A recent post over at Cranky Philosopher argues that it isn’t.
His post opens with:
states. To establish this thesis all I need is one good example. So
consider the felt pain that ensues when I plunge my hand into extremely
hot water. This felt pain or phenomenal pain is a conscious mental
state. But it does not exhibit intentionality.
He follows with an argument for the said claim which is a bit too long to post here.
His view, if I understand it, is that the experience of properties such as pain, color, warmth, are not intentional, as a mental state is only intentional if it “essentially” has an object. The notion of an ‘object’ is left undefined, but he got quite bellicose when I asked for clarification (I suggested that properties can be objects).
I realize that more traditional phenomenologists would say that pain is not intentional, but I am frankly not convinced by ol’ Cranky’s argument to this effect. My favorite philosopher, Dretske, would certainly disagree: pains are representations that function to pick out damaged states of the body (this isn’t sufficient for a representation to be a pain, but it is part of his story). Rosenthal has a different account. He would say that conscious pains are representations of pains (where the unconscious first-order pains can be analyzed in a roughly Dretskian fashion).
So in Dretske and Rosenthal’s accounts of conscious contents, the conscious representations have a target other than themselves, a target I would be perfectly happy to call the intentional object of the representation. While I’m not saying either of them has provided a final theory of consciousness, this aspect of their theories is plausible.
So, even though the pain arguably doesn’t wear its object on its sleeve (e.g., a caveman didn’t know what a stomach ache picked out, but just had an unpleasant experience in his midsection), that doesn’t mean it has no object. Perhaps phenomenologists like Cranky, by focusing on examples from the visual system (which provides us with an experience of clearly individuated objects), get it wrong.
Note I bring this discussion here because Cranky’s response when I asked questions hoping for clarification was
the civilized “You are an idiot please never post comments at my blog
again.” So, I thought I would bring the discussion to a site more
amenable to actual discussions.
I follow this rule of thumb:
-emotions are always intentional states because they direct to something (e.g. object, person, event…)they have an aboutness.
The only caveat i have is that despite the suggestion made by Aristotle some 2000 years ago, considering “pain” as a sensory emotion, the actual view based on current sicentific evidence is that pais is not an emotion but an homeastatic signal.
I don’t think that pain is mental state at all. It is something that we feel. And feeling is intentional act, which might have pain as intentional object.
Now, we might say “I’m in pain”, but what does that mean except that we feel some pain in some place of our body (or in phantom limb, or something), and that this pain is affecting us somehow?
That those two things – the feeling of pain, and how it affects us might be separated, can be seen in examples of pain asymbolia.
I’ve always had a problem with the philosophical notion of intentionality. As far as I’m concerned, any phenomenal event, even if it is not a palpable object (e.g., pain, color, taste, etc.), necessarily occupies a particular spatio-temporal location in the brain’s representation of egocentric space and should be considered an intentional event. For example, in my model of the cognitive brain, selective attention (the heuristic self-locus) would be *directed at* the locus of pain.
I’m confused by your first para. I consider all conscious states to be mental states (though not vice versa).
As for your second paragraph, I am talking exactly about pain in some part of the body.. E.g., the burning sensation when boiling water hits your hand. I don’t think we are ever just in unfocused “pain.” Pain asymbolia is interesting, though it just seems to show that there are dimensions to <i>normal</i> pain that can be disrupted by damaging a brain. E.g., I have never seen it argued that one of these aspects of pain (the aspect that remains for pain asymbolia) is intentional while the other aspect (typical mental responses to that aspect) is not. I’m not sure why we should be surprised that pain has a structure: color vision has an interesting structure revealed by bichromats etc..
To explain my first paragraph: The pain is something we might feel, ‘feeling’ is the intentional act, and pain the intentional object. Same as when we see a rabbit, the rabbit is the intentional object, and the seeing intentional act. So same as rabbits are not mental states I don’t think that pains are mental states.
I agree that those are two aspects, but not ‘of the pain’, but of the feeling of the pain. So, the hurting of the pain, doesn’t belong to the pain itself, but to how it affects us, which is part of feeling pain. Again to make an analogy, when we hear some music, and it affects us somehow (e.g. makes us nervous), the nervousness is not aspect of the music, but is aspect of our hearing of the music.
Hope I was more clear this time.
OK, I see what you are saying. It seems to be similar to Rosenthal’s view.
But isn’t there another layer of ‘aboutness’ there? The pain is an object of a conscious act of feeling (so pain is an intentional object), but what about the pain? Would you say it is merely correlated with, or informative about, bodily damage, but is not about bodily damage the same way that the feeling is about pain?
I sort of agree with Arnold about the word intentionality. It is an old overextended termthat is fractionating, i.e., being analyzed into multiple more well defined phenomena (as was the general term ‘memory’ which has fractionated into semantic memory, working memory, etc).
I think the word intentionality is weird too. I discuss this elsewhere in the comments.
Sure. We can talk about some aboutness in the case of pain itself. E.g. we can say that pain means body damage, similarly to how we say that smoke means fire.
But this doesn’t, seems to me, have anything to do with the intentionality, which as I understand is supposed to be about the aspect of our mental acts to be directed towards some object. (One can’t see without seeing something, one can’t love without loving somebody, one can’t feel without feelings something, etc). Nor it makes the pain itself (as object of an intentional act – that of feeling) a mental state.
So, I don’t know… Depends on the words we want to use, we can talk about two different levels of ‘aboutness’ here, but seems to me that both are of different nature. One is the intentionality where the act is directed towards something (we are feeling the pain), and there is some objective correlation where the pain is correlated (or means, or is caused, or whatever) by the bodily damage.
So for you, the ‘feeling’ of pain (which is a kind of attitude taken toward pain) is doing the real mental work. I wonder if, in Rosenthal’s related view, only the higher-order representations have intentionality, or if the first-order representations have intentionality but just not consciousness.
So do you think there is unfelt pain? I.e., pain that we are not conscious of? It seems so, so then for you the title of the post should be ‘Is felt pain an intentional state?’ In which case the answer is yes. I probably should have been more clear in the title, as I even mention Rosenthal’s theory which includes the possibility of first-order unconscious pain states.
Incidentally, Dretske’s information-based view wouldn’t grant fire-aboutness to
smoke. In his view, information is necessary but not sufficient for a state to have intentional contents. For him (and me) a pain state isn’t just a correlation with body
damage (if that were the case, pain receptor activation in the toe
would count as being about the toe), but is part of a larger story. At
any rate, I don’t want to sidetrack this discussion into a deep
analysis of Dretske. I discuss his views in a little depth here.
Yes, I think there is unfelt pain, illusions of pain, and all kind of phenomena which we notice for other intentional objects (e.g. there are unseen rabbits, illusions of rabbits, hallucinations of rabbits, dreams of rabbits, etc…). Examples would be… sometimes when we have prolonged pain, it may happen that we are distracted for some time, after which time we might check if the pain is still there. This commonsense attitude points that at least on intuitive level we see pain as something that is there even if we aren’t aware of it. We have illusions of pain like in Thermal Grill illusion, etc…
But, as I said, I don’t believe that pain (be it felt or not) is a mental state of any kind, same as I don’t believe that rabbits are mental states.
Maybe ‘feeling a pain’ might be a mental state, but then, I’m not sure even about that, as the analogue ‘seeing a rabbit’ doesn’t seem a mental state – it is an event in the world which includes the seer, the rabbit, photons bouncing off the rabbit, and whatnot.
I agree that carrying information about something, isn’t same as being in intentional relation (at least not in the sense in which I’m using the word in the context of intentional acts/objects). Thanks for the link! Looking forward to read it.
Wait, ‘seeing a rabbit’ isn’t a mental state? Is a visual experience of the rabbit a mental state in your view?
Hi guys!
This is an interesting discussion.
By the way Eric, that guy seems like a real jerkoff…I don’t even read his blog…
But back to the discussion. On Rosenthal’s view the first-order pain states do not represent in the same way that intentional states do…the qualitative properties are homomorphic to the perceptible bodily properties and so ‘reflect’ them. They don’t represent them sentetially or propositionally as intentional states do and he explicitly denies that his view is a representationalist view afccording to which “the metal character of qualitative properties is intentional” (from “Sensory Qualities, Consciousness, and Perception”)
Richard: thanks for jumping in. So for Rosenthal, the first-order representations are nonpropositional, so intentional states must be propositional? Are the higher-order representations of the first-order representations propositional in his view? In that case, if that is sufficient for consciousness, why aren’t conscious pains intentional contents? Because the targets of those representations aren’t propositions? So would he only say that higher order propositional thoughts about other propositional thoughts are conscious intentional contents?
Even Dretske has the analog/digital distinction in which only digitally encoded information is a true ‘semantic’ structure that can be true or false etc..
Being raised by the Churchlands, I guess I don’t have the
strong view that intentional contents must be propositionally
structured, or that only ‘digitized’ information has true semantic properties. That is, contents may be nonpropositional all the way up. This may get at the crux of my lack of acceptance of the standard views of intentionality. Or at least my agnosticism.
I think that “John sees a rabbit” is (literally) true only if there is a rabbit, so, it can’t be that seeing a rabbit is a mental state, as the rabbit is necessary constituent of the whole event. But I can see how “John sees a rabbit” can be used metaphorically for the cases where it seems to John that there is a rabbit in front of him. (hallucination, illusion). But in those metaphorical cases, literally John isn’t seeing a rabbit.
(Probably this goes besides the topic,… but in short, as for ‘visual experience’ I don’t believe that there are such things as ‘phenomenal/visual experiences’ where the experience is to play role of representation and be characterized by what-it-is-like to have it. I think the word ‘experience’ properly refers to event in the world in which the subject takes part and by which event the subject is affected or from which gains some knowledge. If in need for some rough categorization – I guess I’m leaning towards naive realism and tend to agree with Ryle/Austin linguistic analysis of perceptual/mental terms.) Eric, I hope you won’t ask me to stop sending comments now because of the differing views 🙂
After spending a couple minutes pinching myself, I still can’t determine whether or not pain is “intentional”. This either indicates that I am either a bad phenomenologist or like others have mentioned, the concept of intentionality just isn’t very useful for talking about such events. Probably a bit of both.
The only way I can really wrap my head around pain is semi-Gibsonian. That is, it seems like the perception of pain is really like any other perception, namely, a pickup of information. When I pinch my skin, I pickup information about the state of my nervous system in the same way that when I perceive visually, I pickup information from the ambient optic array. Trying to squeeze intentionality in here just seems awkward and forced. I think the ecological approach is much more useful.
I’ve just kicked SwampMan in the nuts and now, teary eyed and grimacing, he writhes on the ground at my feet. It seems pretty clear that he’s undergoing pain, but it’s hard to see how, lacking an evolutionary history, he has states that serve the function of indicating tissue damage.
I was once on an operating table during a procedure that only required local anesthetic. My wife, out of great empathy and curiosity, was holding my hand during the procedure. I was very nervous about the procedure, and found that as the procedure wore on, I was becoming more and more nervous. She noticed that I was squeezing her hand tighter and tighter, and finally she leaned over and asked me “are you in pain?” I suddenly realized I wasn’t nervous, I was in pain! A lot of pain! “Yes” I said “I’m in pain!” The surgeon snapped to attention, called for more anesthetic, and as the numbing buried the pain I felt far less “nervous”.
If interpretation is intentional, so is pain.
Hee hee….
Unlike (some versions of) Dretske I am not wedded to historical accounts of function.
In practice we never confirm functional claims in neuroscience by doing evolutionary biology, but by doing functional decompositions a la Cummins and Bechtel. We pick out a higher-level thing we have good reason to think is important biologically (e.g., the ability to behaviorally discriminate between sour and sweet), and then once that higher-level thing is fixed as our explanandum, we do a Cummins-style functional decomposition. It works well, and we don’t have to become fuzzy-headed evolutionary biologists.
If a bunch of neuroscientsts were handed 100 swampmen, and told they were swampmen, we’d still be perfectly happy saying their sensory transducers function to convert light into voltages or whatever. This would actually be an interesting thought exercise: what kinds of things would actual neuroscientists say about swampman? What about swampman and swampwoman’s kids? Would the sperm of swampman have a function?
I like the efforts Schroeder has made to resuscitate control theory inspired analyses of functional claims (I think he has only two papers on this topic, both published in 2004: Functions from Regulation in The Monist and New Norms for Teleosemantics in the book Representation in Mind). I think he has his finger on something true, at least as true as Wright functions and Cummins functions, and am surprised it hasn’t received much attention. I agree with him that people jettisoned the Weiner function prematurely. I am a pluralist wrt function I guess.
Ha! I did that to him last week. Will he ever learn?
Hmmm I thought you were going to say, since you had a pain you weren’t conscious of, pain isn’t intentional, or something like that. An interesting experience, especially to think about for people who say our judgments about what we are experiencing can’t be false.
Good answer. (That’d be my take as well). Unfortunately, I was eating breakfast when I got to the part about the swamp sperm.
NSFB*
.
.
.
*Not safe for breakfast
Poor Swampy. Those Dretskians and Millikanians are always kicking him in the nuts. And the Cartesians? Fuggetaboutit.
In the honor of this weekend, Go Patriots. I’m flying to New Hampshire tonight to watch Sunday with the fam…
Eric, you ask,
“So for Rosenthal, the first-order representations are nonpropositional, so intentional states must be propositional?”
Yes on both accounts
“Are the higher-order representations of the first-order representations propositional in his view?”
Yes, the higher-order thought characterizes the first-order state using the concepts at the creature’s disposal…it is propositional…
“In that case, if that is sufficient for consciousness, why aren’t conscious pains intentional contents? Because the targets of those representations aren’t propositions? So would he only say that higher order propositional thoughts about other propositional thoughts are conscious intentional contents?”
For Rosenthal a conscious pain consists in having a higher-order thought to the effect that one is in pain. A conscious pain just is being conscious of oneself as being in pain. I would take this to mean that the conscious pain is an intentional state, even though the first-order state that one is conscious of is not.
As far as swampman goes, Millikan never denies that the kind of approach that you talk about is good or useful. Nor does she deny that swampman has functions in that sense. What she does deny is that that kind of function is ALL there is to biological explanation.
Thanks a lot for clarifying Rosenthal’s view. It seems fairly standard that ‘intentional contents’ is used as a synonym with ‘propositional contents.’
True about Millikan. Swampman just can’t have bits with Proper Functions. Swampman is quite Improper.
If the brain’s ‘intentional contents’ is synonymous with its ‘propositional contents’, would the intentional object be a propositional symbol or a sensory event to which the symbol refers?
That’s exactly the kind of case that Rosenthal appeals to in arguing that pain and what it’s like to be in pain come apart (his example is the Dental Fear phenomenon, but Jim’s is a nice example to…though I wouldn’t want it to happen to me!!!!)…at any rate, why would you think that this kind of example would cut against the intentionality of pain?
And, yes Eric for Rosenthal ‘intentional’ and ‘propositional’ are synonyms…
Arnold,
I’m not sure I get what your question is, but as far as I can see to say that a state has intentional content is just to say that the state is semantically evaluable…what exactly that means, I suppose, is open for debate…
O.K. What does the phrase “semantically evaluable” mean for YOU? I’d be interested in what this means for others as well.
Usually it means it has certain ‘semantic’ properties such as being true or false, extension (e.g., if it is a propositionally structured thought, some items in the proposition refer to things in the world (e.g., ‘mom’ has my mom as its extension)), and sometimes ‘usability in an inference’ (this is especially popular for advocates of propositional attitude psychology). Some would say truth is reducible to reference and syntax (i.e., the truth-value of a propositional thought is fixed by the the extension of the terms used, and their syntactic relations to one another).
Some might add ‘intension’, but for historical reasons much of semantics has focused on reference and truth rather than intension. (Intension is any part of meaning not captured by reference, so ‘super man’ and ‘clark kent’ have same extension different intension). Note all of these categories are really inherited from talking about semantics of natural language, and it is the tradition in phil mind to import them into the head and assume they also apply to cognitive contents.
I’m not sure any of this is what Richard meant, but I think it is pretty standard.
Semantically evaluable means true or false when applied to sentences, satisfied or not when applied to predicates. It is pretty standard.
So:
1. An intentional state is a semantically evaluable state.
2. A semantically evaluable state is a propositional state in which propositional sentences can be evaluated as true or false, and predicates can be evaluated as satisfied or not satisfied.
If I’ve got it right, this seems to expose a weakness of the notion of intentionality when we try to understand human cognition.
First, with regard to deciding whether a proposition is true or false, in closed formal systems the decision is simply determined according to the formal rules of the system. But in a natural open system like the cognitive brain, the subjective validity of an internal proposition may be decided by a multitude of internal criteria — some innate, others culturally determined. Second, with regard to deciding whether a predicate is satisfied or not satisfied, it is claimed that the extension of the predicate must truly exist in the world. Does this mean that if I think “An ant is smaller than a unicorn”, I am not in an intentional state? It seems to me that by the nominal criteria given, I would not be. If this is the case, the formal concept of intentionality does not seem to be very useful in understanding the cognitive brain. We are in much better shape when we can point to circuit diagrams of putative biological mechanisms that can be shown to correspond to sentential propositions, examine their neuronal predicates and point to their sensory or imagined referents within a minimal model of the relevant brain mechanisms. Then we can empirically test the causal competence of each candidate model and explore the consequences of its particular design.
>First, with regard to deciding whether a proposition is true or false,
in closed formal systems the decision is simply determined according to
the formal >rules of the system.
I’m not sure what you mean by a closed formal system. Is sentential logic a closed formal system? If so, I’m not sure what you mean by saying that the formal rules of the system will tell you whether some atomic sentence is true or false.
>Second, with regard to deciding whether a predicate is satisfied or not
satisfied, it is claimed that the extension of the predicate must truly
exist in the >world.
I’m not sure about this idea here either. I’m no expert in this area, but I think philosophers have been working for about a century on what to say about sentences with terms like “unicorn” or “the present king of France”.
But, stepping back, I think that intentionality is often not invoked as an explanatory hypothesis, but as some feature of the mind or the cognitive brain that, at least prima facie, has to be explained. There is a large philosophical literature devoted to trying to understand how it can be that thoughts are about things. So, it might not be that big a deal if it does not help you understand the cognitive brain, since it is a feature of the cognitive brain that is noted as something one wants to understand.
YEp, that’s the kind of stuff that I meant. So, for instance, a sensation of red cannot be true or false and (arguably) does not have an extension but my thought that I see red does.
Arnold, as Ken points out, there are many different attempt to spell out what exactly is going on when you think ‘an ant is smaller than a unicorn’, which is why I said what exactly one means is debatable…but one plausible hypothesis is that one is in a mental state that can be evaluated as saying ‘there is a class of things called ants whose (typical/usual/normal/whatever) representitives are smaller than the representitives of the class of things called unicorns’. We might represent this formally as Ex Ey (Ax & Uy & S(x,y), where ‘A’ stands for the predicate ‘is an ant’ and ‘U’ stands for the predicate ‘is a unicorn’ and ‘S’ stands for the relation ‘smaller than’. So your thought can be true or false, and its [redicates do or do not have extensions (/are or are not satisfied), etc. But again, whether it is evaluated as true or false depends on one’s theory of predicates like ‘unicorn’, like Ken says.
But, I am not as sure about Ken’s claim that intentionality is not invoked as an explanatory hypothesis. Isn’t this exactly the way it shows up in, say, Brentano (I am no expert on Brentonao! I can’t even spell his name!! :)). It is invoked as an explanation for certain of our behaviors…but then of course it does present an independent problem of how anything could exhibit intentionality in the first place…
Kenneth wrote: “There is a large philosophical literature devoted to trying to understand how it can be that thoughts are about things.”
Then if I understand you correctly, an intentional state is *not* necessarily a sentential propositional state. It is just a thought about something. So if I simply have the experience of something far away from me, it is an intentional experience because it is about a relationship between something and me. If this is the case, then wouldn’t any perspectival experience be an intensional state? Or are you defining *thought* as an exclusively linguistic event?
My claim was weaker than you suggest, Richard. I wrote, ” I think that intentionality is often not invoked as an explanatory hypothesis”. In the 100+ years of philosophical discussion of intentionality, I figure lots and lots of things have been tried. Intentionality could be a terrible or useless explanans, but is a plausible explanandum.
>Then if I understand you correctly, an intentional state is *not*
necessarily a sentential propositional state. It is just a thought
about something. So if I> simply have the experience of something far
away from me, it is an intentional experience because it is about a
relationship between something and >me.
Well, I am not an expert on this, but I figure there are some philosophers who think you can have an intentional state with just some property, say, red, or individual, such as a tree, as its content, but then there are other philosophers who think that intentional state contents must be propositional, e.g., there is something red, there is a tree. I don’t know what would favor one view as opposed to the other.
>If this is the case, then wouldn’t any perspectival experience be an intensional state?
Philosophers have debated this one. Is pain perspectival? Is it intensional?
>Or are you defining *thought* as an exclusively linguistic event?
I happen to believe that thoughts are linguistic in the sense that they have a combinatorial syntax and semantics, but not in the sense that they are elements of some natural language. There is one’s natural language and one’s “mentalese”. But, I don’t take that as a definition of a thought. That’s what I take to be a discovery of what thoughts are like. But, I know there are philosophers who reject combinatorial syntax and semantics for thought and reject a distinction between one’s natural language and “mentalese.”
When I write,
>”There is a large philosophical literature devoted to trying to understand how it can be that thoughts are about things.”
I am thinking, in the first instance, of things like Dretske’s teleoinformational semantics, or Fodor’s asymmetric causal dependency theory, or Rob Cummins’s picture theory of representation. But, there are tons of competitors. Incidentally, just today I sent in a draft of a Stanford Encyclopedia of Philosophy article on “Causal Theories of Mental Content.”
Ken