fall down, and he had said 'Good, the yearnings I expressed in The Tempest are satisfied', I would have been disappointed; I thought his longings were much more profound. But even if we grant that a transcendent desire is not a desire for more of the same, still our many desires for more of the same would be examples of non-satisfiable desires. I think the evolutionary psychologist would do well to provide argue for a link between 'more of the same' unsatisfiable desires and the transcendent desire that falls short of identity.

I hope that answers your question - of course, it also seems to raise many others. I look forward to your response.

When I offered the case for unsatisfiable natural desires, I said that I was playing devil's advocate. However, I think that this devil's position might have angelic support, if
there is a path from unsatisfied desire to God that does not involve accepting that most of our natural desires are satisfied.

An example of following this path can be found in T. S. Eliot's Murder In The Cathedral. There Thomas a Becket is visited by three tempters who offer him ways to escape his impending death, which he rejects. The fourth tempter offers the path of martyrdom as a temptation, leading Thomas to conclude, eventually,

The last temptation is the greatest treason
To choose the right deed for the wrong reason.

One wrong reason for choosing martyrdom would be to achieve the posthumous blessing of history, and one can understand why Thomas has this desire: it fits the pattern of evolutionary psychology I outlined above. He has a natural desire to be respected and praised by others (thus achieving a dominant position), and it is natural that he should want to be praised and revered even when he is not present (extending his dominance beyond his physical presence). The awareness that life will go on after he is dead naturally leads to a desire to be revered and praised in this period, and to a desire that this reverence should never cease. However, the same awareness of time that makes him capable of desiring posthumous glory also makes him aware that eventually, the judgement of history will change and, as the tempter reminds him
...later is worse when men will not hate you
Enough to defame or to execrate you
But pondering the qualities that you lack
Will only seek to find the historical fact
When men will declare that there was no mystery
About this man who played a certain part in history.

This leads Thomas to ask:

But what is there to do, what is left to be done?
Is there no enduring crown to be won?

To which the tempter responds:

Yes, Thomas, yes, you have thought of that too
What can compare with the glory of saints,
Dwelling forever in the glory of God...
Think of the miracles by God's grace
And think of your enemies in another place.

Of course, Thomas realizes that this too is a temptation: the desire for a heavenly crown, as described here, is not a true desire for the beatific vision, rather it is a perverted desire that will lead him to hell rather than heaven, and so he has to go through a final purification before he is ready truly to embrace martyrdom, by renouncing desire.

Now, the more I think about it, the more I agree with Trent's comment that the kind of transcendent desire evoked by Lewis is not more of the same. Lewis would have recognised the vainglorious desire for sainthood that Thomas rejects as the corruption of something holy that it is. However, in following this path, we are having to point out that temptations often involve desires that cannot be satisfied, and that human nature, left unchecked, will lead us down the path of these desires. Once we point that out however, we cast some doubt on premise (2) of the original argument.

I'm aware that the objects to satisfy a desire can exist even though it is impossible to satisfy the desire, and premise (2) states only that the right objects exist. That is why I limit myself to casting some doubt on premise (2). If nothing else, this gives me a reason for quoting Murder In The Cathedral, which is no bad thing.
I should add that, although I'm not persuaded that the argument works, I certainly have been persuaded that it has been unjustly neglected, and that there is plenty of good to be done by writing a serious academic study of it.

Inateness is the truth all of these theories. Everyone knows what is "right" and what is "wrong" Our capacity to to enable action over knowledge makes the integrity find a marker. The scale of "correctness" has many bars. Where we as individuals are able to fall guides society as a whole. Bad childhoods, abuse, or complete acceptance and support certanly play a part, however many abused people accomplish great deeds, and many fortunate children do not. Each individual has capacity, how much of that inate ability they use determines the fate.

As to heaven, I think how we are thought of determines our fate. Energy is constant. If it is encouraged it will flourish, if not it will disipate.

In short... be a good person, use the golden rule. Keep it simple stupid:)

Mike Scottsdale AZ

Ben, thank you for the follow up. 

So far, I've not heard a reason to accept that there is a plausible naturalistic explanation of the desires that cannot be satisfied in the natural order.  Since your dilemma is premised on this, that makes it hard for me to evaluate that dilemma.  I will do my best to address the concerns you raise, but I wish to make clear that the primary job of the naturalist is to offer a concrete alternative explanation.

You use the terms "transcendent desires" and "desires that cannot be satisfied in the natural order".  If you mean for these to be disjoint sets you'll need to tell me how they are different.  They sound like different names for the same set to me but they could reasonably be names for different sets. 

I'm very confused about the third paragraph.  You write:

"the desire for an eternal home is a desire for a supernatural home, which is ultimately an expression of our deeper yearning for the beatific vision."

Then in the very next sentence you write:

"In this way, we could maintain that all of our desires may be satisfied naturally apart from one desire, the transcendent desire" (emphasis added).

I'm not seeing the "in this way" part.  Is the idea that there is only one transcendent desire, namely the desire for the beatific vision, and that all other transcendent desires are merely aspects of this one desire?  If so, I'm fine with that.  I'd like to say that is true, though it's not part of the argument, nor do I see how it affects it.  In particular, when you say:

"in identifying the non-satisfiable desire for housing with the transcendent desire, we have admitted that there is a naturalistic explanation for our having this one unsatisfiable desire"

I just can't see what you mean.  What's the naturalistic explanation we've admitted.  I don't see it, sorry.

I really like your Shakespeare illustration.  It would indeed be surprising and disappointing if the great poet were that easily satisfied.  I admit that the link between ordinary desires and transcendent desires seems like a promising line to track for the naturalist: I would like an apple.  I would like two apples even better, three more yet... but I think your Shakespeare illustration effectively neutralizes that approach unless someone can work it out in detail.  I haven't been able to do that myself. 

An important point is that our desire for more of the same is not unsatisfiable in the sense I intend.  For almost any physical object of desire, if nature includes one, it includes more than I could ever consume in a human lifetime.  I'll try to make the relevant notion of satisfiability more clear but it has to do with whether something is *intrinsically* unsatisfiable in the natural order.  The desire for more apples than I could ever want is not *intrinsically* naturally unsatisfiable, anyone who's had the experience of sehnsucht knows that the object of that desire is *intrinsically* such as to be beyond the natural realm.  I see now that it is important to make this more clear, so even thought I don't see a naturalistic explanation in the neighborhood, I do see a new important problem: clearly defining the notion of satisfiability.  Any suggestions are warmly appriciated.

I also like the use of Eliot.  However, when you write:

"temptations often involve desires that cannot be satisfied, and that human nature, left unchecked, will lead us down the path of these desires. Once we point that out however, we cast some doubt on premise (2) of the original argument"

I'm unable to see how the doubt is cast.  Can you spell that out for me?

Finally, if our discussion brings more attention to the AFD our primary desire will be satisfied!a

Trent, I'm taking 'desires that cannot be satisfied in the natural order' to be a set of desires that fit a definite description. Lets term this {N}.

'Transcendent desire (s)' in this context is probably best understood as a rigid designator that picks out either a single desire that many people share, or a certain kind of desire that many people experience. I think we can agree that on certain occasions it is clear that someone is describing having a transcendent desire, without being confident that, as yet, we have any clarity about the essence of such a desire. It is because of this unclarity about the essence of such a desire that I leave the question about the plurality of transcendent desires undetermined. (We might decide that Taoists experience transcendent desire and so do Sufis, while being unsure whether to say that the desire of the Sufis is the same as the desire of the Taoists). However, for simpicity, let's assume a single transcendent desire, T.

I take it that T is an element of {N}. The question is whether there are many more elements of {N}. The evolutionary psychologist claims that there are, the desire for an everlasting house, for an endless supply of affection etc. If we admit many members of {N} other than T, that weakens premise (2), which relies on our saying that for natural desires other than T, they are satisfiable.

A large part of the problem, as you point out, relies on the entry conditions for {N}. What is the relevant sense of 'can' in 'can be satisfied'?
Thomas has, and ultimately rejects, the vainglorious desire for false sainthood. Shakespeare, we agree, desires more than a permanent theatre, but his lesser disiple, Speareshake, would be satisfied to know his theatre is indestructible.

Your argument, as I understand it, is that the objects so desired exist (life, fame, power, theatres), just not with the desired property of pemanence. So, the relevant sense of 'can' is starting to look something like this: for most natural desires other than T, the desire is for a state of affairs that involves a combination of objects and properties such that every object is of a type that exists, and property is a type that is instantiated, even if the combination so desired never occurs.

So construed, I don't think the natural desires I've described can pass as members of {N}. This is, of course, a major point in favor of the argument: these desires do not constitute a counter-example to (2).

However, re-reading my previous contributions, a point that I was edging towards but did not state explicitly is that this way of establishing membership of {N} might seem ad hoc to the evolutionary psychologist.

After all, it is arguable that anything that is clearly conceivable or describable passes the combinatorial test I've described. It is not surprising then, says the evolutionary psychologist, that the only desires for which objects don't exist are those that transcend the limits of normal life, since not having objects of satisfaction is logically equivalent to transcending limits of normal life.

Furthermore, when people are taught to avoid temptations, they are told, 'This desire cannot be satisfied', and the desire in question is a desire that the evolutionary psychologist was just arguing should be included in {N}. Now granted, there are legitimate senses of 'can': I can travel around the world in eighty days and then again (economically) I can't. So in desiring everlasting posthumous fame Thomas desires something he can have (logically) and can't. But, even though it is a logically impeccable move, the jump from saying to the evolutionary psychologist 'You can have this' to Thomas's recognition 'I can't possibly have that' creates a certain tension. Or maybe not. It is hard to find an objective way of settling issues of ad hocness, and you might not think there is any tension at all.

I do have more ideas, but I need to wait until another time to combine the ideas with the time.

Ben, your explicatioin of the membership criterion and the combinatorial principle are very much what I had in mind, and I agree that this avoids the counterexamples proposed.

I don't think there's anything ad hoc about it thought (not that I know what "ad hoc" even means here). It just take it to be a datum that we do have such a desire and the reference class just is what it is as a consequence of the nature of the desire. If it were a different kind of desire, it would have a different relevant reference class.

The Humean principle that all which we can concieve is generated combinatorially out of our experiences doesn't have much going for it as far as I'm concerned, but even if it was true, I think it's intelligible to say that the inarticulate desire is for something for which no known combination can suffice.

I find such a desire quite strange and want to know why I and others have it. The simplest explanation I can think of is that, just like our other spontaneous desires universal in temporal and spatial extent, there is something corresponding to that desire for which we are "made". We're open to alternative explanations, but so far, I've yet to recognize an alternative naturalistic explanation. I'm very very glad you pushed the combinatorial line, because I think when one realizes that nothing of that sort could satisfy this desire--the nature of the desire and its strangeness and its resistence to naturalistic explanation really comes through.

I am not sure how much this will effect the argument but one thing you might condsider is the way in which desires are satisfied. With natural desires such as food and saftey once I have them I am no longer desiring. I no longer desire food if I have just eaten. Are these truly desires or just needs that become desires when there is a lack?

Other desires though are not satisfied in this manner. Love is never satisfied. I do not stop desiring the beloved simply because I am with him. Rather the desire continues but feels no lack as long as it is met and returned. In other words these are true desires as opposed to those that are caused by physical needs. Maybe this is a way to define transcendent vs natural desires. My desire for sex is natural. Once the physical act occurs the desire fades. It can be satisfied in time. My desire for love is transcendent because it never ends. If my desire is thus eternal then only eternity can satisfy it. Our desire for recognition or respect falls into this transcendent category too. As does our desire to live. In fact I would suspect that what God offers us-love, life, recognition, knowledge etc. are all transcendent desires of this type, that are satisfied not by being filled but by being equaled.

Hi guys! I too have long thought that this was an interesting argument that has received little treatment. I'm glad to see that others think the same.

It seems to me that there are bunches of unsatisfiable natural desires. We often desire that we could relive various parts of our lives or that our kids were young again. These seem to be natural in that they naturally and commonly occur in (I'm guessing) quite a number of humans. Now, as you say, one (or a few) counterexample(s) does not refute an inductive argument, but there being a few counterexamples does invite one to ask whether the case in question is more like the cases where the object of desire exists or more like the cases, such as I've mentioned, where it does not. If its more like the latter, then an inductive argument given all of our evidence will be inconclusive.

So, which is the desire for transcendence more like? What seems to be going on in the above examples I described is that the desires arise as ways of expressing regret or fondness or something of the like. In fact, upon reflection we tend to recognize this; we recognize that these desires can’t really be satisfied. So our possession of these unsatisfiable desires is, plausibly, explained by our need to firmly express certain emotions towards an event, fact, person, place, or thing. Could our desire for transcendence be explainable in a similar way? Well, perhaps this desire of ours is explainable by our firmly feeling that the world isn’t quite just or right. I don’t know if this is a good explanation for our desires, but it is one on which the object of our desires (transcendence) does not exist and so needs to be dealt with if the inductive inference is to go through.

There are other examples of relatively natural desires that are unsatisfiable, such as a man’s desiring to care for his child when he doesn’t actually have a child – he’s been deceived into thinking that he does in some way. Such desires are unsatisfiable, but we can explain why he has an unsatisfiable desire by the fact that he has a base disposition to care for his offspring should he have any together with the fact that he mistakenly believes that he has a child. Now, the question with the desire for transcendence is whether it can be explained by a basic disposition together with a mistaken belief (or, to be more neutral, a belief that may turn out to be mistaken). Here’s a candidate for such an explanation: we have a disposition to desire to be in good states, which when coupled with the belief that there is a transcendent state that is supremely good, gives us a desire for transcendence. Where do we get such a belief (we can set aside whether it is true or false at this point)? Perhaps we absorb it from our surrounding culture. So, this is an explanation of our desire on which it needn’t turn out that our desire has an object. Indeed whether it does depends more upon whether the belief that helped to generate our desire was warranted or justified. But, then we can’t use the argument from desire to conclude that belief in something transcendent is justified because it presupposes that our belief in something transcendent is justified (i.e. the arg. from desire turns out to be circular).

So, it seems to me that to get the argument from desire to go through we need to see whether these alternative explanations are as good as the explanation that it is implanted into us by the transcendent thing, whatever it is. The more general point is that once we see that there are varieties of unsatisfiable (relatively) natural desires, and we see how the having of these desires gets explained, we see that it isn’t at all clear that our desire for transcendence can’t be similarly explained, and so we need to do some work in order to show that the desire for transcendence isn’t as like those cases as it might seem.

What do you guys think?