Conceivability, i.e. the capability of being conceived or imagined, is an important device employed by many philosophers - Descartes, Hume, and Chalmers, to name just a few - in developing their arguments about modality.  Is conceivability a guide [1] to possibility?  I want to argue that it is not; at least, I want to show that one argument that it is such a guide - that put forth by Steven Yablo [2] - fails to establish its conclusion.  In doing so, I will argue that there exists a circularity in the argument that whatever is conceivable is possible, in a way that parallels Joseph Butler's circularity objection to Locke's memory theory of personal identity. [3]   Specifically, I wish to argue that Yablo's argument for conceivability as a merely defeasible guide to possibility fails to address the essence of the circularity between conceivability and possibility.  In other words, even if conceivability is indeed only a defeasible guide to possibility, Yablo's argument still falls victim to the circularity.  Moreover, I want to point out an improper analogy between modal and perceptual intuition, an analogy that Yablo attempts to exploit.  Finally, I suggest a possible way to avoid the circularity by appealing to an objective sense of modal intuition.

 

Conceivability Evidence

Hume's famous maxim, stated in his Treatise, originated the issue of whether conceivability is a guide to possibility:

 

'Tis an established maxim in metaphysics, that whatever the mind clearly conceives includes the idea of possible existence, or in other words, that nothing we imagine is absolutely impossible.  We can form the idea of a golden mountain, and from thence conclude that such a mountain may actually exist.  We can form no idea of a mountain without a valley, and therefore regard it as impossible. [4]

 

What does Hume mean here by possibility?  There are a number of different kinds of possibility, including physical possibility, logical possibility, epistemic possibility, and metaphysical possibility.  Here Hume clearly seems to have in mind metaphysical possibility, or so-called ontological possibility.  Here's a rough definition of metaphysical possibility:

 

(MP) Something is metaphysically possible if it is consistent with the laws of logic as well as fundamental ontological principles. [5]  

 

For example, that an object is green all over and red all over is logically possible but ontologically impossible, because the ontological laws tell us that both red and green are colors and no objects can have both colors all over at the same time.  So, metaphysical possibility is stronger than logical possibility (it is governed by the laws of logic as well as additional principles) but weaker than physical possibility (something can be metaphysically possible, but impossible given the laws of nature).

How, then, should we define conceivability?  I plan to postpone offering a positive account at present, and to return to the matter at the end of the paper as part of my suggestion for how to proceed in this matter.  However, it is important to distinguish conceivability from possibility in their philosophical sense.  Conceivability is an epistemic notion concerning what can be thought, while possibility is an ontological notion concerning how things can be. [6]   If this is right, then we must take care to distinguish conceivability and possibility as different sorts of features of propositions, [7] such that one does not, at least not automatically, entail the other.  That is, that p is conceivable cannot suffice to show that p is possible; there must be some premise bridging the epistemic and the ontological in order for such an argument to be valid.

 

The Circularity Objection

There seems to exist some ambiguity regarding conceivability and possibility in Hume's maxim.  For, the question might arise, concerning some particular mental state, whether it includes the idea of possible existence and thus whether it counts as an act of conceiving. [8]   If it does not include the idea of possible existence, then it is not genuinely a conceiving, and so cannot provide evidence of possibility.  How, though, do we know whether the mental state includes the idea of possible existence?  Presumably, by way of some independent criterion of possibility.  But this suggests that Hume's maxim is circular: he has offered conceivability as a guide to possibility, but conceivability itself presupposes possibility. 

Interestingly, this same sort of circularity can be found in Locke's memory criterion of personal identity.  In "Of Identity and Diversity," [9] Locke proposes a memory criterion that can be construed roughly as follows:

 

(L1) Person P₁ at time T₁ is identical with person P₂ at some earlier time T₂ if P₁ has memories at T₁ of P₂'s actions or thoughts at T₂.

 

Joseph Butler then suggested, in "Of Personal Identity," [10] that (L1) is flawed because whether a present mental state that represents a past mental state counts as a genuine [11] memory or not depends on whether both mental states were had by the same person.  Thus, Locke's theory is circular: he defines personal identity over time in terms of memory, but memory itself presupposes personal identity. 

Yablo offers a statement of the circularity between conceivability and possibility as follows:

 

(C1) Conceivability is a guide to possibility only as constrained by prior modal information tantamount to the information that p is possible. [12]

 

Yablo then goes on to suggest four distinct kinds of circularity objection that evolve from (C1), among which argument D is considered the strongest.  Accordingly, I shall focus on this version of the objection; call it Yablo's type D circularity objection:

(YD) If all it takes to find a proposition conceivable is to be unaware that it is impossible, then since impossibilities go unappreciated all the time, they are just as often conceivable.  Before relying on conceivability evidence in any specific instance, then, you need a reason to think that in this case, p's conceivability signifies that it is possible rather than that, although it is impossible, you are unaware of this.  That is, you need a reason to deny that

 

            (*) although you are unaware that p is impossible, p is             impossible.

 

Because (*)'s first conjunct is true, and known to be-you are unaware that p is impossible-you can be reasonable in denying (*) only if you are in a position to deny its second conjunct.  But its second conjunct is that p is impossible!  So you must already know that p is possible before you can conclude that it is from its conceivability. [13]

 

Yablo's Response to the Circularity Objection

How should we respond to the (YD) circularity objection?  Yablo argues that this strong form of the circularity objection does not do justice to Hume's maxim, in that conceivability need not be a perfect guide to possibility; conceivability arguments, Yablo maintains, do not have to be completely infallible.  All the Humean needs to claim is that conceivability arguments for possibility are sufficiently reliable to permit us to draw prima facie conclusions about possibility.  Of course, conceivability evidence per se does not always justify possibility, but as long as there exists a generalization that is not too weak, then conceivability can be a guide to possibility.  In other words, conceivability is only a defeasible guide to possibility. 

To support his argument, Yablo offered as a challenge to the circularity objectors to the Humean view that they have too much confidence in thinking that propositions whose impossibility is unappreciated are almost always conceivable, because they "mistook conceivability for the believability of truth" as well as "mistak[ing] it for the believability of possibility." [14]   Thus Yablo distinguishes what is presented as possible from what is possible:

 (Y) To conceive or imagine that p is ipso facto to have it seem or appear to you that possibly p.. Just as someone who perceives that p enjoys the appearance that p is true, whoever finds p conceivable enjoys something worth describing as the appearance that p is possible.. In slogan form, conceiving involves the appearance of possibility. [15]

 

Apparently, as will become important a bit later, he distinguishes the appearance of possibility from possibility for an even more important reason: to establish an account of modal intuition that is on a par with our faculty of perception.  

 

Reply to Yablo

I want to argue that Yablo's response does not avoid the circularity objection.  To know that conceivability is a trustworthy guide to possibility even sometimes, and in fact even a single time, requires an independent criterion of possibility, and thus we're back in the circle.  For, to know of even a single instance of conceivability indicating possibility, we must know that some p is both conceivable and possible; but this requires that conceivability and possibility be independently ascertainable.  So, if conceivability is meant to be our criterion of possibility, then the argument to establish that it is a reliable guide - even a defeasible one - is circular.  And, of course, if we have some other criterion of possibility, then there is no need for conceivability to act as a guide.  So, Yablo does gain some argumentative strength by maintaining that conceivability is merely a defeasible or prima facie guide to possibility; for, he is now immune to the problem that we are at least occasionally wrong in the use of our faculty of conceivability, and he merely needs to argue that conceivability is usually (but not necessarily always) a guide to possibility.

            It might be illuminating to compare this exchange to the epistemological debate with the skeptic about the external world.  Since Descartes, epistemologists have responded to the skeptic by insisting that sense perception need not be an infallible guide to the external world, and thus defended themselves against arguments based on the fact that our senses sometimes deceive us.  That is, most of us disagree with Descartes when he claims that it is unwise to trust our senses, given that they have deceived us on occasion. [16]   Sense perception can be a guide to the way things are, even if it's only a defeasible guide.

            This does indeed provide for a response to the skeptic who insists on certainty; but it does not help us avoid a circularity similar to what we have seen above.  For, in claiming that our senses at least sometimes - even if not always - provide evidence of the way the external world is, we need to point to an instance in which things seem a certain way and they in fact are that way.  But, of course, that requires some independent access to the way the world is, other than the way it seems to be, and we are back in the skeptic's circle.

 

Yablo's Possible Reply to My Objection

I think that Yablo's discussion indicates how he might respond to this type of objection.  He would likely reply that the analogy with sense perception is indeed illuminating, and that it in fact suggests a new line of reply to the circularity objection.  He argues that modal error is "a fact of life," [17] and that the real issue is not simply that our faculty of conceivability is fallible, but whether this fact can lead to an argument that defeats our claim to be at least prima facie entitled to regard whatever we find conceivable as metaphysically possible.  The objector's challenge, in other words, is to "find a defeater q of the conceiver's modal intuition." [18]   And, accordingly, Yablo's burden, as a proponent of the claim that conceivability is a guide to possibility, is to offer some positive, non-circular grounds for thinking that our faculty of conceivability is trustworthy.

The backbone of Yablo's line of argument to this effect resides in his analogy between modal intuition and perception.  Yablo suggests, that is, that our modal faculty of conceivability is importantly and relevantly similar to our faculty of perception.  As he suggested earlier, "if the worst that can be said about conceivability evidence is that it is as bad as perceptual evidence, that may be taken as grounds for relief rather than alarm." [19]  

 

My Further Objections to Yablo's Reply

How sound is Yablo's analogy of modal intuition to our faculty of perception?  I do not think that it is very sound.  Although I have not here presented the details of Yablo's account, I think that, in fact, our contemplation on the essence of conceivability may well lead us to conclude that sense perception and modal intuition are quite different faculties, although they have similarities in some ways.  In other words, whatever Yablo's argument, there is an important disanalogy between the two faculties, and this will suffice to render his argument defeated. 

One obvious difference between conceivability and perception is that, while perception is necessarily derived from a subjective/first-person point of view, conceivability implies an objective/third-person perspective.  For, what I perceive and what you perceive will differ, based on the fact (among others) that we occupy different perspectives; but, presumably, this will make no difference with regard to what we can conceive.  To give an example: when I conceive the possibility that I could have been the purple towel in my bathroom, my modal intuition tells me:

(P1) I could imagine there is such possible world in which "I," Laura, in the form of a human body, had never existed.  Instead "I" am a purple towel in the bathroom of an apartment in Jonesboro, Arkansas. [20]

 

P1 comes from a sort of faculty of conceivability, but one which is extremely prone to modal error, of course.  But note that I find myself with another intuition, one that that contradicts P1:

 

(P2) I could not have been a purple towel, in that I know for a fact that a purple towel does not have a conscious faculty that fulfills the idea of self-identity.  If there is no such idea of "I" related to a purple towel then I could not have been a purple towel. [21]

 

Notice that in this case, there exists a counterbalanced force between P1 and P2 that is not seen when we execute our faculty of perception.  Of course, there are times when we cannot decide what we perceive.  For example, I cannot decide whether the person I saw in the dark the other night was a man or a woman.  But this is different from the case of P1 and P2.  What triggers P2 is my "intuition" gained from learning and knowledge through the years, while P1 is an example of, in Yablo's words, "natural intuition" (assuming there is such a thing), which we are born with.  That is to say, P2 seems to result from a faculty that resulted from knowledge and training in logical reasoning but that, in essence, is not innately possessed by anyone (and yet can be gained when someone has developed a sufficient level of reasoning ability). 

Thus the concept of modal intuition cannot be treated as the common concept of intuition we have in mind when we think of the analogy with sense perception.  Compare the relations between P1 and P2 to the case in which I could not decide whether my perception is right.  There is no such fundamental difference in the two perceptions I try to distinguish.  They are both the same type of perception, an inborn faculty of human beings that results from the sense organs. 

What I am attempting to point out is that Yablo's analogy of perception to modal intuition is flawed.  There is one species of perception, and it is a first-person affair.  But there are two species of modal intuition: the first-person sort, which is indeed analogous to sense perception but which is quite unreliable, and the third-person sort, which is more reliable but is not analogous to sense perception.  Furthermore, once this analogy fails, we will have to reconsider the plausibility of his reply to circularity objection.  In other words, the circularity objection still holds.   

 

Conclusion

In closing, let me suggest what we require from a positive account of the essence of modal intuition, in hopes that such an account could help us to understand the seriousness of the circularity objection to Hume's maxim.  If I am right, then we have learned that the analogy with sense perception is not the place to look for an account of a modal faculty which will provide reliable intuitions about possibility.  Instead, we need to focus on a modal faculty which is objective and shared, rather than a faculty which is indexed to some particular perspective.  So, if one wants to show that conceivability is a guide to possibility, one must defend the claim that this third-person objective faculty is something human beings all have.  To solve the circularity problem and argue that conceivability is indeed a guide to possibility, then, we need to begin with a clarification of modal intuition along these lines.

 

Bibliography

Butler, Joseph. The Analogy of Religion, Natural and Revealed, to the Constitution and Course of Nature. New York: Mark H. Newman & Co., 1847.

Chalmers, David J.  The Conscious Mind.  Oxford: Oxford University Press, 1996.

Descartes, René. Meditations on First Philosophy, in The Philosophical Writings of Descartes. Translated by John Cottingham, Robert Stoothoff, and Dugald Murdoch. Cambridge: Cambridge University Press, 1984.

Hume, David. A Treatise of Human Nature, 2^nd ed. Edited by Peter H. Nidditch. Oxford: Oxford University Press, 1978.

Kripke, Saul.  Naming and Necessity.  Cambridge: Harvard University Press, 1972.

Locke, John. An Essay Concerning Human Understanding. Edited by Peter H. Nidditch. Oxford: Oxford University Press, 1975.

Tidman, Paul. "Conceivability," in The Cambridge Dictionary of Philosophy. Edited by Robert Audi. Cambridge: Cambridge University Press, 1999.

Yablo, Stephen. "Is Conceivability a Guide to Possibility?" Philosophy and             Phenomenological Research, Vol 53, Issue 1 (Mar 1993), 1-42.