Could a philosophical zombie conclude "cogito ergo sum"?



Could a philosophical zombie conclude "cogito ergo sum"?

Assume a philosophical zombie which is a hypothetical being that is indistinguishable from a normal human being except in that it lacks conscious experience, qualia, or sentience.

In particular, I mean a behavioral zombie that is behaviorally indistinguishable from a human, regardless of its internal machinery.

I guess it would probably insist there is no such thing as that silly thing philosophers call qualia, but that is OK since most of my good friends insist the same.

I wonder if such a zombie could conclude cogito ergo sum since I believe Descartes really meant (inner) experience entails existence, and that the purpose of I in that statement is simply to point out that first person, direct inner witnessing, rather than some concept of a self.

But, a zombie has no qualia, so...?

EDIT - to clarify, I do not question whether philosophical zombies can reach conclusions in general. I do not mind granting that even AI can be said to reach conclusions. I am asking if it can reach this particular conclusion.


Posted 2014-08-05T20:57:21.567

Reputation: 4 231

You said indistinguishable then it can. You said has no qualia then it probably can not. Your self Contradicting definition means that it is probably impossible to make such a zombie. Or to define well what human behavior actually is. Good news in both cases. Besides <-> great(real) humans were in non of the aspects normal - sO yEAH. – Asphir Dom – 2014-08-05T23:31:49.107

@AsphirDom, I do not know that a philosophical zombie is self contradicting, and I do not see how it follows from your argument. Some intelligent people deny cogito ergo sum, or qualia, so a philosophical zombie could as well, without being self contradicting. In addition, philosophical zombies are not made, but conceived. You can alternatively imagine a highly intelligent alien race lacking qualia. – nir – 2014-08-06T07:16:49.577

People denying qualia or cogito ergo sum are zombies. Difference is if they are real zombies or they can be awaken. Most of the religions and logical investigations (together with human desires) suggest that they must be not real zombies but rather sleeping beings. Kinda like kids who do not care about philosophy or existence. – Asphir Dom – 2014-08-06T13:04:32.187



Cogito ergo sum works, logically, in a generalized form that philosophical zombies can manage.

I have a detector of some sort, D(_). Does anything exist at all? If the detector is passed any argument, then I can say: yes! I might not know what it is, but I surely know it's there. Just one problem: I might imagine a fake argument and pass it instead. Do any non-imagined things exist at all? Oh, but wait, we have a detector, by premise. D(D(_)). Ta-da! We have passed an argument to D, therefore something exists, and it is in fact the detector that we have. I can't be faking D because it is acting as a detector and if it acts as a detector it is a detector. Also, since I am (hypothetically) a philosophical zombie, my D is sophisticated enough to know that that D is me.

I, as the zombie, can't conclude much of anything from this other than that the universe is not the empty set. (Nor can Descartes, actually.)

Rex Kerr

Posted 2014-08-05T20:57:21.567

Reputation: 15 388

how does doubt fit in this picture? – nir – 2014-08-05T21:45:09.230

@nir - It doesn't explicitly. Implicitly (or analogously), unless you pass something into D, the zombie doesn't know about it, so the zombie "doubts" that it exists. – Rex Kerr – 2014-08-05T21:47:44.670

I think you are talking about a recursive form of the ultimate machine invented by Marvin Minsky and Claude Shannon :)

– nir – 2014-08-05T22:12:31.717

I suppose the zombie puts all kind of other arguments, first. It puts in the image of a banana, and the detector says something exists, it puts in the sound of an alarm clock, etc... why would it bother putting D in the detector if it already "knows" something exists? – nir – 2014-08-05T22:23:43.157

@nir - I skipped some of the reasoning (where arguments might be imaginary); I've added it in an edit. – Rex Kerr – 2014-08-05T23:25:06.753

Cogito ergo sum is not logical. It is exactly about qualia. It is about qualia of existence. He understood that everything is real. In some sense it means that he got reconnected to "real" reality. Again -- nothing to do with logic, because logic has no qualia neither perception (which is same but who cares). He could have just said - I exist! But simple/sleeping folk would not get what it means. Existence is most important word in whole philosophy. – Asphir Dom – 2014-08-05T23:39:25.890

@AsphirDom - Is cogito ergo sum about existence or about qualia? They are not the same; Descartes uses a qualia-laden event to detect existence but I've shown a qualia-free one that does so in the same manner. Also, I'm not sure the empirical matter of whether it is possible to have a sense of doubt about one's own existence has been settled; it wouldn't be rational but we're not strictly constrained to be rational. – Rex Kerr – 2014-08-06T05:22:14.913

@RexKerr, why is something which acts like something not fake? can't an imagined voice act like a real voice? – nir – 2014-08-06T12:04:05.110

@nir - A detector that consistently returns true when you pass it itself is at least a self-detector by definition. That's all you need. – Rex Kerr – 2014-08-07T18:22:05.653


In Descartes context, the Cogito Ergo Sum is important because you don't necessarily know that anything exists, except yourself. Relates to the Evil Demon thought experiment, with the idea that even if this is the case, you know that you exist because you can doubt your own existence, and that doubt isn't a perception or qualia that can be forced onto you like essentially every other perception is.

This seems to break down in the case of a Philosophical Zombie. The Evil Demon argument doesn't really apply, because if the zombie isn't able to experience sentience and qualia anyways, it can have no doubt that the things it "perceives" are real. Thus I'd be inclined to answer that the zombie can't use Cogito Ergo Sum because it can't doubt the existence of anything, including itself. I guess the zombie knows it exists and has no need of a proof of it, because it cannot be tricked or illusioned the way a sentient human can.


Posted 2014-08-05T20:57:21.567

Reputation: 634

+1 for approaching the idea that a P-Zombie cannot conclude anything. – Magus – 2014-08-05T23:14:56.550

Why can't a philosophical zombie be tricked? Voice recognition software seems to be tricked all the time that I said something that I didn't. Phone support systems even seem to realize that being tricked is possible and sometimes repeat what I (supposedly) said to check. – Rex Kerr – 2014-08-06T05:24:12.973

@Magus, a p-zombie can conclude things by definition. – nir – 2014-08-06T07:18:12.120

@Cain, First, I think doubt is just a thought process, which an omnipotent demon should be able to make you believe you are having, just as with voices and images. Second, I do not see why qualia is required for having doubt corresponding to the information receives by the senses. – nir – 2014-08-06T08:13:47.963

@RexKerr The zombie can be tricked as to the identity of a thing, just not the exist of another thing. VR software can easily identify the wrong voice, but couldn't dream or hallucinate a voice without sentience. – Cain – 2014-08-06T15:52:48.950

@nir I think if self-doubt could be simulated by the Demon, then Cogito Ergo Sum would fail, as it would become, "I doubt my own existence, therefore I am OR a Demon is creating my feelings of doubt about myself. The reason qualia is needed for doubt is because we know that the qualia we experience is not necessarily the same as the senses we receive. With only senses and no qualia, there is no room for a third entity to be feeding false information. – Cain – 2014-08-06T15:58:09.607

@Cain, First, the certainty is not of the reality of doubt but of the experience of doubt. Second, an omnipotent demon could hack into the zombie hearing or optic nerve and stimulate it in whatever fashion required to make the zombie preceive a particular illusory voice or image, without requiring qualia at all. – nir – 2014-08-06T18:55:21.057

@Nir, But that's exactly it. The only way the Demon could trick the P-Zombie is through manipulating physical features of it, which clearly indicates that the zombie exists. For a real person, there is another alternative, namely manipulating the experiences of the person without the need for a physical existence, which means another form of proof is required. – Cain – 2014-08-06T19:46:24.780

It's pretty easy to program in hallucination: just allow some internal source to generate data and insert it into a stream of data from the outside. It's not a useful property so we usually don't. (Actually, for software testing it's done all the time, but we tend to be very explicit about which input data is real and which is fake.) – Rex Kerr – 2014-08-07T18:25:11.823


A philosophical zombie could not conclude anything inside their mind, because by definition they don't have any minds. They could make it seem like they have made the conclusion, though; but you (again, by definition) would have absolutely no way to tell if this is really so.

The fact that the argument in question is "cogito ergo sum" does not change a single thing.

Michael Smith

Posted 2014-08-05T20:57:21.567

Reputation: 811

why did you complicate conclude with "inside their mind"? It seems tangential to the question. – nir – 2016-08-06T12:09:19.940


They could say the words, but could never believe them because they have no beliefs. They could act like they believe the words but no actual belief. They could act like they have reached such a conclusion, they could act like they believe it.

If by "reach a conclusion" you mean have the subjective experience of reaching the conclusion then by definition (they have no subjective experiences) they cannot reach any conclusion about anything.

If by "reach a conclusion" you mean act just like a non-zombie who has had the subjective experience of reaching the conclusion then yes they can act like they have reached the conclusion. (Also by definition since zombies can act just like non-zombies.)

Vector Shift

Posted 2014-08-05T20:57:21.567

Reputation: 371

what is the subjective experience of reaching a conclusion? and what does it have to do with the veracity of a conclusion? – nir – 2016-08-05T08:34:04.193

@nir, If that is a legitimate question, you are a zombie -- and you need a tune-up. You are supposed to be preventing all of us from knowing that you have no subjective experience -- get with the program. – None – 2016-08-05T19:50:35.603

@jobermark, what? I apologize but I find your comment unintelligible. – nir – 2016-08-05T20:00:21.167

I am not falling into that trap. You either know enough to interpret that answer, or you don't know enough to ask the question above. Since you did the latter, I am totally confused as to how you expect me to take this response seriously. – None – 2016-08-05T20:21:06.857 The whole force of your original question is based on the ambiguity of the statement "can a zombie conclude.." You don't specify if you mean behaving like they concluded or have a conscious experience of the conclusion. It s also spurious to bring up the veracity of the conclusion since even incorrect conclusions can be reached, and often are. Your question was whether or not they could conclude not if they would be correct in doing so. – Vector Shift – 2016-08-06T06:46:06.017

@jobermark, trap? your comment is simply unintelligible. either you are purposely being silly, or I am afraid that you are too smart for your own good:

– nir – 2016-08-06T07:14:16.093

@VectorShift, I believe you understood my question differently than I intended. I made an edit to clarify it. – nir – 2016-08-06T08:19:01.803

@nir Drawing a conclusion is a conscious act. I think you should find a way to poise your question in a way that doesn't require a zombie to have consciousness. I am not sure of the distinction that you wish to draw between inner experience and a sense of self. A zombie has neither by definition so it doesn't seem a good way to draw the distinction by requiring a contradiction. – Vector Shift – 2016-08-08T00:48:29.733

@VectorShift, 1) here is Turing on attributing machines with thinking: "I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted." 2) Suppose that by the word conclude I mean whatever the zombies are doing when they do things that we would call conclude in our world. 3) I am not trying to draw the distinction you mention, and I do not understand your problem you are having with the question. – nir – 2016-08-08T05:45:00.350

@nir Casual language does not distinguish between someone acting like they are conscious and their actually being conscious. We say that someone is conscious when we see them acting like they are conscious. Of course the whole point of the zombie construct is that these aren't the same. There is no way to determine if another person is actually conscious, this is 'the problem of other minds'. It's a bit like the difference between "seeing a car" and "seeing the light bouncing off the car". – Vector Shift – 2016-08-09T10:32:16.823

@nir Ordinary language doesn't distinguish between the two but the latter is a more precise description of the situation. Turing was talking about a machine acting in a sufficiently complicated manner that we would say that it was thinking like we say that a person is thinking when they act in particular ways. You need to go over the problem of other minds to learn to avoid use of terms like "conclude" , "think" which presume consciousness thereby conflating observable behavior with actual consciousness. – Vector Shift – 2016-08-09T10:42:31.250

I am familiar with the problem of other minds. thanks. I think that it is time to conclude that there is a gap in our understanding of the question and the problem, which prevents us from discussing the them fruitfully, and that this gap cannot be bridged in this forum. – nir – 2016-08-09T11:25:07.730

@nir So you're willing to talk about what zombies think, believe, feel, conclude , etc., because you ignore the problem? You attribute conscious activities to an entity that by definition has no consciousness? – Vector Shift – 2016-08-10T16:26:06.067

@VectorShift, I've explained myself the best I can in the question and in these comments. We do not understand each other. Let's agree to disagree. – nir – 2016-08-10T17:00:49.167

@nir Not this time. You are basing everything on ignoring a contradiction without any argument to the contrary. Is this a technique that you embrace? Your best explanation for assigning cognitive states to zombie is.."I think that it is time to conclude that there is a gap in our understanding of the question and the problem, which prevents us from discussing the them fruitfully, " ? – Vector Shift – 2016-08-12T01:01:20.003