Wednesday, August 16, 2006

Does Saying "I'm Thinking of a Pink Elephant" Make It True?

Suppose I say "I'm thinking of a pink elephant". I'm sincere and there's no linguistic mistake. Does merely thinking such a thought or reaching such a judgment, silently or aloud, make it true? Tyler Burge and Jaakko Hintikka (among many others) have endorsed this idea; and it has often been thought key to understanding introspective self-knowledge.

I'll grant this: Certain things plausibly follow from the very having of a thought: that I'm thinking, that my thought has the content it has. Any thought that manages to assert the conditions or consequences of its existence will necessarily be true whenever it occurs.

But, indeed, anything that's evaluable as true or false, if it asserts the conditions or consequences of its existence, or has the right self-referential structure, will necessarily be true whenever it occurs: the spoken utterance "I'm speaking" or "I'm saying 'blu-bob'"; any English occurrence of "this sentence has five words"; any semaphore utterance of "I have two flags". This is simply the phenomenon of self-fulfillment. This kind of infallibility is cheap.

If I utter an infallibly self-fulfilling sentence, or if I have an infallibly self-fulfilling thought, it will be true regardless of what caused that utterance or thought -- whether introspection, fallacious reasoning, evil neurosurgery, quantum accident, stroke, indigestion, divine intervention, or sheer frolicsome confabulation. If "I'm thinking of a pink elephant" is of this species, then despite its infallibility, no particular introspective capacity, no remarkable self-detection, is required. And very little follows in general about our self-knowledge.

But I'm not sure that it is really necessary to think of a pink elephant to utter sincerely and comprehendingly, "I'm thinking of a pink elephant". Surely I needn't have a visual image of a pink elephant. Nor need I have, it seems, a sentence in inner speech to that effect (especially if the thought is uttered aloud). What is it to "think of" something? Is it merely to refer to it? To include it in a silent or spoken judgment? That seems a rather thin notion of "thinking"; and if we do adopt that notion, the vacuity of the infallibility claim becomes even more obvious. It becomes tantamount to "this thought makes reference to the following object: a pink elephant". Then it really is structurally no different from the utterance "I'm saying 'blu-bob'".

So, despite some philosophers' quest for and emphasis on the infallible in self-knowledge of the mind, to me the matter seems rather trivial, unimportant, and in fact utterly unconnected to the issue of the trustworthiness of introspection. (Take that, Descartes!) Or am I missing something? Maybe a fan of the importance of self-verifying thoughts can help me out?

10 comments:

Tanasije Gjorgoski said...

Hi Eric,
I would agree that pronouncing "I'm thinking of a pink elephant" in case you present, where "thinking of" is made equal with "referring to" is a trivial truth, but it is such because you chose a trivial case. (similar to Lakoff's witty "Don't think of an elephant").
1.That is, in my opinion, too thin notion of thinking. Thinking is not a sequence of disconnected references to things. (How do I know? By introspection.) We do think about things, but we think about them in context and in their relation to other things. So, when I usually say "I'm thinking of an pink elephant", I'm saying that I'm having some motivated thoughts about pink elephants, I'm considering something about them or in relation with them; Be it if they exist, or if I would like one for my birthday, etc... So, in general case, I don't think that reports of form "I'm thinking of..." are trivial.
And, except in this special case with very thin notion of thinking, I don't see how your attack would work on other reports like "I feel pain in my finger" and "I see dead people".

Eric Schwitzgebel said...

Thanks for the comment, Tanasije! I agree with everything you say. I don't mean to imply that claims like "I feel a pain in my finger" or "I see dead people" are trivially self-verifying. But I also think that we can go radically wrong about such things! My thought is, in fact, that the two types of case are so unlike that we can make no inference from the infallibility of the one to the reliability of the other.

pete Mandik said...

Hi Eric, I agree with your claims re vacuity in this post. However, I'm curious about how some of this relates to some stuff you said in comments on the Dennett and fiction post. In particular, I'm curious how your view that we can be very wrong about our introspective reports squares with your statement in that other comment thread about how you take phenomenology to be ontologically basic (or something like that). These two positions strike me as odd bedfellows, but I confess to not being able to state precisely what bothers me about them. I guess most people I encounter that champion the irreducibility of phenomenology view phenomenal facts as having some special epistemic status that strikes me not sitting well with the fallibilism you seem to like. Does this worry make sense?

Tanasije Gjorgoski said...

I agree that those two are of different nature, and the infallibility of the reports of our view of the world (if one is to argue it), will not be based on this kind of tautological truth as in "I'm thinking of a Pink Elephant".
My proposal that infallibility of reports can be based on infallibility of knowledge of self-agency (as response to one of your previous posts) didn't seem to convince you much :) ...
Let me try the negative approach:
When one reports sincerely "I have pain in my finger", what is that makes the report sincere?

Eric Schwitzgebel said...

Thanks for your comment, Pete! I think you accurately convey the general state of the field. I've never understood, though, why belief that phenomenology is ontologically fundamental needs to go with any especially strong "privileged access" thesis. I do think we have a means of knowing what our experience is that no others share (introspection) -- but it's not an especially trustworthy means! In this, I follow Titchener, Wundt, Kuelpe, Angell, and the preponderance of the early introspective psychologists, though they, more than I, put faith in training as a way of overcoming our introspective deficiencies.

Maybe my position is unusual because it is pessimistic? I've found that people don't generally stress the importance of phenomenology or introspection unless they think they know of a reliable means to learn about it.

Tanasije, I apologize for having failed to be convinced! ;) I do think that some self-knowledge is the self-knowledge of agency -- I think Moran, especially, is helpful on this -- but I don't think even such knowledge rises to the level of infallibility. Not unless it's also vacuous ("I'm about to do *this*"). And I don't think such knowledge is genuinely introspective or takes us very far in understanding our phenomenology.

I'm afraid I have no positive account of sincerity. (I don't think a statement is sincere just in case I believe it, for example.) Are you thinking that a sincerity condition of a report of pain be the actual presence of the pain? If so, I'm not sure why that would be. I happen to think people can be sincerely mistaken about their pains. Sorry to be giving such unsatisfactory answers....

Tanasije Gjorgoski said...

Thanks for the answer Eric.
No need to apologize, I didn't convince myself either. I believe in the infallibility of the reports, but I don't think my attempt was nearly enough.
Anyway, I asked about sincerity, as I don't see how you can give account for the sincere/lying distinction, if you admit no fact about which the subject will be sincere.
If sincerity is the case when one reports what he believes (though you seem to not accept this account of sincerity?), then the sincere reports by definition will implicate that at least "I believe that X is a fact" need to be considered as infallible introspection. (And there is not much distance from "I believe " to "it seems to me")

Eric Schwitzgebel said...

Ah, Tanasije, I think people sincerely attribute to themselves beliefs they don't have quite often (in earlier posts I talked about belief in Heaven and belief in racial equality). In such cases, one might reach a sincere judgment that doesn't accord with one's long-standing and enduring belief; and maybe you're almost always going to be right about your own current judgments. But I would distinguish claims about current judgments from claims about current phenomenology. I think the language "it seems to me..." can be treacherous at this point, since it invites equivocation between expression of a judgment and description of phenomenology. (Too quick, I know. It deserves a proper post in its own right.)

Pete Mandik said...

Eric,
So, why do you think that phenomenology is ontologically basic?

Eric Schwitzgebel said...

Ah, Pete, you force me to reveal my presuppositions! Actually, ontology -- the serious a priori kind of philosophical ontology -- gives me the heebie-jeebies. Maybe, in principle, at the end of the day we can show the identity of phenomenology and some material condition (at least identity enough for science); but we're not there yet, and in the meantime if we're going to take the mind seriously we've got to take seriously this stuff, this phenomenology, whatever the best ontological account of it at the end of the day. It's no good trying to work around it for the sake of being a good materialist.

So I guess when I say "ontologically basic" I mean to be saying that the best accounts we can construct now of what it is to believe, to be happy, to intend, etc., our best accounts of the overall structure and function of the mind, can't leave it out or work around it or reduce it to something else. For now, at least, it's ineliminable.

Perhaps this is too pragmatic, and not metaphysical enough, for most philosophers' tastes. That's how I happen to think, though -- and it's what lay behind (without my articulating it to myself that way at the time) my perhaps idiosyncratic use of "ontologically fundamental" above.

Pete Mandik said...

Fair enough, Eric. Thanks for spelling that out!