Avicenna’s flying man

Peter Adamson’s new book Ibn Sīnā (Avicenna): A Very Short Introduction is an excellent primer on the great medieval Islamic philosopher.  After a biographical chapter, it treats Avicenna’s views on logic and epistemology, philosophical anthropology, science, and natural theology, and closes with a discussion of his influence on later philosophy and theology.  Among the things readers will find useful is the book’s discussion of Avicenna’s famous “flying man” argument.  Let’s take a look.

The flying man thought experiment is one of the means (not the only one) by which Avicenna aims to establish the incorporeality of the human soul.  He presents it at the end of the first chapter of his treatment of the topic of the soul in his work The Cure.  One place you can find the relevant passage is Jon McGinnis and David C. Reisman’s anthology Classical Arabic Philosophy, which translates it as follows:

For the purposes of establishing the existence of the soul… [I]t has to be imagined as though one of us were created whole in an instant but his sight is veiled from directly observing the things of the external world.  He is created as though floating in air or in a void but without the air supporting him in such a way that he would have to feel it, and the limbs of his body are stretched out and away from one another, so they do not come into contact or touch.  Then he considers whether he can assert the existence of his self.  He has no doubts about asserting his self as something that exists without also [having to] assert the existence of any of his exterior or interior parts, his heart, his brain, or anything external.  He will, in fact, be asserting the existence of his self without asserting that it has length, breadth, or depth, and, if it were even possible for him in such a state to imagine a hand or some other extremity, he would not imagine it as a part of his self or as a necessary condition of his self...  Thus, the self whose existence he asserted is his unique characteristic...  Thus, what [the reader] has been alerted to is a way to be made alert to the existence of the soul as something that is not the body – nor in fact any body. (pp. 178-79)

The basic idea of the thought experiment is as follows.  A man who comes into existence in the bizarre circumstances Avicenna describes would have no sensory experiences.  For one thing, he has from the start somehow been suspended in midair, in a manner that does not involve even the air pushing against him – perhaps by miraculous divine action.  Hence he has never experienced external physical objects exerting any pressure on his skin.  Because his arms, legs, fingers, etc. are all spread out away from one another, he also has not felt even his own body parts pressing against him.  Because he is veiled (presumably in a manner that does not involve a veil making contact with his body) he has never seen anything.  Presumably his ears, nose, and tongue are similarly prevented from being affected by any stimuli.  Hence he has no awareness of any physical object, not even his own body.  As Adamson notes, while some might object that such a man would still have proprioceptive experiences of his limbs, it is not difficult to extend the thought experiment in a way that would prevent that.  We could imagine, for example, that the miraculous suspension of the normal operation of the relevant nerves is a further part of the situation. 

Now, the man would, Avicenna claims, nevertheless have awareness of himself.  He would know that he exists, even though he would not know that his body exists, and indeed would not know that any physical world at all exists.  In that case, though, he must be distinct from his body and from anything corporeal.  For if he were corporeal, how could he know he exists without knowing that anything corporeal exists?

Parallels to Avicenna?

I’ll come back to some of the remarks Adamson makes about the argument, but first let me make some observations of my own.  Avicenna’s argument might seem similar to arguments later developed in the Cartesian dualist tradition.  For example, in his Sixth Meditation, Descartes argues that he could in principle exist without his body existing, if God willed to create him that way.  And in his book Engines of the Soul, W. D. Hart argues that it is possible in principle for a person to have visual experiences while lacking a body, in which case it is possible for a person to exist without a body.

However, Avicenna’s argument is importantly different, in several respects.  First, Avicenna emphasizes that the man in his thought experiment has had no sensory experiences at all.  By contrast, Hart’s argument involves a disembodied person who does have such experiences.  And at least earlier in the Meditations, in Meditation One, Descartes suggests that it is possible for someone to have sensory experiences even in the absence of the existence of his body or of any material world at all, if a Cartesian demon caused a disembodied mind to hallucinate. 

Second, the key premise of Avicenna’s argument is epistemic, whereas the key premises of the Cartesian arguments mentioned are ontological.  Descartes and Hart start with the idea that it is possible for the self to exist without the body, and conclude from that that the self is distinct from the body.  Avicenna starts with the idea that one can know the self without knowing the body, and concludes from that that the self is distinct from the body.

Third, and relatedly, the thought experiments Descartes and Hart appeal to presuppose that the self could in fact exist apart from the body, whereas Avicenna’s thought experiment does not.  That is not to say that Avicenna doesn’t think the self could survive without the body, but only that that would be a further conclusion of the argument rather than a presupposition of the argument. 

The reason these differences are important is that they make Avicenna’s argument immune to certain objections that might be raised against Descartes and Hart.  First, one might question the assumption that sensory experience really is possible without a body.  If that assumption is wrong, then Hart’s argument will fail (though whether Descartes’s argument would fail will depend on how seriously Descartes wants us to take the Cartesian demon scenario).  But Avicenna’s argument makes no such assumption. 

Second, because they presuppose that it is possible for the self to exist apart from the body, Descartes and Hart might be accused of begging the question.  They are trying to get from the possibility of the self existing apart from the body to the real distinction between self and body.  But a critic can object that the claim that it is possible for the self to exist apart from the body presupposes that there is a distinction between self and body, and thus can hardly cogently be appealed to in order to establish such a distinction.  Avicenna is not open to such an objection.

If we are looking for arguments from the tradition that are similar to Avicenna’s, it seems to me that a more plausible parallel is to be found in some arguments earlier than his, which were developed by St. Augustine in On the Trinity.  Augustine held that the mind can know its own essence with certainty, but does not know with certainty that corporeality is part of its essence.  Hence, he concludes, corporeality is not part of the mind’s essence.  He also held that the mind can know itself without the mediation of any imagery, but cannot know material things that way, and concluded that the mind must not be material.

Augustine’s and Avicenna’s arguments are similar, then, in starting with what the mind knows or doesn’t know about itself and about material things, and from this epistemic premise drawing a conclusion about the distinction between the mind and anything material.  The key difference is that Avicenna appeals to a novel thought experiment in order to make his point about what the mind knows.

Some objections

As Adamson notes, one objection that can be raised against Avicenna’s argument would be to deny that the flying man really would or could know of his own existence.  One could hold that it is only after the mind has had some perceptual experiences that it comes to know itself, by way of reflecting on those experiences.  Note that one can hold this on the basis of the moderate empiricism of Aristotle and Aquinas, without committing oneself to the more extreme modern empiricism of Locke and his successors.  And as that indicates, one could hold this without rejecting Avicenna’s conclusion that the mind is incorporeal, but only the flying man argument’s particular way of arriving at that conclusion.

Adamson also notes that Avicenna’s argument has to be understood in light of his broader epistemological commitments, which include the thesis that the self is always at least tacitly aware of itself.  I find these broader commitments dubious, but for present purposes will simply note that the need to defend them in order to get the flying man argument off the ground at the very least makes it a considerably less punchy argument than it might appear to be at first glance.

Another objection noted by Adamson is that to know one’s self without knowing one’s body does not by itself entail that the self is different from the body, any more than the fact that Lois Lane knows that Clark Kent is at the Daily Planet without knowing that Superman is there entails that Clark Kent is different from Superman.  Adamson suggests that one way Avicenna could reply to this would be to argue that to know a thing’s essence, specifically, requires knowing its essential constituents.  If we say that the flying man knows his essence while not knowing anything about his body, then the body cannot be among the self’s essential constituents.

This interpretation of the argument underlines its parallels with Augustine’s arguments.  I refer the reader to my discussion of those arguments, which is not unsympathetic even though they are not my own preferred way of establishing the mind’s immateriality. 

Related reading:

Avicenna’s argument from contingency, Part I

Avicenna’s argument from contingency, Part II

Avicenna on non-contradiction