Here is an argument that has its roots in the work of Nagel, and is also probably highly influenced by 'Zen and the Art of Motor Cycle Maintanence.' I just wondered if anybody out there had any comments.
1. We have qualitative experiences, such as an experience of the redness of red.
2. Science always proceeds quantitively, by discovering laws that can be formalised mathematically.
3. You cannot quantify a quality.
4. Therefore, science cannot account for the qualitative aspect of our experience.
(3 votes)
- Thomas Rickarby's blog
- Login or register to post comments
The dubious premise...
Toby Wardman on Wed, 11/05/2008 - 9:03am...is 3, of course. I suspect there are plenty of examples of qualitative data being reduced to quantitative data.
Off the top of my head, here's a candidate: my background is linguistics, and I did some work on 'the sociolinguistics of conversation' as an undergrad. The idea of this is that you take a recorded conversation and you analyse it to determine all kinds of things about the speakers. What is interesting is that if you listen to a conversation before you do any analysis, you get a 'sense' of all kinds of things: the attitude of the speakers towards the topic, the nature of the relationship between the speakers, even such vague things as the time of day. Some of these impressions are conscious, others are unconscious and take a bit of teasing out, but they are all quintessentially qualitative.
Then as a conversational analyst, what you learn to do is go through and pick out specific features of the discourse (number of tag questions, length of micropauses, choice of pronouns, length of overlaps, etc) which can be shown to generate different intuitions in the minds of naive listeners. These are robust, quantitative factors: you can vary them and generate different intuitions in listeners; you can show how skilled actors unconsciously use them to create different impressions; etc.
That seems to be an instance of the qualitative being reduced (or re-expressed, if you like) to the quantitative through the work of science.
[edit for typo]
Have you quantified the quality?
Thomas Rickarby on Wed, 11/05/2008 - 1:01pmFirst of all, thank you for the comment.
Its certainly the best response I've had so far :P.
In your example, I do not think that the qualitative aspect of the conversation, which certainly goes along with the quantifiable aspects of the conversation, is being explained.
There is nothing in the 'number of tag questions, length of micropauses, choice of pronouns, length of overlaps, etc' which in themselves mean what its like to listen to the conversation. (Hand those quantites to a lay person and they probably won't be able to imagine what the conversation was like merely from the data).
All that you know is that conversations of a certain quality also have certain quantifiable aspects in certain quantities. That they go along together doesn't entail that they are identical.
You haven't yet explained how the quantities give rise to the quality of the conversation, which is what I think you need to be able to do to show that you have indeed quantified the quality of the conversation.
Tom
'If you have ice cream I will give you some. If you have no ice cream I will take it away from you.' - Jack Kerouac
what is a quality?
colette sciberras on Wed, 11/05/2008 - 3:24pmIt all depends on what you mean by quality. It seems that you are making use of the distinction between primary and secondary ones. It is normally said that primary qualities can indeed by quantified e.g. length can be expressed in cm. It seems to me that so can secondary ones (of which red is a classical example.) By stating the frequency of its lightwaves, we can know what colour an object is; thus either higher or lower frequencies (I don't remember which) would be "redder."
Of course, you are right that this doesn't come anywhere near to describing an experience of red. Consequently it has been generally accepted that redness is, perhaps, a subjective quality that exists somewhere in the mind of the experiencer, as opposed to real, 'hard' qualities like length, which exist as properties of the object. However, Einstein's theory of relativity shows that length too is, in a way, 'subjective,' and depends on the relative velocity of the observer.
So if by quality you mean the experience of it, then, in all cases, it seems to me, a quality can be redescribed in terms of quantities, although this will not capture what it is like to have that experience. If you mean something 'in the object itself,' an inherent property or whatever, then it seems that the only way we can talk about qualities is by quantifying them; anything else will smack of subjectivity.
But this is not my area and it would be good to hear your comments.
Colette
On The Ontology of Quality
Thomas Rickarby on Thu, 11/06/2008 - 8:15pmFirst of all, thanks for the comment.
You have an appreciative style, which is certainly refreshing in a philosopher.
I wasn't really thinking of the distinction between primary and secondary qualtities. That distinction comes with so much baggage that I think I'd rather leave it alone.
I do think you make an interesting point about the fact we seem to be more easily able to measure length than colour, I think you're right not to lend that too much credence, given that we could also quantify aspects of the red that we see e.g. intensity.
I think that its in your second paragraph where you really touch on what I'm trying to get at. When I refer to the qualitive aspect of experience, I mean precisely what it is like to have such an experience.
On the question of whether qualities are merely aspects of a perceiving mind, or whether they also exist apart from perceiving minds I want to remain silent. Mainly because I don't know on what basis it could be shown either way.
I do think that it is at least possible that what being ultimately consists in is not something entirely physical, though being must have some sort of physical aspect - given the significant technology physical descriptions of being have enabled us to develop. I say this because I think that there is a strong physicalist trend in both the academic and lay population.
Ironically, this trend may be stronger than it is amongst physicists (at least the physicists that I know), given that physicists tend not to believe anything that cannot be backed up with experimental data. I think most physicists would take my position - that we do not know what being ultimatley consists in, only that we have descriptions and approximations of it that are good enough to create nuclear weapons, radio transmittors, optical communication systems and so on.
However, it does not matter whether qualities are merely aspects of a perceiving mind, or whether they also exist apart from perceiving minds. This is because, either way, there are still scientific accounts that may want to accommodate it. Whether those are neurological accounts, or purely physical accounts, there is still the underlying need to explain qualitative experience in terms of a physical account if one wants to maintain the position that all being is ultimately physical being.
I think that is really what I'm trying to get at.
'If you have ice cream I will give you some. If you have no ice cream I will take it away from you.' - Jack Kerouac
Aha... so...
Toby Wardman on Sat, 11/08/2008 - 10:13am...so this is a debate about qualia - i.e. what it is like to have a particular experience. There's a large and forbidding literature around this issue and I don't pretend to have got my head round it all!
But I still think that it would be very difficult - if not in principle impossible - to show that there is at least one quality that can't be reduced to a quality. In your reply to me earlier, you wrote:
Well, that's true, of course. There is nothing in there which in itself means what it's like to listen to the conversation. But you might as well argue that there is nothing in the lightwaves that in itself means what it's like to be red; or, to take another favourite example from the qualia debate, there is nothing in the physical description of water molecules that in itself means what it's like to be wet. But does that mean that science can never fully account for red's redness or water's wetness? I think not. In fact, the opposite is true. Surely it is a good thing - in fact, surely it's absolutely necessary - for an explanation not to have to re-postulate to the thing it was explaining? If you want a (quantitative, (meta)physical, reductionist) explanation of a particular phenomenon, say 'wetness', you had better make sure that the phenomenon itself does not recur in your explanation, or else you'll end up with an infinite regress. In quantifying wetness, you want there to be no actual wetness in your account, or else you haven't explained it at all. In quantifying redness, you want there to be no actual redness in your account, or else you haven't explained it at all. And similarly in the case of the conversation analysis.
I suspect you'll want to bite the bullet and say, 'All right then - science can never (fully) account for wetness, redness, or what it is like to listen to a conversation'. There is a long tradition of philosophers saying this (have you read the Monochrome Mary thought experiments?). But it seems to me to be baseless. If we can quantify the causes and effects of a phenomenon, including its physical nature, the effect it will have on my neurological structure and the resulting effects on my linguistic behaviour, belief structure and future dispositions, then what exactly is supposed to be left unquantified?
By the way, I think it is revealing when you say, "Hand those quantites to a lay person and they probably won't be able to imagine what the conversation was like merely from the data". Right. But a linguist would. Equally, a lay person would not necessarily recognise the colour red from an electron-level physical description of a light wave. But a scientist would. This is not literally the same as undergoing the experience, of course, but why should it be? All that is necessary is that the experience has been quantified.
Biting the bullet.
Thomas Rickarby on Sat, 11/08/2008 - 11:09amWow. Those are some pretty good arguments. But I think I can come up with some sort of a reply. And yes I do want to 'bite the bullet'. Actually I think that was the point of my original argument!
'In quantifying wetness, you want there to be no actual wetness in your account, or else you haven't explained it at all.'
This is certainly true, and I have no problem with the way science gives scientific accounts of certain phenomena by excluding the 'what it is like' aspect of the phenomenon from the explanation.
However, when science wants to begin to account for the 'what it is like' aspect of the phenomena it has a problem. And there are many fields in which it may want to do this, specifically in the neurological and cognitive fields.
'If we can quantify the causes and effects of a phenomenon, including its physical nature, the effect it will have on my neurological structure and the resulting effects on my linguistic behaviour, belief structure and future dispositions, then what exactly is supposed to be left unquantified?'
What it is like to experience the phenomenon! Furthermore, it is possible that the 'what it is like' aspect of the phenomenon actually plays some role in the life of an agent, and is not merely an epiphenomenon. For example, when I like the look of one painting over another I am specifically appealing to what painting is like for me, rather than to some quantitive description.
'Equally, a lay person would not necessarily recognise the colour red from an electron-level physical description of a light wave. But a scientist would.'
On this point I think you are correct. However, I think there was something that I was getting at, so I hope you don't mind if I take a step backwards and redefine the the terms of the argument.
After all, it may be the case the scientist can imagine what the conversation or the light is like with the aid of the experimental data. However, what I'm trying to get is that the quality is not implicit in the data. When one sees the mathematical expression for the quality of red, one does not 'see red'. This is analogous to the fact that when one sees the word 'dog', one does not see a dog in the mind's eye. One might take the expression and use that as a cue to begin imagining what is expressed, but that imagined representation is not implicit in the expression - one has to bring something to it.
The best way of unpacking this point is by noticing that no-one actually knows what it is like to experience a quark, precisely because we do not have the required perceptual faculties. And yet science can still make headway with physical descriptions of the quark, because it doesn't need to understand what it would be like to experience a quark in order for quarks to take a certain role in scientific theory.
Now if we take a blind person, who is also a scientist and we ask him to imagine what the colour red is like merely from the physical description of red light, it seems like he wouldn't be able to do this. For him, in terms of quality, the understanding of a quark and the understanding of red light would tell him exactly the same thing. The only reason a scientist who can see can use the physical description of red light to imagine 'redness' is because he has already had experiences of 'redness', which then enable him to take the physical description as a cue for his imagination.
Hope that clears everything up. :)
PS I haven't read about those experiements, but I am certainly interested to do so.
'If you have ice cream I will give you some. If you have no ice cream I will take it away from you.' - Jack Kerouac
Something about Mary
Toby Wardman on Sun, 11/09/2008 - 2:01pmThanks for your thorough reply! OK, firstly: the Monochrome Mary thought experiment. Your reference to the blind scientist towards the end of your reply is very similar to this thought experiment. As far as I can tell, the argument you're making is the same as the one made by Frank Jackson, the guy who invented the thought experiment. I am much more tempted by the kind of response suggested by Daniel Dennett and others, which are also outlined in the SEP article.
Maybe we're talking at crossed purposes here, but this seems wrong to me. My claim is that a scientific explanation does not exclude 'what it is like'; it redescribes it quantitatively, and in doing so, it accounts for it.
Well, I can't dispute that the scientist who sees red is in a different position from the scientist who reads about red and thoroughly understands it. But the question is, what is the nature of this difference? Is there some more information gleaned by the experience of seeing it compared to the experience of reading about it - some extra fact or facts, something that the scientific description doesn't account for? Your answer is yes; my answer is no. You say "the quality is not implicit in the data"; I say precisely that it is.
Remember, the scientist is supposed to have a complete physical account of red in front of him. For the physicalist, that includes a description of how his brain will react to seeing red, how his neurological state will change, what new beliefs and opinions he will form, what words he will utter ("what a pretty shade of red!"), and so on. Everything. The only difference between this knowledge and the act of actually perceiving red lies in the de se fact that the scientist is not himself undergoing that particular experience right now. We can agree on this, but I say it is not relevant, because this single fact is not part of the description of redness; it is not part of the description of what it is like to see red. It is just a plain old fact about the scientist's current state. And anyway, it is a fact that the scientist also knows, of course.
So what I'm saying is this: of course seeing red is different from having a complete scientific understanding of red. But I say the difference is simply in the de se fact that I'm experiencing it. I deny that there is any difference in the knowledge I have when I see red from when I totally understand its quantitative description. To put it another way, I deny that there are any (qualitative) facts about 'what it is like' to see red that are not also present, redescribed quantitatively, in a physically complete description of the phenomenon and the accompanying experience.
Really? Are you saying that it would be beyond the wit of a psychologist to experiment, discover what you like in paintings, and quantify it? Surely not. People do this all the time. The role that your experience of looking at a painting plays in your life is entirely describable in quantitative terms. The way you respond to a particular painting is a consequence of certain perceptive acts, certain brain state changes, certain beliefs and opinions, certain speech acts and so on. By a series of experiments, a psychologist could discover facts about what features of paintings appealed to you. (Of course you might not be aware of these facts - you may simply have your preferences delivered to your conscious mind on a plate when you evaluate paintings - but that doesn't mean the facts aren't there. Equally, I may not know that what causes me to want to eat Chinese takeaways is their high MSG content; I just know that I like Chinese takeaways.) If we had all the physical knowledge about your likes and dislikes when it came paintings, we could use it to select paintings that appealed to you and to avoid ones that didn't appeal, just as your wife might try to do when she chooses you a Christmas present. Again, there is nothing mysterious, nothing irreducible, nothing essentially unquantitative about this - it's just that we ordinarily lack that kind of complete qualitative knowledge.
Can 'Redness' be formalised mathematically?
Thomas Rickarby on Tue, 11/11/2008 - 3:20pmThe problem with the Monochrome Mary thought experiment outlined in the article is that it requires that we already have the most complete physical description of the experience of colour (as we see it) that we are going to get. Yet this account is not yet formulated, or at least not yet understood as such. Which is to say that even upon the highly doubtable premise that the account of colour we have is the most complete account we are going to get, we still do not understand our current account to be the most complete account.
My argument is intended to show why there will be some difficulty in having a complete physical understanding of qualitative experiences so long as we understand the physical as that which can, in principle, be formalised mathematically.
Unless you find a way of quantifying 'redness' as we see it, you cannot formalise it into a mathematical description. It doesn't matter how far scientific accounts go in describing the physical brain-states which give rise to redness, unless they can include 'redness' as a quantity in their equations, they cannot explain it scientifically. Even if a scientist develops a perfect understanding of the physical mechanism which underlies our perception of redness-as-such, unless he can show how redness as we see it is included in the equation, he has not shown how the physical being of a body perceiving a lit surface has given rise to this quality.
If you cannot find a quantifier for the redness of red as we see it, then you must agree with me also, or at least agree that there is some difficulty in quantifying the redness of red as we see it.
'If you have ice cream I will give you some. If you have no ice cream I will take it away from you.' - Jack Kerouac
Getting to the nub of the disagreement!
Toby Wardman on Wed, 11/12/2008 - 8:46amThis is the heart of where I disagree with you. I believe exactly the opposite. I believe that as long as an explanation retains reference to the thing it is trying to explain in the same terms, it is no explanation at all. Explanations are logically debarred from having their explanans recur in their equations. As a point of logic, if an account of redness did include redness qua redness in its account, it would fail (apart from anything else, it would immediately introduce an infinite regress).
I'm not sure I understand.
Thomas Rickarby on Thu, 11/13/2008 - 9:09pmIf an explanation does not end up with what is explained, then how can it be said to have explained anything at all?
If I ask why Jill walked into the lampost, and you reply that its because she wasn't paying attention to where she was going, then you have offered an explanation.
Surely the inference in this case is that the explanation entails the explanandum. i.e.
P (Jill walked into a lampost)
IFF Q (Jill wasn't watching where she was going) then P
Therefore Q
R (We have an experience of Red qua red)
IFF S (Insert scientific explanation here) then R
Therefore S
I think what I meant to say, and I should have been more clear, is not that red must form part of the explanation for red, but that it at least must be quantifiable if a scientific explanation is able to entail it, given than mathematical equations can only entail quantities.
I think the actual nub of the disagreement is the ability to quantify red as we see it. I'm asking you to try and do this.
'If you have ice cream I will give you some. If you have no ice cream I will take it away from you.' - Jack Kerouac
This works, doesn't it?
Toby Wardman on Mon, 11/17/2008 - 10:11amFair enough, then. So we have:
R (We have an experience of Red qua red)
IFF S (Insert scientific explanation here) then R
Therefore S
This seems fine to me. And it seems fine to say that S can be a quantifiable scientific account. In fact, I think that is exactly what scientists do when they provide an account of redness.
Of course, S might be a long conjunction, especially if you need to exclude things like "and the subject is not unconscious, and the subject is not hallucinating", and so on. And the relationship between R and S will only be a physical contingency, not a logical necessity, that R follows from S, but of course science only deals with the way the world actually is, not the entire scope of logical possibility. And I'm not necessarily saying that any particular scientist today will be able to completely define S.
But I see no reason at all to think that S should be undefinable as a quantity.