In some philosophy books I have encountered that conceivability of something is taken as an argument for its possibility.

In "The Conscious Mind" David Chalmers presents an argument that a world is possible physically identical to ours, but the people are zombies without conscious experience. The possibility arises as a consequence that we can conceive that world logically. I think this argument makes no sense. Here is an absurd scenario similar to it:

Imagine we are in the year 1210. I want to comprehend the nature of thunderlight. I put the argument that the light comes from a giant lamp hidden in the sky, where from time to time emits light. I can conceive this, so it is possible that this giant lamp exists. Now I start to build a whole theory about the lamps in the sky, etc...

I want to ask why conceivability is considered in philosophy even to sustain an argument for such possibility, and if you know some treatise that treats this idea of conceivability.

  • 3
    It's a good point that the notion of "possibility" is very vague. At least we can say this, however: if we can coherently conceive of a situation, then the situation is not logically self-contradictory. Chalmers is not saying p-zombies do exist - he claims that it's not a contradiction for them to exist. Then he says that, if it is not a contradiction for p-zombies to exist, then the definition of consciousness must be logically decoupled from intelligent behavior. (Whether consciousness is, in our world, actually decoupled from intelligent behavior is a separate matter.)
    – causative
    Commented Sep 8, 2021 at 2:59
  • 3
    Well, if consciousness by definition was logically coupled to intelligent behavior - so that wherever you have intelligent behavior, you must have consciousness as a matter of logical necessity - then p-zombies would be logically impossible by definition, because p-zombies are the presence of intelligent behavior without consciousness. So (taking the contrapositive), if p-zombies are possible, consciousness must not be coupled to intelligent behavior.
    – causative
    Commented Sep 8, 2021 at 3:40
  • 1
    See SEP, Conceivability that reviews Yablo's and Chalmers's theories concerning counterfactuals, and references there. On "conceivability entails possibility" arguments as applied to philosophical zombies and physicalism specifically, which involve many more participants, see SEP, Zombies.
    – Conifold
    Commented Sep 8, 2021 at 5:15
  • 2
    The conclusion of logical possibility is very weak: teapots orbiting Jupiter are conceivable and logically possible, but nobody thinks they exist. Logically possible may very well be what we call "absurd" or "ridiculous" in common talk. Chalmers's point is that even such weak conclusion is enough to make trouble for physicalism, because it severs consciousness from matter while physicalists hold that consciousness is merely a paraphrase of material happenings, necessarily tied to matter like shape to a statue.
    – Conifold
    Commented Sep 8, 2021 at 7:21
  • 1
    I've always been interested in arguments from conceivability, at times finding them incredibly persuasive and at other times finding them close to worthless. It took a discovery to show that the Evening Star and the Morning Star are both the planet Venus. So it was conceivable that "they" might have been two distinct entities. But self-identity seems to be a necessity, so although it was conceivable, surely it wasn't really possible ... was it? Possibility given the (actual) state of the world and possibility given the state of our knowledge seem to come apart. Commented Sep 11, 2021 at 10:02

3 Answers 3


Personally I have always had a problem with the use made of conceivability as a surrogate for possibility.

The reason for this for me is not that I think that actual conceivability wouldn’t imply some form of possibility between logical possibility and physical possibility (this usually being called metaphysical possibility) – I think it probably would imply this.

The issue that I have is that I don’t think people in fact can conceive things that aren’t physically possible. They may think that they are doing this, but that is because their conceptions are very vague. Example – people often talk about possible worlds in which the laws of physics are classical; we can shut our eyes and do some type of mental process that we may think equals conceiving of this. But at a level of detail, classical physics does not work (e.g. all the atoms would collapse); if you try to ‘conceive’ in full detail of a world in which classical physics is true, you won’t be able to.

I have never seen an example in the literature of an example of something physically impossible which I would agree that people can conceive in full detail. So I don’t think it is a thing.


Could you precisely construct semblances of all the people you interacted with today out of clockwork, and rerun your day? There you go, it's conceivable. Because you can conceive how it's possible. They aren't different in this case. In fact, it would be considerably simpler than constructing the entities we think we see, if we had to do so from scratch.

It's just a return of something very similar to Cartesian scepticism. The point is not that your body could be made out of glass, nor whether other people are zombies. The point is, how do you know what you think you know? How does the knowledge you feel you have about yourself, compare to what you know about the internal states of others?

One view goes, if it quacks like a duck it's a duck - in which case passing the Turing test means it qualifies as a mind. Basically no one is happy with that, unless it's hedged with caveats that basically say only a Turing test we can't define fully yet could demonstrate a mind.

At the other end of the spectrum, is the idea humans are imbued with magical mindliness, a soul maybe, & we need never worry humans have it even if braindead, and we can be sure no non-human has eg because we are made in god's image. Almost no one likes this either, & it struggles with aliens, & long term of evolution.

P-zombies help us examine positions in between these extremes.


This is a very interesting question, brings to mind firstly the idea that the pictures of objects in our mind are amalgamations of objects we've experienced directly. Kant said in critique of pure reason that one can define anything as long as it's logically consistent, I can define a Barney as a purple animal with horns and 10 legs. A barney is an amalgamation of the properties I've experienced directly in other objects. The reason I know the traits of purple , and the objects horns and legs is because I've experienced them before. Because I haven't traversed the entire universe, I don't know whether a Barney exists physically. So in that sense , conceivability is definitely a sufficient condition for possibility. Regarding conditionals, there is a lot of controversy about what they really signify.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .