As I mentioned earlier I am sitting in on David Rosenthal’s Mental Qualities class and I wanted to jot a few things down in the wake of the class while they are fresh in my mind.

What is this class about? The title is ‘mental qualities’ but what are those? Rosenthal has been suggesting that the mental qualities are those properties of mental states that are produced by the senses. So this would include visual qualities, auditory qualities, tactile qualities, etc but also bodily sensations like pain, itches, and tickles. On this way of doing things it will be the case that there is no cognitive phenomenology.

Before we begin a little in the way of full disclosure. Those who know me, or have looked around this blog long enough, might know that I personally do think that there is cognitive phenomenology (see here). In fact I am tempted to think that what makes a mental state mental is that it has some qualitative property, and that there is thereby something that it is like for me to be in that state when it is conscious (note that is perfectly compatible with the claim that the quality can occur unconsciously and that there is nothing that it is like when it does), but leave that aside. It is controversial whether there are mental qualities associated with thoughts. Indeed, we would only need posit them if there were cognitive phenomenology and we thought there must then be some corresponding mental quality in virtue of which there was phenomenology when the quality was conscious. But why think that there is cognitive phenomenology in the first place? Can we even make sense of what it would mean for the to be cognitive phenomenology? Rosenthal doesn’t think so.

His central challenge seems to be that there is no well defined thesis worth defending in the area. We seem to have a handle on what it means to say that there is sensory phenomenology, but what do we mean when we say that there is cognitive phenomenology? If we are tempted to say, as I would be, that we mean that there is something that it is like for one to have cognitive experiences then David will reply that this phrase is useless. It is used in many different ways by many different people. So we can ask again ‘what does it means to say that there is something that it is like to think a conscious thought?’

One thing we might mean is just that thoughts can occur consciously. But if that is all that we mean, Rosenthal says, then it is not interesting. Everyone knows that thoughts can occur consciously. So we must mean more than that thoughts occur consciously. But what else do we mean? Well, that there is something that it is like to have the thought. But why think that? One line of argument is that we can find cases where the sensory qualitative properties are held constant and we vary the intentional content we get a difference in phenomenology (this line of argument was pushed by Zoe Jenkin). He admits that this can happen (and that it is common) but denies that it supports the existence of cognitive phenomenology.

So, suppose that I am looking at my dog Frankie and that I clearly see her. I thereby have a conscious visual experience as of certain shapes and colors, etc. But I also perceive that she is a dog. Rosenthal conceded that the phenomenology in this case is distinct from that of a similar experience with different intentional content. So if in one case the intentional content is ‘that there is a dog’ (or whatever) and in the other it was ‘that there is an animal of some kind’ then we would have a difference in phenomenology. But this does not show us that there is cognitive phenomenology. Phenomenologically the two properties (that is the qualitative and intentional properties) seem to be intertwined, or co-mingled, in some intimate way. In fact they are so co-mingled that one might think that it is a mistake to talk about qualitative properties without talking about intentional properties and vice versa. Since this is true in the case of perception we may be tempted to think that this is also true in the case of thoughts. This would be one way to give content to the claim that there is cognitive phenomenology. But it does not by itself give us a reason to think that the two properties are the same.

He then suggested that we have some reason to think that the two properties, though intimately intertwined, are distinct properties. He argued that since we theorize about these properties in very different ways we have prima facie reason to think that the two things are different properties. So, he concludes, treating them as the same kind of property doesn’t buy us anything theoretically.

There is a lot more to be said about all of these issues, but I’ll save that for another day.

On Wednesday I attended the inaugural session of the Graduate Center’s philosophy colloquium.  The speaker was Joe Levine and he wanted to examine two of the arguments for the phenomenology of thought as given by people like David Pitt and Charles Siewert and argue that they were not up to the task that supporters thought they were.

The two arguments were what he called the self-knowledge argument and the phenomenological argument. The self-knowledge argument claims that the only way we could have genuine acquaintance-like knowledge of our cognitive states was if they had a phenomenology. Levine rejects this argument as question begging. The second argument he takes more seriously. The phenomenological argument points to several distinct kind of phenomena. So, take an ambiguous sentence like ‘visiting relatives can be boring’. When one understands it to mean that the relatives who are visiting are boring and when one understands it to mean going to visit relatives is boring there seems to be a difference and this difference intuitively seems to be phenomenal. Or take listening to someone speaking a language you don’t understand versus one that do. When people are speaking a language you do not understand it often sounds as though they are speaking really fast and that there are no spaces or pauses in their speaking but this is very different from listening to a language you do understand. The idea is supposed to be that there is a distinctive cognitive phenomenology that goes beyond any associated internal monologue or mental imagery. Levine admitted that he felt there was something to theses kinds of cases and argued that intuitively it is just as string an intuition as that there is something that it is like to see red or feel pain. I agree. The question, then, is what does this force us to conclude about the phenomenology of thought?

As a contrast Levine introduced a null hypothesis, what he called the Non-Phenomenal Functional Representation thesis. NPFR, as he calls it, is basically a standard higher-order view about self-knowledge. When one knows what one is thinking one tokens a higher-order state the content of which is that one is in the first-order state. This is why the self-knowledge argument doesn’t really pull any weight. Both camps have an explanation of how we have self-knowledge. What about the phenomenological argument?

In order to respond to this Levine distinguishes two versions of the claim that there is a phenomenology that is distinctive to thought that he calls a pure and and an impure view. On the pure view there is a phenomenal character of an occurrent thought that is not tied to any sensory state while on the impure view “attributes phenomenal character only to sensory states, but allows that cognitive states can create phenomenal distinctions among otherwise identical sensory states,” (from the handout). The pure view is the just the usual idea that there is a distinctive phenomenology for thought. The impure view is a bit harder to get ahold of but the basic idea seems to be based on an analogy with the way sensory states work. So, take the higher-order view about consciously seeing red. On the HOT view there is a first-order sensory state that has phenomenal character and then there is a higher-order state that represents oneself as being in a red sensory state. One can a higher-order thought to the effect that one is in a generic red state or that one is in a specific red state and this will determine what it is like for you to have the experience but the HOT itself has no phenomenal character. So by analogy then Levine’s impure view seems to be that we have a first-order state, say a hearing or seeing  of ‘visiting relatives can be boring’ and one’s higher-order state can then represent it as either being about the relatives coming or your going to them and this will result in a distinctive phenomenology. That is to say that on the impure view what it is like to hear the sentence will be different when one is aware of it one way or the other but there is no cognitive phenomenology. All there is is two different kinds of auditory phenomenology.

I think my own view about cognitive phenomenology is similar except that I think that this can happen in the case of a propositional attitude and not just through some sensory state. For instance when one has a conscious belief that p I claim that it will be like believing that p for you and this is because one is aware of oneself as believing that p. This makes it a version of the pure theory. So, is there any reason to prefer the impure theory to the pure one? Levine argued that the phenomenological argument supported only the impure account and so it was no reason to think that the pure view was correct. His idea seemed to be that since the data was hearing a sentence one way versus hearing it another we only had evidence that there were two different ways of hearing the argument.

At the end of his talk he introduced another distinction between transparent and opaque cognitive phenomenology. On the transparent view “what the cognitive state is about, what it is representing, constitutes the “look” of the cognitive state while on the opaque view there is only a contingent relationship between what is represented and the cognitive state. The issue here seemed to be diagnosed buy whether one thought that there was any possibility that one could find out that one was radically mistaken about what one thought. His example seemed to be the standard brain in the vat scenario. If one came to be convinced that one was a brain in the vat, or that Quinian indeterminacy of referce, were correct one might come to find out that one was radically wrong about what one thought. On some theories of mental content one wouldn’t be mistaken, but let that slide. The point he was trying to make was that we could “wrap our heads” around the idea that our cognitive states are not transparent. He compared the opaque view to Block’s view about mental paint.

During discussion Levine discussed a comparison with people like David Pitt and Sussanna Seigel. Seigel argues that the content of our perceptions is richer than we thought (e.g. it is part of our perception of a tree that it is a tree)  and in so doing end up making perception more like cognitive states while people like Pitt argue that thoughts have a phenomenal feel and thereby make thoughts more like perceptions. This led some to wonder how we might distinguish between the two states. On my own view this is wrong headed. What we should take this as is a trajectory towards a unified account of the mark of the mental.