By BRENT SILBY
Copyright (c) Brent Silby 1998
Higher-order thought theories (or HOT theories) state that a mental state is conscious only when it is accompanied by a higher-order thought (HOT).1
The thought that makes a state conscious is not conscious in itself, but having that thought is what make the state of which it is about conscious. If there is no HOT about a mental state, then that state is not a conscious state. On this view, a state can only be a conscious state if we are aware of that state, or, have a HOT about that state.
Fred Dretske, on the otherhand, believes that mental states can be conscious without the subject being aware of those states. For Dretske, what makes a mental state conscious is not the subject's awareness of the state, rather, it is the way in which the state makes the subject conscious of something in the world. When we look at an object, we are not conscious of our experience of looking at the object, we are conscious of the object itself.
In this paper, I will suggest that HOT theories lead to a cluttered picture of the mind. I will then offer Dretske's account of consciousness as an alternative to the HOT theories. My goal will be to show that Dretske's model of consciousness should be the preferred option.
Higher-Order Thought Theories of Consciousness
According to David Rosenthal, for a state to be conscious, it has to be the object of a higher-order thought. Rosenthal believes that these higher-order thoughts are usually unconscious and this is why we do not notice them. We would only notice a second-order thought if it was accompanied by an even higher-level thought, which itself would be unconscious. The reason Rosenthal claims that the higher-order thought must itself be unconscious is because if it was conscious, we would have a potential explosion of higher-order thoughts. We would have to ask: what is it that makes the higher-level thought conscious? The answer would be that it is also accompanied by an even higher-level thought, and that thought is accompanied by still a higher-level thought. By stating that the higher-level thought is unconscious, Rosenthal has found a way of preventing such an explosion of increasingly higher-level thoughts.
On Rosenthal's account, a mental state is conscious only when we are conscious of being in that state. This means that every conscious mental state is accompanied by a higher-order thought. At first glance, this seems like a reasonable account of conscious experience, but I find it hard to accept. For a start, the higher-order thought theory seems to lead us back to a warmed over Cartesian model of the mind. The theory seems to suggest that there is some intelligent process, over and above the primitive workings of the mind, and this process somehow focuses on mental states thus making them conscious. This opens up questions that are very difficult to answer; for example, how is this process instantiated? How does the process choose which states to focus a higher-order thought upon? What is it about a second-order thought that makes a mental state conscious?
The second problem with the higher-order thought model is that it seems implausible to suppose that every one of our conscious experiences is accompanied by a higher-level thought. Think of your experiences of every bit of detail in your visual field. Surely each of these experiences cannot be accompanied by a higher-level thought. Such a multitude of higher-level thoughts would seem to be uneconomical and unnecessary in the design of a cognitive system.2
Having a second-order thought about every detail in the visual field would be an unnecessary redundancy. Supposing that every detail of experience needs both a first-order and second-order judgement leads us to a very complex and cluttered model of the mind. A more economic construction of the mind would keep experiences localised at the first level of thought without requiring higher levels of thought to make them conscious. Before accepting a higher-order thought model, we should consider an alternative view that does not require higher-order thoughts to make a state conscious.
Fred Dretske draws on a distinction that was made by Rosenthal. He contrasts creature consciousness with state consciousness. A being has intransitive creature consciousness is it is a conscious being. A being has transitive creature consciousness if it is conscious of things, or conscious of facts. This is where the first distinction is to be discovered. Creature consciousness in the transitive sense, requires a being to have states of consciousness, or state consciousness. For Dretske, the purpose of creature consciousness is obvious. If you take it away, then the being cannot see, hear, smell, or be conscious of anything in its environment. To say that an animal is creature conscious is to state the obvious. If an animal is a conscious being, then it is conscious of things in its environment - it can see, hear, smell, and so on. The interesting side of the distinction involves state consciousness. This is what makes a being creature conscious.
What is Dretske talking about when he speaks of state consciousness? What makes a metal state a conscious state? There seem to be two possible answers. First, we could say that a mental state (S) is conscious because we are conscious of it. Alternatively, we could claim that a mental state (S) is conscious because we are conscious with it. Dretske believes that the second option is the one to go for because it more accurately captures the nature of state consciousness. The first choice is a common misconception about the nature of conscious experience. It treats the mental state (S) as an object of consciousness. On this line of thought, we would have to say that: "I am conscious of this conscious state." and this is a problem. We do not want to say, for example, that a cat is aware of its experience of a dog. Surely the cat is simply aware of a dog, not of its experience of a dog. For Dretske, a mental state is not an object that we are conscious of having. Rather, it is an act that is making us conscious of something. An animal or human can be conscious of something without being conscious of the fact that it is conscious of something.
What characterises state consciousness? Dretske breaks state consciousness down to two types of awareness - object-awareness and fact-awareness. We can be aware of objects without being aware of facts about those objects. For example, you might be looking at a tree upon which a camouflaged stick insect is sitting. You are not aware that the insect is there and yet you are seeing it. The information is entering your visual system and is being processed. On Dretske's account, you are object-aware of the insect but not fact-aware of the insect. You see it, but are not aware that it is an insect. This leads Dretske to a nonepistemic category of seeing, or "seeing without knowing". You can see something even if you do not know that you see it. Dretske offers various examples of object-awareness without fact-awareness. In one of these examples, he asks us to look at a pair of pictures.3
The two pictures are almost identical and contain a random scattering of large spots. The only difference between the two pictures is that one of them contains an extra spot. When people see these two pictures for the first time, they usually do not notice the difference and think that the pictures are identical. But the difference is there. Dretske believes that this is an example of object-awareness without fact-awareness. When we look at the pictures, we are object-aware of the difference (the extra spot in the first picture), but we are not fact-aware that there is a difference. This leads to the conclusion that differences in conscious experience need not be reflected in conscious belief.4
Although Dretske refers mainly to object and fact-awareness in vision, his distinction should also apply to other sense modalities. In the case of hearing, for example, we can point to similar cases. There is a situation that most people have found themselves in at some time. Imagine you are at a party and are engaged in a conversation with someone. There are a lot of people talking in the background but while engaged in conversation, you are not aware of what their voices are saying. This is to say you are not fact-aware of their conversations. However, the sound of the voices is being picked up by your ears and is being presented to your auditory system, so on Dretske's account you are object-aware of their conversations. On the face of it, one might claim that it is not obvious that there is any type of awareness of the background conversations at all. Surely if you are not aware of what all these people are saying, their conversations are not being heard. But Dretske would claim that they are being heard. You are simply not fact-aware of them. Notice what happens if, on the other side of the room, someone mentions your name. The sound of your own name rings like an alarm bell in your mind. Suddenly you become fact-aware of part of a background conversation. Now, this effect would not occur if the background voices were not being processed somewhere in your auditory system - if you were not object aware of them.
Assessing The Distinction Between Object and Fact Awareness
Dretske has told us that we can be object-aware of items in our visual field without being fact-aware of those objects. On his account, when we look at the plant containing a camouflaged stick insect, we see the insect even though we are not fact-aware of it. We do not see it as a camouflaged insect, but it is still being presented to our visual systems and so we are object-aware of it. Daniel Dennett objects to this point. He asks: does it follow that if you could have seen the insect, then you did see the insect?5
This is a good point and it betrays a problem with the use of the word `see'. It is far from clear that the stick insect is seen at all. The visual information is entering the eye and is being projected on the retina, which is sending signals to the visual cortex, but is this enough to count as `seeing'? It would seem that something else has to occur. There has to be some sort of conceptual uptake. For Dretske, this conceptual uptake would lead to fact-awareness but before this occurs, there is object-awareness. In the experiment involving the two almost identical pictures of spots, Dretske claims that we are object-aware of the extra spot in the first picture, but we are not fact-aware of the difference. In cases where the difference is not noticed, it is obvious to say that the difference in the two pictures is not registered or conceptualised in any way. So Dretske is right to say that in these cases there is no fact-awareness of the difference. But can he really claim that we are object-aware of the extra spot in the second picture? Can merely having something presented to your visual system give rise to object-awareness? It would seem that being object-aware of something would do us little good on its own, and does not tell us much about conscious experience. Our introspective view of our conscious life seems to point only to fact-awareness. Our seeing something involves conceptualisation and categorisation, and this only occurs in fact-awareness. Anything that we are object- but not fact-aware of, seems to have little to do with directing our behaviour and attention. Without being fact-aware of an object, the object seems to be invisible. Can object-awareness really count as `awareness' in any sense of the word? Dennett suggests that when visual information contained in two slightly different pictures is present in the primary visual cortex, but is not being used or "looked at" by other "homuncular agents", then the only sense in which the differences in the pictures is being registered is the sense in which the differences would be registered inside a video camera that is not recording.6
For Dennett, this does not count for much. It certainly does not count as `seeing' in any interesting sense of the word.
Dennett's objections are well founded, but this is not to say that Dretske's account is incorrect. The problem lies in our use of words such as `seeing', and `awareness'. Dennett admits that if Dretske simply means nonepistemic seeing in the sense of a video camera that is not recording, then he is welcome to the concept. However, for Dennett, this view does not provide a convincing model of conscious experience. This may be true, but is Dennett right to compare object-awareness to a video camera that is not recording? If object-awareness is like that, then Dennett would be right in thinking it an unconvincing model of conscious experience. The information entering the visual system would not be used by other "homuncular agents" and as such would not be registered in any way. But remember the party scenario. You were not fact-aware of the background conversations, but when someone on the other side of the room mentioned your name, you suddenly became fact-aware of that fragment of conversation. You were object-aware of the background conversations, but you were not attending to them. You were not fact-aware. It seems that object-awareness is necessary to give rise to fact-awareness. If the auditory information from the outside world was merely being presented, but was not used by any system in the brain, it would seem that fact-awareness could not have occurred at all. Something in the brain must have been listening. Rather than compare object-awareness to a video camera that is not recording, it may be better to describe it in terms of a video camera that is recording but contains a very short tape, which loops and re-records every few seconds. The tape is being monitored and conceptualised in a very low level way, but when something important appears on the tape (like the sound of your own name), the information is passed to other systems for further analysis. When this happens, fact-awareness occurs.
Fred Dretske has offered us an account of consciousness that should be taken seriously. He has given us reason to believe that consciousness does not consist only in our awareness of our mental states, or our experience of those states. HOT theories state that a mental state is conscious only when it is accompanied by a higher-order thought. If there is no higher-order thought about a mental state, then that state is not a conscious state. Dretske, on the otherhand, has shown that it is possible for a subject to be conscious of something in the world without being aware of any facts about the thing in the world. This conclusion was drawn after making a distinction between object-awareness and fact-awareness. What makes a state conscious is the fact that it makes us conscious of something in the world. We can be object-aware of things in our environment even if we are not fact-aware of those things. The point that Dretske wants to make is that we can see things, which puts us in various conscious states, without being aware of any facts about those things. HOT theories lead us to a cluttered and uneconomical picture of the mind and should be avoided. Dretske's account offers us a way of describing what it is that makes a state conscious without appealing to higher-order thoughts. He has not offered us a complete theory of consciousness, but he has shown us "where not to look for it."7
Chalmers. D., The Conscious Mind, Oxford University Press, 1996.
Dennett. D., "Dretske's Blind Spot".
Dretske. F., "Conscious Experience" in The Nature of Consciousness, Edited by Block. N., Flanagan. O., Guzeldere. G., MIT, Bradford 1997.
Rosenthal. D., "A Theory of Consciousness" in The Nature of Consciousness, Edited by Block. N., Flanagan. O., Guzeldere. G., MIT, Bradford 1997.
7 Dretske. F., "Conscious Experience" in The Nature of Consciousness, Edited by Block. N., Flanagan. O., Guzeldere. G., MIT, Bradford 1997, Page 786.