The External World: Further Experimental Evidence of Its Existence

(collaborative with Alan Moore)

It occurs to me to doubt whether the external world exists -- that is, whether anything exists other than my own stream of conscious experience. Radical solipsism is of course crazy. But can I show it to be wrong? Or is my only recourse simply to assume it's wrong, since any attempt at proof would be circular? Can I break out of the solipsistic circle, using facts about my own stream of experience, knowable from within the solipsistic perspective, to license an inference to the existence of something beyond my stream of experience?

The most famous philosophical attempts to prove the existence of an external world from solipsism-compatible assumptions and methods -- for example, Descartes's and Kant's -- are generally acknowledged by the philosophical community to fail.  I'm inclined to think those attempts fail because they set the bar too high: They aim for certainty through deductive proof.  Better, perhaps, to aim lower -- to use scientific methods and scientific standards, licensing only tentative conclusions as the current best explanation of the available evidence. In another way, too, I aim low: I aim only to refute solipsistic doubts in particular, not other sorts of skeptical doubts such as doubts about memory or reasoning. In fact, I will take the general standards of science for granted, insofar as those general standards are compatible with solipsism. I aim to do solipsistic science.

Last year on my blog I presented the results of two experiments designed to provide evidence that an external world does indeed exist. The first experiment suggested that something exists ("my computer") that is swifter than I at calculating prime numbers. The second experiment suggested that the world has some constancy to it that exceeds the constancy of my memory. Today, I will report on a third experiment.

If solipsism is true, nothing in the universe should exist that is better than I am at chess. Nothing should exist with chess-playing practical capacities that exceed my own. Now I believe that I stink at chess, and my seeming student "Alan" seems to have told me that he is good at chess. If solipsism is true, then he should not be able to beat me at rates higher than statistical chance. I agreed with this seeming-Alan to play twenty games of speed chess, with a limit of about five seconds per move.

Here is what seems to be a picture of our procedure:
(photo credit: seeming-Gerardo Sanchez)

The results: seeming-Alan won 17, I won 2, and one stalemate. 17/19 is statistically higher than 50%, p < .001 (hand calculated).

It occurs to me that I might have hoped to lose, so as to generate results confirming my preferred hypothesis that the external world exists. Against this concern, I comfort myself with the following thoughts: If it was an unconscious desire to lose, then that implies that something exists in the universe besides my own stream of conscious experience, namely, an unconscious part of my mind. So radical solipsism is false. If it was, instead, a conscious desire to lose, I should have been able to detect it, barring an unusual degree of introspective skepticism. What I detected was a desire to win as many games as I could, given my background assumption that if Alan actually exists I would be hard pressed to win even one or two games. I found myself repeatedly and forcefully struck by the impression that the universe contained a practical intelligence superior to my own and bent on defeating me -- an impression, of course, confirmed by the results. The most natural scientific explanation of the pattern in my experience is that that impression is correct.

I might easily enough dream of being consistently defeated in chess. But dreams of this sort, as I seem to remember, differ from the present experiment in one crucial way: They are vague on the specific moves or unrealistic about those moves. In the same way, I might dream of having proven Fermat's last theorem. Both types of case involve dream-like gappiness or irrationality or delusive credulity -- the type of gappiness or irrationality or delusive credulity that might make me see nothing remarkable in my daughter's growing wings or in discovering a new whole number between 5 and 6. Genuine dream doubt might involve doubt about my basic rational capacities, but if so such doubts are additional to simply solipsistic doubt. Whether I am dreaming or not, if I consistently experience specific clever and perceptive chess moves that repeatedly exploit flaws in my own conscious strategizing, flaws that I experience as surprising, it seems hard to avoid the conclusion that something exists that surpasses my own conscious intelligence in at least this one area.

[Cross-posted at The Splintered Mind]

Comments

[Titus Rivas]

As I understand it, radical monistic solipsism is not a position that states that reality can be reduced to my conscious mind, but that it can be reduced to my mind in general, including its unconscious dimensions. My unconscious mind is indeed by definition external to my consciousness but not to my mind in general, i.e. not to me as the only real, substantial entity. Starting from a radical solipsistic position any empirical evidence is simply evidence of processes within my mind, both consciously and un- or subconsciously. Thus, no empirical evidence can ever disprove solipsism, not even a smart computer or any other apparent entity that seems more intelligent than I. For the radical solipsist, anything that happens, would take place solely within myself.
In my view, only a strong intuition that this is false can save us from solipsism. No analytical or empirical argument will do.

[Eric Schwitzgebel]

Titus: That's not my understanding of solipsism as traditionally defined. But anyhow call your version Solipsism-1 and mine and Alan's Solipsism-2. Our only target here is Solipsism-2. An argument against Solipsism-1, if possible, will have to wait for another day!

[Aaron Boyden]

At the end you consider the problem that when you are dreaming, you generally don't notice the gappiness and irrationality, which of course raises the possibility that there was gappiness and irrationality in your apparently consistent and detailed chess experiment, but you just didn't notice. But you dismiss this as an "additional" doubt beyond the "simply solipsistic" doubt. I'm not satisfied; this distinction between "simple" and "additional" doubt looks fairly suspect to me, more convenient than principled.

[Eric Schwitzgebel]

I discuss the principles a bit more explicitly in the introduction to the post on Experiment 1. Here's what I say there:

*************
A fair game needs ground rules. The skeptic’s position is obviously unassailable if I must prove every premise of any potential argument. On the other hand, I shouldn’t, like G.E. Moore, invoke premises that already assume the falsity of radical solipsism. For purposes of this study, then, I’ll allow myself tentatively to accept the following:

• introspective knowledge of sensory experience and other happenings in the stream of experience,
• memories of past experiences from the time of the beginning of the experiment (but not before),
• concepts and categories arrived at I-know-not-how and shorn of any presumption of real grounding in the external world,
• the general tools of reason and scientific evaluation to the extent those don’t build in any assumptions about the things beyond the stream of experience.
****************

I hope that list seems reasonably justified. Doubting everything hands the game to the skeptic. What I choose not to doubt seems -- I hope! -- reasonably modest, principled, and not question-begging against solipsism.

Where under this principles should I fit the assumption that I am not bizarrely irrational, the way I sometimes am in dreams? Probably under the fourth, but maybe under the first too.

[David]

Possible skeptical problem: (?maybe?)

"If solipsism is true, nothing in the universe should exist that is better than I am at chess."

You've probably noticed that other things in the universe are better at living heavy things than you are. Is this enough to refute solipsism?

My line of thinking is in this 2ish page thing I just now made to show where I'm coming from:

https://docs.google.com/open?id=0B6mXwvHlgUVbX3hnd2tnVXZVbG8

Good blog post that may or may not exist.

[David]

Also, back in the day Von Neumann helped program the chess super computers. If Von Neumann can make a chess computer, and that computer can beat him at chess at a rate that is greater than statistical chance, then it is possible that a solipsist can create an automaton-thing that can beat them at chess. To note: this is done consciously and not subconsciously. However, if the subconscious can affect the solipsistic universe, then it is possible that it can affect the s-universe in a way on par with the conscious-mind writing software that can beat the conscious-mind at chess.

[Eric Schwitzgebel]

Thanks for the comments, David!

I'm inclined to think that the heavy-lifting argument doesn't work. I have visual experience as of, say, a forklift lifting a car. But that doesn't establish, on the assumptions of this exercise, that there is a forklift lifting a car.

If Von Neumann (or anyone else) can manufacture a chess program that regularly beats him at chess, then that would provide evidence of an external world for Von Neumann, on Alan's and my account, if he conceptualized that exercise in the right way. In a way, that would be combining the present experiment with our second experiment of 2011 which involved memory failure and our first experiment involving computational failure -- since if he could remember all his algorithms and quickly compute them he would be as good as his computer. With the forgetting and the defeat, the best explanation, I suggest, is that his own conscious mind is not the entirety of the universe.

[Titus Rivas]

I generally view radical solipsism as a special, extreme case of ontological idealism. In ontological idealism, one normally accepts that there are other beings with their own minds, but no objective (and non-mental) physical world outside these minds. If radical solipsism is only about the conscious mind, then this would also have to be the case for idealism in general. If so, any argument for processes external to the conscious mind would already count as an argument against idealism. This is problematic, because it would imply that there could be no mental unconscious, only non-mental neurological processing etc. Any unconscious process would automatically have to be non-mental. This seems hardly reconciliable with the dream-metaphor for idealism. Most idealists see our experiences of physical reality as some kind of dream-like illusion. Now, dreams are usually not the direct product of conscious processing. In other words, the very fact that one would dream up the physical world would already imply a subconscious mind and therefore disprove idealism and thereby all of its manifestations, including radical solipsism. That's why I'd suppose that (at least contemporary) idealism and solipsism accept the existence of an unconscious mind.

[Eric Schwitzgebel]

Hm, interesting! Do you see evidence of belief in an unconscious in, say, Berkeley?

Bradley on solipsism: "I cannot transcend experience, and experience must be my experience".

Quine: "It leaves each of us in the position of solipsism, according to which there is nobody else in the world, not indeed any world but the pageants of one's own sense data".

Russell: "What grounds have we for inferring that our percepts and what we recollect do not constitute the entire universe?"

I agree that solipsism is often characterized as the view that my mind is the only thing that exists in the universe, but I think *most* of the characterizations along those lines implicitly commit to a conception of solipsism on which the mind is wholly constituted by experience. Probably there are some explicit exceptions to this. I'd be interested in references if you have them.

[Josh Weisberg]

Eric,

Works against pure solipsism (maybe), but not against an evil demon, whom, I take it, is better than you (or me) at chess. So you're not getting at Descartes' worry, for what it's worth. Likewise for B-I-V's and so forth.

So, sure, the external world may exist, but you may have VERY few true beliefs about it

Also, I'm with Aaron here--I think there's something suspect with your simple/additional list. Look, if solipsism has always been true, then even if your chess principle is true, you should still expect to be beaten. A theoretical explanation of this is you are somehow creating Alan and beating him at chess--mental compartmentalization, whatever. Such a thing must be possible if solipsism is considered as a viable scientific hypothesis. So your evidence doesn't tell against solipsism, when it's properly made explicit as a testable theory.

Cheers!

Josh

PS Hi!

[Eric Schwitzgebel]

Hi Josh! I fully agree about the evil demon, brain in the vat, and Descartes. The claim is only the minimal denial of radical solipsism.

I'm not sure I agree with your last point though. Suppose I am creating seeming-Alan. Here are three possibilities: (1.) I really do create something distinct from my conscious experience. Then radical solipsism is false. (2.) I create a part of my introspectively accessible conscious experience that is seeming-Alan conscious experience. But it's unclear how there could be an introspectively discoverable part of me capable of consistently overcoming and surprising what I think of as my own best efforts at chess. (3.) I create something that is a conscious part of me but whose conscious experiences are inaccessible to the main stream of my conscious experience and which seems to behave beyond my control. But then it's obscure why I should say this is part of *me*.

One helpful thing that is coming out of these discussions is that Alan and I need to think more explicitly through our background suppositions about introspectibilty and the boundaries of the solipsistic "self".

[Titus Rivas]

Here are a few quotations from a website (http://solipsistic.askdefine.com/), although they do not seem to derive from a specific book or article:

"On this scale, solipsism can be classed as idealism, specifically subjective idealism. Thoughts and concepts are all that exist, and furthermore, only 'my' thoughts and consciousness exist. The so-called "reality" is nothing more than an idea that the solipsist has (perhaps unconsciously) created." [...] "The Descartes body could only exist as an idea in the mind of the person Descartes Descartes and dualism aim to prove the actual existence of reality as opposed to a phantom existence (as well as the existence of God in Descartes's case), using the realm of ideas merely as a starting point, but solipsism usually finds those further arguments unconvincing. The solipsist instead proposes that their own unconscious is the author of all seemingly "external" events from "reality"."

As far as I know, Berkeley believed in a divine mind to explain anything that continued to exist as a (mental) fact even though it was not consciously experienced by any human being. So he seems to have postulated a divine consciousness rather than a human unconscious.

[Eric Schwitzgebel]

Thanks for the reference, Titus! I think the core issue here is what explains the patterns in my experience over time. It could be something external and not-me (thus, not solipsist in either sense), something in my unconscious (not solipsist in the strong sense of "solispism" that I prefer), or rather the patterns could simply be laws of relationship among my own experiences (consistent even with the strong sense of "solipsism" that I prefer).

[Josh Weisberg]

Hi Eric.

Hmmmm... It seems that maybe you're idea of solipsism is tied up with a transparency thesis: all introspectively discoverable parts of me must fail to consistently surprise me. This seems too strong to me. Further, we have some evidence that dreams can do this, even though we create the dreams. As can drug-induced hallucinations and disorders like schizophrenia. It's not that I'm saying we might now be on drugs or schizophrenic; rather, it's that we have evidence that the mind can in some circumstances create surprising characters. Now, why not all the time? Inference to the best explanation? Simplicity?

I agree that the self-boundary stuff is important, especially given the history of this debate, as Titus points out.

Cheers.

[Eric Schwitzgebel]

I think you're right, Josh, that there's a transparency assumption of some sort in there which Alan and I should make more explicit. However, I deny being committed to the view that on solipsism nothing should be surprising. Things can be *random* (the outcome of my solipsistic seeming die toss can be unknown to me in advance); things can follow experiential laws of which I am ignorant (my afterimages might be in perfect complementary colors though I fail to realize that). But being beaten at chess seems a different matter from mere randomness or an experiential law unknown to me -- *unless* I am willing to tolerate bizarre, post-hoc-seeming experiential laws that operate by strategic and semantic rules unbeknownst to me. My acknowledgement of that possibility is why I do three different experiments (the previous two are in earlier posts) and why I claim only the force of inference to the best explanation. The solipsistic explanations seem to fall back on such bizarre laws; but that seems desperate. I admit that my sense of what counts as desperate post-hoc maneuvering, as opposed to what seems the most natural prediction of the solipsist view, is intuitive and informal.