Mike -
Based on your last CE comment, I infer that you have access to the Harvard edition of EPM with an intro by Rorty and a study guide by Brandom. I found the study guide helpful at times, so I suggest that you keep referring to it even if it doesn't always turn out to be so.
Although your question was about section 10, just to make sure we're on the same page I'll set out where I understand Sellars to be headed. As always, the following is caveated by the observation that I speak with no authority whatsoever, so "IMO" should be understood to accompany every statement.
There are several things that make reading EPM difficult. One is that Sellars apparently was notorious for generally being hard to follow, a reputation that seems validated by EPM. Also, he throws in occasional seemingly extraneous material, eg, sections 8, 9, and 9 bis (there are two section 9's in the original!) which are possibly better skipped on the first reading. And his terminology - at least in the early sections - is often problematic since it relates to "sense data", which apparently is a somewhat nebulous concept that has changed between its inception early in the 20th C and the present and appears not to be currently popular in any event. Fortunately, as the essay evolves, references to sense data decrease and I believe they are mostly - if not totally - gone in the later sections.
In reading the essay the first time, I tried - and I suspect you are now trying - to translate the language into concepts that are more familiar today, eg, those from neurophysiology. However, Sellars explicitly avoids using neurological vocabulary, presumably for reasons that emerge only toward the end of the essay. So, such translations may or may not help.
As is usual in philosophy of mind discussions, the emphasis is often on the visual rather than sensations in general. This could actually make it more difficult for someone like you who is very familiar with visual processing. Even with my superficial knowledge of that processing I occasionally thought a statement seemed incorrect. Fortunately, little if any of the essay addresses that level of detail, so knowing "too much" about visual processing may not pose any real problem.
Now, on to section 10. Even before I started reading EPM I had encountered mention of those two distinguishable types of "inner episodes" several times, but out of context I couldn't quite see the point. I think part of the problem was that examples tend to be based on visual sensations, in particular those caused by "seeing red"; and the fact that colors - especially red - are so familiar makes it hard to accept that we have to learn about them (eg, the "Mary's Room" thought experiment). We are inclined to think that we "just know red when we see it". But if one focuses instead on sounds, eg, C#, it becomes easier (at least for those of us who aren't musicians with perfect pitch) to understand the difference between these two inner episodes:
(1) experiencing the sensation caused by hearing a musical note that within certain linguistic communities is called "C#"
(2) noninferentially knowing (ie, based on no additional information) that the musical note that corresponds to that sensation is called "C#" within those linguistic communities
Not distinguishing these two types of episodes is one example of the "myth of the given" - that one just "knows" about sights that are before one's eyes, sounds that are in one's ears, etc. (In trying to break entirely free from such misunderstandings, I have found it useful to try to imagine myself in the position of a baby who "knows" almost nothing about anything and has to learn even things that only a few years later will have become so familiar as to seem to have been known all along, ie, to have been "given" just by virtue of being alive and having an intact sensory system.)
The importance of all this is that the essence of the myth is that there is a body of knowledge that can be known noninferentially (the "given"), from which other knowledge can be inferred, and which is therefore foundational. The first part of the essay has the goal of refuting the existence of any such foundational knowledge.
I don't fully understand the significance of the references to "verificationism" and "operationalism" in this section, but it doesn't seem to matter - they never appear again, as best I recall. In the study guide, Brandom relates them to attacks on the myth by logical positivists (AKA, "verificationists", IIRC), who apparently denied the existence of inner episodes. Sellars argues against this denial of the myth later in the essay.
Another attack on the myth is by Wittgenstein, et al, who denied that inner - and therefore private - episodes could be premises (ie, foundational) for inferential knowledge since they are not amenable to public discourse and justification. Sellars notes that this opens the door for a version of the myth based on so-called "looks talk" - how something looks (ie, appears to be) from a person's 1-POV. While private, such reports are generally accepted as incorrigible since although people can be wrong about how a visual scene "is", they can't be wrong about how it "looks" to them. Sellars argues against this denial of the myth in the following sections.
On a first reading, it is easy to find this confusing since in the course of refuting the myth of the given, Sellars is also refuting attacks on it by others. Only after completing the essay does it become (relatively) clear why he does this.
I hope this helps, or at least doesn't just add to the confusion.
Subscribe to:
Post Comments (Atom)
14 comments:
Thanks Charles. That helped and puts me in a better mindset of how to read him; which has so far been more along the lines of a critique on Russellian sense-data (the "triadic" relation as Sellar's puts it I think), rather than Searlian qualia like most people go on about on the Con.Ent. blog (which does seem derivative).
His A,B,C of Sec.6, about the incompatibility of our intuitions (that having "looks" is an unacquired skill yet is somehow still epistemic) I found to be resonant with Wittgestein's and Anscombe's:
"If we cannot speak of doubt we cannot speak of knowledge either. It makes no sense to speak of knowing something within a context of which we cannot possibly doubt it."
"We can speak of knowing, and not merely say it, when there is a possibility of being right or wrong: there is a point in speaking of knowledge only where a contrast exists between 'he knows' and 'he (merely) thinks he knows'."
The baby analogy, and his bit about the shop-keep, is the start of a good argument about how "looks" must not only be learned/acquired but that they can go "stale". I'm anxious to see how he develops it.
And also how he develops/supports the idea of an "inner episode". It does seem odd that his aim is to do away with "the myth of the given", (i.e. that sense-datum or qualia are unshakable epistemic content) and yet maintain "inner episodes". Hopefully he's not maintaining Descrates's theater.
I'm also curious as to what you think about sec.9. It sounds very much like the point you and I had labored on the Con.Ent. blog about how any theoretic explanation of qualia will require, and rest upon, public language and the physical objects it goes on about (as Sellar's puts it, the "cash value" of any theoretic explanation of qualia would seem to rest on public language and, ultimately physical objects).
Then in the second paragraph of sec.10, he seems to say further that he agrees with this BUT it doesn't disprove "inner episodes".
I don't follow his reasons why; actually I'm not even sure he stated them! or if I am to wait till later on in the text.
Thanks again Charles,
-MS
Mike - Sorry to be late in replying, but altho I've been checking back here regularly, I was just checking to see if the post count had incremented rather than paging down to the comments. Doh! (I'm not used to being the blogger rather than a commenter.)
Glad my comments helped. I'll reply later today to your specific issues. But one quickie - I don't think Sellars ever really defines "inner episodes", but they don't lead to the Cartesian Theater - they aren't "watched" by the "eye of the mind". I think of the term as just being an abbreviation for "everything that's going on in the brain in response to a specific sensory stimulation". Then the two distinguishable "episode" types are (1) just the neural activity attendant to (what I call) the phenomenal experience (eg, hearing a musical tone) and (2) that activity plus whatever additional activity is attendant to recognizing the tone as C#.
"I think of the term [Sellar's 'inner episodes'] as just being an abbreviation for "everything that's going on in the brain in response to a specific sensory stimulation"."
But was that ever denied by anyone? If you're right it makes what he's critiquing seem like a bit of a straw-man: Earlier on around sec.10 he says some people have denied "inner episodes" (from what I can tell he's talking about Ryle). Albeit that Ryle certainly underplayed covert-behaviors to much, I don't think anyone has denied that there are covert-behaviors/neurophysiology events.
Or maybe I'm overly sensitive to the term. Both words, inner and episode, strike me in a certain way due to their use historically. E.g. an episode is something you expect to find on a stage, or a frame of sorts (comic or movie). And "inner" in scare quotes is almost always claiming some domain out of reach to any kind of instrument.
Hah - I was about to compose this comment when I read your last one, which addresses just the issue I'm was going to raise: to what extent is our confusion due to the fact that Sellars wrote this over half a century ago, in the interim we know more than was known back then, and words may have meanings today that are somewhat different from their meanings back then.
I'm glad you pushed me on sections 8 and 9 because I had not really understood them before (and no doubt still don't!) Anyway, here's my current take after struggling with it some more.
I understand Sellars to be saying that a "sense content" was understood to be a property of the object being sensed, so that what he calls the "sense datum language" would describe things in terms of properties of the sensed object. His representative sentence is "X presents S with a phi sense datum" in which the sense datum phi is seen as "moving" from X to S. Therefore, the sense datum phi is what S works with in creating the corresponding "inner episode" (which I take to be the phenomenal experience of - for example - "seeing phi".
Today, we think of it differently. For example, illuminating an object with certain light reflecting characteristics under a certain set of conditions causes an SPD to be received by an eye and this results in a certain pattern of neural activity. We would describe this as "X looks phi to S", where "phi" is now a neural activity pattern rather than a property of the illuminated object.
I think what Sellars is getting at with the stuff about the "sense data language" is that in the former case, the sense datum "moves" from X to S's neural activity processors, so there is essentially an isomorphism between sense data at X and processor inputs at S. But in the latter case, we know that a specific set of processor inputs can be caused by multiple scenarios of object light reflecting characteristics, illumination, opponency decision thresholds, etc so that there is no such isomorphism. Or put differently, the "sense datum language" can't be translated unambiguously into the "looks" language.
I'm still fuzzy about all this, but perhaps my musings will give you some ideas, and between us we can sort it out.
I guess he was focusing on what was the contemporary thing to do in '56, focus on a philosophical problem by opening the hood on language. And then assessing the logic of such in an epistemic context. I still think there is much to be said for this approach. It strikes me that many people today consider such an analysis of language as having little traction on their 'qualia' or whatnot. Perhaps, but if they want to tell us about it they have to meet us along that path.
A good example is what Brandom has to say about Sec(16)s. About how Sellers is attempting, not to "explain away" the incorrigibility of qualia, but explain it as linguistic phenomena.
From here the question becomes what is the horse and what is the cart: Is said linguistic phenomena causal of the way we understand qualia to be incorrigibile; or is qualia causing the incorrigibility of said linguistic phenomena.
Others in the Con.Ent. blog have argued recently to me qualia is the horse. I'm not so sure, and I understand Sellars to be exploring the other option. Either way, this dilemma shows that what was the "in thing" for Sellars's period (as mentioned above) is still pertinent.
"Today, we think of it differently."
Well, some people do. I still think the majority of people (everyone seems to have at least ad-hoc philosophical episodes in their life) consider there to be a Cartesian appearance vs. reality distinction.
I.e. there /is/ the property of an object, and then (since I may be mistaken) there is how it /looks/ to me. The /is/ and the /looks/ are different.
This sort of thinking is as alive today as it ever was. And those of a spiritual persuasion let it lead them into Idealism (and optionally Dualism). And those, like myself, from the naturalism persuasion let it lead them to Realism (and optionally Identity-Theory). And I think such ontological commitments are both erroneous for the same reason; what I've called "a program of refication" (Sellars calls it hypostatisation I think).
And this is part of what I was going on about on Con.Ent. about how the crass dichotomy in philosophy of mind is a phenomena to be understood itself. Anyhow, I've digressed from Sellars a bit, but from what I've read of his so far he's jesting in that direction too.
By "we" I mean you, I, and others with a common conception of how all this works. I had hoped to expand our ranks by selling EPM - or at least the summary posted by M Baggot (comment 16 in the CE "Consciousness Meter" thread) - to other CE readers, but that proved to be a quixotic tilt.
Re qualia, I've decided to just ignore that concept. I see only two possibilities: the qualiaphiles are right, in which case there will have to be a whole new subscience developed - analogous to the study of dark matter - that will be too foreign for a geezer like me to adapt to; or they are wrong in which case every minute devoted to trying to understand what they have in mind will have been wasted.
As for the reality-appearance distinction, I look at it as follows: there are "objects" out there that "occupy space" in some energy-field sense and have to be dealt with (the "kicking the stone" argument), ie, we must treat them as "objects ala realism" rather than "objects ala idealism" or risk getting hurt. But they aren't phenomenally "like" anything - they don't have intrinsic color, taste, sound, etc - we do add all of that in our brains, presumably for evolutionarily beneficial reasons. So the "we create the world in our minds" idea is partly right and partly wrong - we don't create the energy-field aspect but do create the "looks/sounds/tastes/etc like" aspects.
This seems so obvious that I must assume that it either is a widespread view the name of which I don't know or is patently wrong in some subtle way that I am missing. What say you?
I might note that in case you are by now wondering "so what's the big deal with EPM?", the meat doesn't start until along about section 30. All the stuff up to that point is - to borrow Rorty's (perhaps in turn borrowed from "big W") - phrase "therapeutic", but then it gets "constructive". So, hang in there!
I'm anxious to get to what Rorty called his "psychological nominalism". Which sounded like he made the claim that human consciousness is a special sort of conceptual awareness that requires language. Rorty made it sound much like the antecedent to what Dennett argued in 1969 with his awareness1 and awareness2 model. And I have a stack of incoherent notes from when I read Quine's W&O that lends itself to such a claim too (which fits in nicely with the Whorf hypothesis in some degree I think). The notion that language is required for the kind of consciousness that is human-esq (i.e. with a narrative 'I') is as fascinating as it is haunting. And thinking of this sort excites the childish curiosity in me to no end.
And I should be picking EatPom up again tomorrow ;)
"As for the reality-appearance distinction, I look at it as follows: there are "objects" out there that "occupy space" in some energy-field sense and have to be dealt with (the "kicking the stone" argument), ie, we must treat them as "objects ala realism" rather than "objects ala idealism" ... we don't create the energy-field aspect"
I struggle with this because of "operational definitions". Time and space exist in some strong sense only in our mind. The physicist Victor Stenger (2009, p.66), "time exist only in the human mind", who is as strong a naturalist or physicalist as one can be, explains this point quite well. And it makes me second guess any explanation I try and give where one is trying to speak of things as existing completely independent of mankind. This is the sort of thing that makes me feel completely ignorant once again.
My current (probably failing) answer is to meet the Idealists half way. Without man and his operations, go his definitions and descriptions. So in this sense we can agree with Idealists, without human minds time, weight, temperature etc. would go away. But that which underlies the descriptions still exists; only without description and definition.
Part of my answer, within the context of colors, you might have read in my OtIoC essay that you have (p.47):
"...the world-as-experienced, i.e. the description found out by our perceptual systems, is likely to be unique to humans in some areas (e.g. our experiences of the coloring of the world may be unique, but our experiences of shape not so). Other species, using different perceptual systems, may inhabit a world with features found to be only commensurable with that system (e.g., “bee-violet” of §vii).
This proposition begs the classical dilemma between Idealism (the idea the world only exists if there are minds to perceive it) and Realism (the idea the world exists independent of minds and the senses provide us with a direct awareness of the external world via sense-data). If the human perceptual system ceased to exist, and took with it its peculiar types of perceptual discernments, doesn’t this commit us to a form of Idealism? Emphatically, no. Only with the insistence of the reification (§v) of these perceptual discernments as outright things would we commit ourselves to such an Idealism; and it simply does not follow that, in purging the world of sentient creatures, the world ceases to exist in every way once conceived. The extermination of a describer and his descriptions does not necessitate the extermination of that which satisfies such descriptions. For example, we can expect such facts as those dependent on a society, e.g. the value of currency and art and nations, to cease. But with the removal of these you still have that which such facts are ultimately dependent on, i.e., that which is describable by physics (e.g., gold and banking system computers). Of this we can be certain; as such a reality existed prior to the appearance of mankind and will continue long after our departure."
I feel this reply isn't entirely secure, and I'm hoping once I learn more I can further strengthen it.
Miscellaneous replies:
"I'm anxious to get to what Rorty called his "psychological nominalism". Which sounded like he made the claim that human consciousness is a special sort of conceptual awareness that requires language."
As I understand it, Sellars' "psych mominalism" comprises roughly the claims that (1) awareness of anything is "a linguistic affair" and (2) "to understand a concept is to master use of a word". The latter makes sense when you think of teaching a child a new word by pointing at an object and naming it. The child not only learns the word but learns to associate it with a certain sort of object, ie, the concept of "an object like this one".
Which leads to another thought re qualia. In my last comment I opined that in the absence of humans to receive and process sensory inputs consequent to their presence, perhaps objects aren't "like" anything. Ie, most of the attributes we attribute to an object are a product of the brain's processing of those sensory inputs. Which, of course, reminds me of the "what it's like to ..." aspect of qualia. Perhaps the same argument applies: I take it that qualia are some special feature of the result of sensory input processing. If so, then just like the other "what it's like" features, they don't exist in the absence of that processing.
Anyway, as I said before I'm taking qualiaphiles at their word when they say qualia are ineffable, and following W's advice (re the beetle in the box) to the effect that there's nothing to say about something about which there's nothing to be said. (A bastardization from DD's paper "Quining Qualia".)
"Victor Stenger (2009, p.66), 'time exist only in the human mind'". I buy that, and the rest of your exposition as well.
You are clearly much, much better read on all this than I. Are you involved professionally? And even if not, do you have formal background in Phil of Mind? I sure hope you've been doing this a long time, because I've been at it several years now more or less 24 X 7 and haven't read a fraction of the works to which your refer. Of course, since I have to read everything 2-3 times (and often still don't get - or at least remember - the material), maybe that's to be expected.
"You are clearly much, much better read on all this than I. Are you involved professionally? And even if not, do you have formal background in Phil of Mind? I sure hope you've been doing this a long time, because I've been at it several years now more or less 24 X 7 and haven't read a fraction of the works to which your refer. Of course, since I have to read everything 2-3 times (and often still don't get - or at least remember - the material), maybe that's to be expected."
Actually, I'm a software engineer (and a bit of an EE as well) by trade. But I've been studying phil. of mind for about 3 years now, I think, day in and day out.
Although, my daily working routine has changed this past month and I don't have nearly the time I use too. I've been wanting to break into it professionally, but I don't have the background; although I'm still fairly young I guess (34). So I'm considering what I should do from here. Supporting oneself on a phil. degree at this point in time seems almost impossible.
But, since so many very intelligent people are going off the rails in tackling an explanation of mental content, the roll of philosophy--in the sense of surveying conceptual topography as I like to put it--has as important a roll as ever. So I'm hoping to find a connection with someone who can point me in a good direction, as locally (New Hampshire) my only option would be a psych. major and that's to narrow in scope for what I'm chasing after I think perhaps. So for now I'm just floating in the margins, with the goal of hopefully writing something articulate and insightful enough that it can't be ignored.
Interesting - maybe it's the common EE background that explains our broad agreement on the Phil of Mind issues! Do you - as I - suspect that having no background in this subject and therefore no baggage to overcome has helped in getting up to speed on contemporary thinking?
I agree with your comment on the continuing role for philosophy, although to make it seem more technical and therefore respectable, I tend to describe the role as "system engineering" (I see lots of disparaging references to "armchair philosophy"). It does seem to me ultimately a system engineering problem requiring high level familiarity with multiple disciplines rather than narrow specialization in one. So, while recognizing the obvious role of neuroscience, I rather doubt that getting a detailed grip on neuron cell biology is necessary in order to attack "consciousness".
Well, I came into the topic with a great naivety. For years on end those around me more or less advocated something along the lines of a Jung & Freud explanation of the mind, and/or a religious one i.e. dualism. And on top of it that the problem was intractable by science. So I got fed up, "if that is the case it is to be demonstrated, not taken as a modus operandi!", I thought.
Having an EE background certainly has contributed to how I got about and see philosophy of mind working (or not). And also a deep appreciation for the topic. There is something it is like to be Charles, but there is nothing it is like to be a spoon, or two spoons. Or a cell, or 2 cells, or 3. But somehow there is something it is like to be 10^13 cells. It seems to be a non-sequitor and if any sole problem is worth dedication to (by an engineer I might add!) it is this one.
Part of the problem is most people have misconceptions of comp.sci. or EE. They fail to realize computation is a pervasive natural phenomena found, not only in black boxes with IBM written on the side, but also living systems. I've been wanting to do writing on this subject. A lot of the good work being done today takes it for granted that the audience has this appreciation I think.
. . .
"I rather doubt that getting a detailed grip on neuron cell biology is necessary in order to attack "consciousness"."
This move (Searle seems to make it) is sort of the inverse of what Chalmers does with his Leibniz Mill, i.e. "Welp I dont see the a 'part of redness' in the brain. Time to rewrite physics!". I can only assume people do not have a proper appreciation for what I was alluding too above and/or they are afraid it will be too "reductionist" dun-dun-dun. When really systems engineering is anything but.
. . .
Anyhow.. I popped on here to ask your thoughts on sec.19 & 20. Which I understood as saying:
"looks read" depends on the "concept of red". And "concept of red" depends on standard conditions and practice.
But what I want to be clear on is how he supports the first point, that "looks" depend on "concepts". As this is something that crossed my mind in my own writing (that someone might object "looks don't depend on concepts; therefor your points on public language don't apply").
The little comment box these blogs give you to type into is brutal ;p
Because this thread of comments is getting a bit unwieldy, I've moved my comments on sect 19 & 20 to a new post. But here are a couple of unrelated replies.
I'm not so sure about "what it's like to be ..." ("WILT"), and in fact I think you make a good argument for my skepticism (unintentionally, perhaps - I wasn't quite sure) with your 2, 3, ..., 10^13 cells. Supposedly, WILT suddenly shows up somewhere in that sequence - but where and why? Maybe it actually is like something to be a spoon - or maybe a thermostat, which after all does make decisions; or the control computer for a very complex system that makes many very sophisticated decisions. Given that I view us as "merely" even more complex systems making (sometimes) even more sophisticated decisions, I find it hard to grab onto a substantive difference. And what about reltively simple people who aren't at all sophisticated in their decision making - does WILT nonetheless apply?
I suggest composing relatively long (almost anything other than a one or two line) comments off-line, then copying them to that "little comment box". I find composing, editing, and proofing immeasurably easier and more effective in a full screen window.
Post a Comment