Lately I’ve been reading Outline of Pyrrhonism by Sextus Empiricus. Sextus collects the arguments for Skepticism as practiced by ancient Greek and Roman philosophers. Since the notion of “skepticism” seems to play some small role here, I thought it would be fun to take a look at what Sextus means by it.
Sextus situates skepticism as the only reasonable response to “dogmatism”. The dogmatists he has in mind are Platonism, Stoicism, Epicureanism, and Aristotelianism (“the Peripatetics”).
He observes, firstly, that the dogmatists all contradict one another — if Stoicism is right, then Epicureanism must be false; if Epicureanism is right, then Aristotelianism must be wrong, etc. What are we to do when dogmatism contradicts dogmatism?
Sextus then observes that none of these positions is “self-evident”, because all of them requires “going beyond the appearances” by making claims about what is “nonevident”. In order to do make claims about what is nonevident, the dogmatist must always either make a circular argument that assumes what they purport to establish or commit themselves to an infinite regress. On this basis he concludes that it is not reasonable to make claims about reality one way or the other. Instead the Skeptic endeavors to live only according to the appearances, and be guided only by what is immediately evident to the senses.
A nice corollary of Sextus’s arguments is that one cannot be a naturalist and a skeptic, since the naturalist does make positive claims about the nature of reality. Naturalism and theism effectively cancel each other out.
The dialectic between dogmatism and skepticism stretches out across the whole history of philosophy. The re-discovery of Stoicism and Epicureanism during the Renaissance re-activated the ancient quarrels between competing dogmatisms (though with a different political dimension, since by this time Aristotelianism had become, thanks to Aquinas and subsequent theologians, the official doctrine of the Catholic Church, which its entrenched power structure). So the quarrel between competing dogmatisms had a political dimension that it seemed to have lacked in antiquity. The revival of Skepticism, most notably (to my mind) with Montaigne, then leads to renewed efforts to establish dogmatism by refuting Skepticism. (This did not prevent some philosophers from attempting to integrate Christianity and skepticism, as Pierre Gassendi did.)
Descartes was, as we know, the most famous (or infamous) of attempts to refute skepticism. But as was pointed out even then, Descartes’ arguments do not avoid circularity. (I believe it was Antonin Artaud who first made this point in first, in his Objections to the Meditations. Descartes’ Reply is, to put it mildly, not convincing.)
The inconsistencies within Cartesian dogmatism led to multiple and contradicting attempts to repair it: Spinoza, Leibniz, Malebranche, and Berkeley being the attempts that have since made it into the Canon (largely because they were all men). At the same time, Pierre Bayle is collecting the new Skepticism into what amounted to a new version of Outlines of Pyrrhonism for the modern era. Following on the heels of all of them, it fell to Hume in his Treatise on Human Nature to demolish all permutations of modern dogmatism by destroying their basis in Cartesianism.
Since then, the dialectic runs back and forth between competing dogmatisms and between dogmatism and skepticism. Kant was perhaps the first philosopher to even attempt a genuine via media between dogmatism and skepticism, but the fatal problems with Kant’s solution are well-known to most casual students of philosophy.
To this day it remains unclear whether there is a via media between dogmatism and skepticism. Some philosophers, including myself, think that the historical arc of pragmatism that runs from Hegel through Peirce and Dewey to Sellars should be understood as precisely an alternative to both dogmatism and skepticism. Others, of course, are not convinced. And so we have the persistence of both multiple forms of dogmatism — naturalism and theism alike — as well as new forms of skepticism.
Can a naturalist be a skeptic? Is skepticism more reasonable than any competing dogmatism? Is skepticism a viable philosophy as a way of life? Is pragmatism a dialectically stable alternative to dogmatism and to skepticism, or must it collapse into one or the other?
KN,
Could you please respond this time? And also to this?
I’m pretty sure that we recognize illusions because of inconsistencies in the totality of what we are perceiving.
As far as I can see, it doesn’t even make sense to talk of comparing to a “correct standard”. What is that “correct standard” and how would we compare?
Something is incorrect if it clashes with measurements or leads to falsified predictions of measurements. I’m not aware of any way to determine that something is fundamentally correct. Not to the satisfaction of philosophy.
Could the same information “lead” different people to different perceptions of what’s on the table or does the information determine the perceptual response?
One wonders if keiths thinks “sense perception” is redundant.
petrushka,
Not to the satisfaction of this (amateur) philosopher, anyway. That’s why I’m a Cartesian skeptic.
Note my wording here:
Those who treat perception as basically veridical regard the lines as being the same length in reality. That’s how they decide that the other perception, of unequal lengths, is the illusion.
To a Cartesian skeptic, both might be illusory. Reality might be significantly different from either perception.
My point is that if you don’t distinguish experience from reality, there can be no such thing as an illusion. Every perception is correct under that goofy assumption.
walto,
Error can creep in at any point in the causal chain, which is why I objected to KN’s sensing/perceiving/judging hairsplitting.
Put yourself in Betty’s shoes. Let’s say there are really no keys on the table, but like Betty, you think that there are. How might this come about?
a. What you’re seeing is a hologram of the keys.
b. God is poofing photons into existence, headed for your eyeballs, that create the appearance of a set of keys on the table.
c. Something else is on the table, but you mistake it for a set of keys.
d. A demon is stimulating your photoreceptors directly.
e. You’re a brain in a vat. You have no eyes, and the neural impulses that would have come from your optic nerves are being fed directly into your brain by the vat apparatus.
…and so on.
The skeptical point is that you can’t be sure the keys are really there, because any effort to demonstrate this depends on the same causal chain whose veridicality you are trying to establish. It’s inescapably circular.
This is where I see almost everybody as going wrong.
There isn’t any “same information”.
The world, in itself, is devoid of information.
Sure, there are properties, such as color, size, etc. But properties are not information. The perceptual system doesn’t just pick up information that is already lying around. Rather, it has to construct information in the first place.
We are perceivers precisely because our perceptual systems are busy constructing information. And we have a meaningful relation to the world, because meaning comes from the details of how we are constructing that information from our engagement with reality.
By contrast, a computer system is given information that was constructed elsewhere. To the computer, the information is meaningless. And it is meaningless, because the computer was not involved in the constructing of the information that it uses.
Above, I said “sure, there are properties.” But, as far as I know, that is contentious. Properties might exist as natural kinds. Or maybe they are our inventions, and we ascribe properties to the world though the properties are not really there (nominalism). I usually try to avoid getting into debates about that, because I don’t see that it matters.
Realism: properties are real. We construct information about those properties, and ascribe properties to the world based on that information.
Nominalism (maybe): properties are our invention. We construct information such as would allow us to exploit reality for our own use, and we invent properties as part of how we use that information to describe the world, and we then ascribe those properties to reality.
I prefer to avoid debates, because I don’t think it even matters. It is our ascribing of properties to the world that is important, and is the basis for consciousness, intentionality, meaning. Whether or not the properties are real is about as important as the number of angels that can dance on the head of a pin.
keiths,
Thanks, but I’m just trying to get a sense of what you mean by “inaccurate information.” Is it that which creeps into every sort of error in your view? When, e.g. you seem to see keys, but it’s something other than keys that’s actually on the table, is the information inaccurate?
Interesting post, Neil. Thanks. But I don’t understand what you mean about property realism–either about what that term conveys to you or why you think it’s not important. Do you take the truth or falsity of property realism to be like whether there are an odd number or even number of planets–something we’ll never find out–or, like the positivists, do you want to suggest the question is strictly meaningless? Or what?
No, to both. It is not like the number of planets, in that we suppose there to be a definite answer (this may be a wrong supposition). And I have never liked the positivist dismissal as meaningless.
That said, I do wonder about philosophical realism. It seems to be asking the question “are real things real?” And I’m puzzled why that isn’t a tautology. It seems to depend on the idea that there are two meanings for “real”. There’s the one we use in ordinary life. And there’s some other meaning that is being asked about. And I’m skeptical of that move, though it makes sense for theists.
I don’t see that the question of property realism matters, because I don’t see that anything important depends on it. If you want properties to be real, then just say that they are real and recognize that this is a matter of social convention (or “meaning” convention).
I take real to mean something like existing “independently of anybody’s thoughts.” I think that’s an extremely interesting question–nothing at all like angels on pins. Saying “who cares” seems ostrichy to me.
But that doesn’t really clarify anything.
There’s a reality that’s independent of our thought. I don’t have a problem with that.
Apart from that: We carve the world at its seams. Except there aren’t any seams. We really carve the world for our convenience and in ways that are congenial to our biology and our social customs.
Suppose that people had never distinguished between cats and dogs. They just treated both as small animals. Would cats then not exist? Or would they still exist, but we don’t know about them. How could they be independent of our thoughts, if it depends on how we characterize and recognize them for something to be a cat?
As long as they could still be there, even before we developed words for them or after no more humans existed–they’re real. To carve up the world at its joints or otherwise, there has to be a world to be carved. That world has (what we now call in English) “properties” or it doesn’t.
Okay. I don’t have a problem with that. I’ve argued that view myself.
The corollary, however, is that there are an uncountable infinitude of real things, most of them unknown to humans.
To me, the question should not be “what exists?” Rather, it should be “of the uncountable infinitude of things that exist, which shall we choose to separate out and name for our own use?”
Sure. That’s realism. Ultra, even!
Neil,
The next time you find yourself Neilishly dismissing a widely-held philosophical position as absurd or a tautology, why not — you know — look it up to make sure you understand it?
What’s meant by ‘philosophical realism’ isn’t exactly a secret. The first sentence of the Wikipedia article:
Oh good. A “Cartesian Skeptic” is going to lecture us about realism. Makes about as much sense as a “Cartesian Skeptic” giving us a lecture about the correspondence theory of truth. But what the hell do I know.
Mung,
Not much, evidently.
You actually think that Cartesian skepticism is anti-realist? Or that it clashes with the correspondence theory of truth?
keiths:
walto:
The inaccurate information shows up wherever in the causal chain the error is introduced. For example, in this case…
…the error is introduced at the points in space and time where God is injecting those misleading photons that appear exactly as if they had been reflected by a set of keys on the table.
I should add that exactly where the error is introduced can be a matter of interpretation.
Consider this case:
You can interpret this in at least two ways:
1) The error is introduced by the hologram. The photons leaving the hologram make it appear that there is a set of keys on the table, when in reality there are no keys.
2) The photons are an accurate “representation”, but not of a set of keys on the table. They are an accurate “representation” of a hologram of a set of keys resting on the table. The error is introduced by the perceptual system, which erroneously interprets those photons as indicating a physical set of keys rather than a hologram.
I think that both interpretations are defensible, and that it doesn’t really matter which one you consider to be “right”. The point is that the perception — of a physical set of keys resting on the table — is incorrect.
No, I think it is irrational. Or should I say a-rational.
Cartesian Skepticism assumes there is something “out there” that is real, but can provide no evidence for that assumption. That is why it has been pointed out in this thread that there is at least one sense in which Cartesian Skepticism is dogmatic.
It establishes a dichotomy that is only available if you accept the assumptions.
And if there is such a thing as “the correspondence theory of truth,” it would hardly follow from Cartesian Skepticism.
So there may not be any such thing as “inaccurate information” after all? (That’s what I was getting at.) It seems to me that information just…..is. It’s not the sort of thing that can be accurate or inaccurate.
walto,
No, there’s definitely inaccurate information. After all, you end up with the perception that there is a set of keys on the table, which is inaccurate. My point is that where the inaccuracy enters the causal chain can be a matter of interpretation.
Sure it can. If Betty tells Bob that there is a set of keys on the table, when in fact there are no keys on the table, then the information she is passing to Bob is inaccurate.
An assertion that p is a particular sort of “information” it has a truth-value, and so I can see how that could be inaccurate. Photons aren’t like that, though, are they? When I asked what you meant by inaccurate information above you gave a list of examples:
Are any of those actually “inaccurate information”? Or do you now mean stuff like somebody being mistaken or lying to someone?
I hope that keiths is finally able to see the problem with treating deviant causal chains at the level of subpersonal cognitive machinery on analogy with deceptive or false utterances between persons.
Mung,
No, Cartesian skepticism does not depend on realism. It is also compatible with idealism and solipsism. For example, suppose that your mind is all that exists, and all of your perceptions are just generated by your subconscious. Cartesian skepticism surely applies in that case.
Don’t pull a Neil. All you have to do is look it up.
Who said that it followed from Cartesian skepticism? Pay attention, Mung.
You claimed that they were incompatible:
There’s no incompatibility.
KN,
What problem? The perception that there are keys on the table can be accurate or inaccurate, just as an assertion that there are keys on the table can be accurate or inaccurate.
That may be–although it’s been controversial since, maybe, forever. But what’s the “information”? Is it like the photons, or is it like you are here construing the perception?
And that is precisely the conflation that I’m arguing against, for reasons I’ve given several times. You only get to treat the states of sensorimotor abilities as being like assertions if take the sensorimotor abilities as themselves being like speakers of a language. And to whom are they speaking? To you? Who is the “you” to whom your senses are “speaking”? Is your brain? Your ego? Your soul?
What I’m desperately trying to get you to see is that you can’t have Cartesian skepticism without a Cartesian theater (a la Dennett). One can’t treat “the deliverance of the senses” as being analogous to utterances without there being some one who receives those utterances and interprets them. This requires a Cartesian theater as being the quasi-ego where it all comes together.
To interpret a linguistic act requires that there be a speaker of the language who receives those utterances and tries to make sense of them. If the deliverance of the senses is like a linguistic act — something that requires interpretation, something that could be true or false, could be accurate or inaccurate — then there has to be an interpreter. And where is that interpreter located? Is it the brain? The visual cortex? What language do photons speak to retinas? What language does the retina speak to the visual cortex?
In short, the position you call “Cartesian skepticism” requires taking an intersubjective, socio-linguistic relationship between embodied persons and transposing into the subpersonal cognitive machinery that implements perception. It’s a category error — a conflation of the personal and the subpersonal.
Cartesian Skepticism depends on creating a dichotomy, without which there would be no reason for the skepticism. Don’t you agree?
As an aside, what do people think about calling the photons a representation of the keys? Is that a proper use of the term representation?
Did you fail to notice the quote marks around “representation”, Mung?
Instead of looking for pointless gotchas, why not try to understand the ideas that are being discussed so that you aren’t perpetually lost?
That’s closely connected to what I’ve been asking keiths about. “Representation” seems like an intentional term to me (although “picturing” according to some philosophers–is not; as I said, this whole area has been very contentious for a long time). But whatever one’s views on whether perception or sensing (or both) is intentional, photons don’t seem like the right sort of things to be representing. Putnam has a famous passage on this matter regarding ant tracks in the sand. In any case, I don’t quite see how photons can be inaccurate.
Mung,
Sure. Take my solipsism example:
Your conscious mind experiences the perceptions, and the subconscious mind generates them. But there doesn’t have to be anything other than your own mind. Realism isn’t necessary.
walto,
Don’t get me started on picturing — we’ll be here all day!
(As a brief aside, I have finally worked out my own interpretation of what Sellars should have meant by “picturing,” given the function of that concept within his overall system. I’ll be writing it up this summer for a forthcoming volume on Sellars and the history of philosophy. My contribution will be on Sellars’ relationship to pragmatism and how the signifying/picturing distinction resolves an ambiguity within pragmatist treatments of truth.)
walto,
That’s why I said that identifying the source of the error — the place where the error creeps into the causal chain — can be a matter of interpretation.
When a person mistakes a hologram for a set of keys, one could argue that:
a) The perceptual system is working properly. It is designed (did you just wet your pants, Mung?) such that it will perceive a set of keys on a table when presented with a stream of photons that could be coming from a set of keys on a table. In fact, the photons are coming from a hologram, but they are “telling” the perceptual system something different. The stream of photons is the source of the error.
b) The perceptual system is broken. It has a design flaw such that when it receives a stream of photons that could be coming from a set of keys on a table, it assumes that they are coming from a set of keys on a table. The photons aren’t “lying”; it’s the mistaken assumption that is the source of the error.
c) The perceptual system is working properly, but it’s receiving insufficient information. The informatio carried by the stream of photons is not enough to disambiguate between “this stream of photons is coming from a set of keys” and “this stream of photons is coming from a hologram”. The insufficiency of the information is the source of the error.
My preference is for interpretation (c). However, to dwell on this is to miss the point. Regardless of the interpretation, you end up with an inaccurate perception — you perceive keys on the table when there are none. The fact that such inaccurate perceptions are always possible and cannot be ruled out is the basis of Cartesian skepticism.
KN,
It’s a repeated mistake of yours to bring social, intersubjective factors into discussions where they are irrelevant.
Betty can perceive a set of keys on the table, and that perception can be accurate — if there really are keys on the table — or inaccurate, if there are no keys. Betty does not have to interact with anyone in order to perceive (or misperceive) that there are keys on the table.
Ditto for believing.
She can even assert to herself that there are keys on the table, and she can utter that thought out loud — to no one but herself. Others are not needed.
Asserting and uttering can (and typically do) involve others, while perceiving and believing need not. In any case, all four — perceptions, beliefs, assertions, and utterances — can be accurate or inaccurate.
You’ve acknowledged that there are such things as illusions — mistaken perceptions. As I wrote to walto above:
Well, it depends on what one thinks the point is. I’m pretty sure everybody here agrees that what you’re calling here inaccurate perceptions are always possible. That’s almost a truism. What I’ve wanted to do on this thread is simply understand what you mean by “inaccurate information.”
FWIW, while I haven’t read it terribly carefully, I don’t disagree with anything else in your post. Lots of places in the causal chain for screw-ups to emerge.
KN,
No. I am only taking perceptions to be like assertions in that they can be accurate or inaccurate. I am also pointing out that perceptions can be the basis of assertions. Isn’t it obvious that when Betty asserts that there are keys on the table, she is doing so because she perceives that there are keys on the table?
Absolutely untrue. Did you miss my conversation with walto on that topic? I’ll try to track it down for you.
ETA: That discussion can be found starting here.
walto,
From there, all it takes to get to Cartesian skepticism is to acknowledge that systematically inaccurate perceptions are possible. If every time I look at the table, God poofs some photons into existence that stream toward my eyeballs, causing me to perceive a set of keys on the table, then those perceptions are systematically inaccurate.
I think KN made a distinction between perception and judgment. And I think that comes in here. My perception never tells me that there are keys on the table. That’s a judgment that I make on the basis of what I perceive. Sometimes it’s an easy judgment, and sometimes it can be tricky. The judgment can be false. The perception is neither true nor false.
Here, I’m going with the Gibson view, that perception just is the acquiring of information about the environment.
I don’t deny that hallucination are possible, obviously. What I deny is that the possibility of hallucination is a good premise from which to motivate Cartesian skepticism.
I know Descartes thought so, but he was wrong about that. Descartes thought so because he had no way of reconciling direct realism with mechanistic physics, and so he had to throw the manifest image under the bus in order to save the scientific image. (Though he then needed to ground the scientific image in theology, etc.).
I think that we actually can reconcile direct realism with our best explanations of the subpersonal cognitive machinery and with evolutionary explanations of the origins of our cognitive machinery. Hallucinations are certainly interesting but they’re not a problem for direct realism that emphasizes the role of integrating cross-modal sensory integration and minimizing discrepancies between exteroceptive and interoceptive predictions and prediction errors. This coheres with the phenomenological results that we distinguish between hallucinations and perceptions in terms of bodily movements, failures to conform to other sensory modalities, and so forth.
Descartes can only get the conclusions he wants by neglecting the role of bodily movement in making perception possible, and he has to do that because he wants to put bodily movement in the category of res extensa. And he has to put sensory information in the category of res cogitans because he realizes that he has to put perception in the same category as he puts action, and intentional action has to go over in the category of “whatever physics can’t explain” because he can’t give up on free will.
Neil,
That distinction is irrelevant to the discussion. See my reply to KN here.
While I don’t personally have the problems with this that KN (who’s probably thought and read more about this than I have), I point out again that the claim that perceptions (however one parses those out of the world, exactly,
I don’t agree with that. Most people don’t, actually.
Yes, that’s roughly my view as well.
I do think that hallucinations pose a problem for the Gibsonian view, though not an insurmountable one. And I do think that integrating a Gibsonian view of perception with an inferentialist (a la Brandom) view of judgment makes it even more clear why perception and judgment belong to different categories. It’s the conflation of those categories that leads to Cartesian skepticism appearing to be a coherent position.
KN,
We aren’t just talking about hallucinations. We’re talking about mistaken perceptions as a whole.
What motivates Cartesian skepticism is the possibility that perceptions can be systematically mistaken. You’ve been unable to demonstrate that this is impossible or unlikely.
walto:
keiths:
walto:
Why is that insufficient? The whole point of Cartesian skepticism is that Cartesian demons, brain-in-vat apparatuses, etc., can produce systematically inaccurate perceptions, and that we cannot rule out those possibilities or even demonstrate that they are unlikely.
It’s a logical possibility, sure. But not one that makes any real difference to practice.
In practice, we have lots of ways of distinguishing perceptions from hallucinations. To mention just two, there’s the question of whether the sensory episode coheres or fails to cohere with episodes coming from other sensory modalities, and then there’s the question whether the actions of other embodied subjects mesh with our own. If I seem to see keys on the table, but they pass through my hands when I try to pick them up, that’s a reliable indication that the visual modality isn’t working correctly. Or if I seem to see a sunset, but my friends don’t, I’m more inclined to think that my eyes aren’t working properly.
There are lots of tests that work in practice for reliably distinguishing between perceptions and hallucinations. Descartes has to brush all those aside in order to generate his merging together of perceptions and dreams into one category. And so do you, it seems.
None of that amounts to a refutation of the mere logical possibility that I’m a brain in a vat or a disembodied spirit being deceived by an evil genius (or any other of the infinite fantastic scenarios that one can imagine), but logical possibilities aren’t sufficient to generate genuine doubt in actual practice.
KN,
We aren’t talking about mere logical possibilities. These are logical possibilities that cannot be dismissed as improbable.
That’s what makes Cartesian skepticism so interesting to me.