Chalmers on the Explanatory, Conceivability & Knowledge Arguments

The Explanatory Argument

David Chalmers says that the “easy problems” of consciousness explain “only structure and function”. They don't explain consciousness. Therefore “no physical account can explain consciousness”.

The Conceivability Argument

We can conceive of a physical system that is note-for-note identical to us but which doesn't have consciousness. (Though what has the psychological notion of conceivability got to do with the problem of consciousness?)

Such as system would therefore be a zombie. Alternatively, it may be a zombie-invert in that some of its experiences are inversions of those of human beings.

The invert-zombie has the same nuts and bolts as us; though nevertheless it has different experiences. So the inverted zombie is still allowed his experiences.

There is also the conceivability of a partial zombie who also has experiences; though not as many as those of human beings – perhaps he can only feel pain.

The point is that all these zombies are physically identical to us from the third-person point of view and their behaviour will also be indistinguishable.

What about their first-person point of view? What is it like to be a zombie of whatever kind? Well, there's nothing it is like to be a bona fide zombie!

On a larger scale. What about a physically identical universe that doesn't, however, give rise to consciousness; though which does give rise to zombies? We can say that such zombies are indeed "naturally possible". However, according to our laws of nature, they probably couldn't exist. That is, given identical physical and bodily facts, then such a universe couldn't help but give rise to consciousness. (This is what some non-reductive physicalists and supervenience theorists believe.)

Let’s take this further.

There could be an identical universe that didn't give birth to consciousness. If this were the case, then consciousness must be something above and beyond the physical if such a counterfactual scenario were possible. In addition, if we can conceive of such zombies in our world or at other worlds, then Chalmers claims that it is "metaphysically possible" that there could be zombies.

What does metaphysical possibility add to the notion of conceivability?Chalmers codifies and simplifies this with a logical argument:

i) It is conceivable that P & not-Q.
ii) If it is conceivable that P & not-Q, then it is metaphysically possible that P and not-Q.
iii) If it is metaphysically possible that P & not-Q, then materialism is false.
iv) So materialism is false.

(Can a mere possibility make materialism false?)

Again, we can see Chalmers’ slide from conceivability to metaphysical possibility. Why should it be that simply because we can conceive of something then that something is metaphysically possible? This has an almost empiricist ring to it in that all conceivables (or ‘ideas’) must come from somewhere or entail metaphysical possibility.

Can we conceive a round square? No. Then it isn't metaphysically possible. Can we imagine a man with five legs? Yes. Then it's metaphysically possible.

The Knowledge Argument

To put the case simply. We could never, and have never, deduced or inferred consciousness from the sum of all physical facts. Though, then again, the same can be said about water. We could study H2 O until the cows come home; though we would never deduce (or infer) the reality of water - its wetness and drinkability - from such physical facts. We can only do so a posteriori – in this case, through science, which tells us that water is indeed H2 O. Though no scientist has ever (or could ever) infer or deduce water’s wetness, etc. a priori. (Or could they? Is water’s transparency an emergent property?)

Does Mary lack knowledge about red? She obviously lacks the experience of red. Is the experience of red a ‘fact’? What sort of fact would it be? Frank Jackson argues that if she finally came to actually experience red, she would learn a ‘new fact’ about red which must be over and above her knowledge of its physical basis and even beyond her powers of deduction from such facts. That is, experience emerges from the physical; though it can't be read off from the physical. That is the essence of emergentism.

Jackson concludes that Mary does indeed know all the physical facts; though not all the facts. There must be non-physical facts (one of which is consciousness).

The strong conclusion to all this is that

i) If there are more than physical facts

ii) and that these things can't be deduced from physical facts

iii) then materialism must be false.

That's because materialism only allows physical facts in its world-picture.

The Shape of the Arguments

Chalmers then attempts to codify and simplify the arguments in strictly logical terms.

Firstly we can think in terms of epistemic entailment, deducibility, explicability and conceivability.

Let us take epistemic entailment.

This is a priori entailment or implication in that it doesn't depend on (further) experience. If we have

P ⊃ Q

we have a material conditional from the physical facts to an arbitrary phenomenal fact. When we know that P is the case, then we must also know that Q is the case without further experience.

In the case of consciousness, P doesn't entail or imply Q a priori. We can't deduce Q from P. Similarly, we can conceive of P without thereby also conceiving of Q. Or, in functional terms, if P is functional, then we can't deduce Q from P because consciousness “is not a functional concept” (as we saw earlier in this debate).

These logical uses of the material conditional can also be applied to the conceivability argument, the knowledge argument and the explanatory argument.

Taken one by one.

If we can conceive of zombies, then zombies are metaphysically possible. If we can't deduce consciousness from all the physical facts, then some facts - those of consciousness - aren't physical. If physical explanations aren't adequate, then there must be non-physical facts that require non-physical explanations.

Now we can talk of another kind of entailment: ontological necessitation.

We can say that P necessitates Q. In the material conditional P ⊃ Q, we can say that P can't hold without necessitating Q. It is ontologically necessary that P necessitates or entails Q. Again, if this were the case, then materialism would be false.

The other interesting point about these arguments is the movement from an epistemic gap to an ontological gap. More precisely, we can argue that:

1. There is an epistemic gap between physical and phenomenal truths.
2. If there is an epistemic gap between physical and phenomenal truths, then there is an ontological gap, and materialism is false.
3. Materialism is false.

The obvious point to make here is the slide from the epistemic gap to an ontological gap. What does that mean?

If we can't slide from our knowledge of P to Q, then that must be because P and Q are ontologically different. If P and Q were ontologically of the same order, then we could move, epistemically, from P to Q.

Why does a lack of epistemic movement from P to Q entail ontological difference? Couldn’t that epistemic gap be accounted for simply in terms of our epistemic limitations or our inadequate knowledge or physical devices? We at one time couldn't move from H2 O to the wetness of water. That epistemic or scientific gap didn’t engender an ontological difference between H2 0 and water or even water’s wetness. Does a lack of knowledge about X entail the fact that X is ontologically weird or irreducible?

David Chalmers on the Hard & Soft Problems of Consciousness

 

David Chalmers makes the helpful distinction between the ‘hard’ and ‘soft’ problems of consciousness.

The soft problem includes the ability to discriminate stimuli, or to report information, or to monitor internal states, or to control behaviour. Prima facie, you can see that these problems can be accounted for in third-person or scientific terms. For example, someone can tell us how they discriminate stimuli with his verbal reports. All this can also be explained neurobiologically. The same goes for the reports of information that we can achieve or carry out. (The problematic ‘easy problem’ is the monitoring of internal states which, prima facie, doesn't seem scientifically kosher.)

 

The ‘hard problem’ is the problem of experience. There is something it is like to experience this, that or the other. What does that mean? It means that we are phenomenally conscious of this, that or the other. Mental states also have their own feel, as it were. When we are having a mental state (if we do indeed have distinct mental states), there is something it is like to be in that state.

What are conscious states in the first place? They include states of perceptual experience, bodily sensation, mental imagery, emotional experience, occurrent thought, etc. There is something it is like to perceive a red apple. There is something it is like to feel a toothache. There is something it is like to imagine Tony Blair. There is something it is like to feel depressed about something. There is even something it is like to make an arithmetical calculation. Of course we will now need to explain what exactly we mean by saying that ‘there is something it is like to…’ and why this makes consciousness ‘unique’ and something non-reducible to the physical.

Chalmers is more precise about this phrase ‘what it is like’. He introduces explanatory technical terms to do so.

For example, conscious states have a phenomenal character with phenomenal properties that characterise what it is like to be in that state.

What, then, are phenomenal properties? The conscious state of perceiving a red apple has phenomenal properties. The properties include the experience of the colour red or the feel of the apple. A toothache can be ‘nagging’ and other pains can be sharp or blunt. The mental image of Tony Blair will have certain phenomenal properties, such as the redness of his face or the emotional responses to the image. Similarly, depressions may be painful or tiring and may engender other phenomenal properties such as lethargy or anger. And even an arithmetical cognition can be accompanied by certain emotions such as that of excitement or tedium.

Chalmers then asks a fundamentally different and equally important question about consciousness.

“How and why do physical processes give rise to experience?”

Intuitively, consciousness, experience or mentality doesn't seem physical at all. How is my mental image of Tony Blair in any way physical? Is my sensation of the red of a red apple in any way physical? And so on. The relation between the physical and the mental is problematic because we can imagine that all the physical processes and events could happen without giving rise to any conscious experience. Such functions and processes “could occur in the dark”. Chalmers calls this problem the “central mystery of consciousness”.

Let's get back to the ‘easy problems’.

What are the easy problems and why are they easy?

These problems are about certain behavioural or cognitive functions. If we lift one leg and are conscious of lifting one leg this can be explained in neurological and neurophysiological terms. As for cognitive functions, these too can be explained in neurological terms, or at least in principle they could be. More explicitly, we can talk in terms of the causal role of a function in a cognitive system and how such causal roles, within the brain, cause us to lift one leg or make a logical inference from p to q. The function takes on the form of a causal role in the production of behaviour and we need only see what exactly the mechanisms within the brain and body are that carry out such causal roles.

In terms of the brain, Chalmers states that such mechanisms are neural or computational in nature (why computational?). As for examples of these functions or causal roles, we can site discrimination, integration, access, report and control. We can intellectually distinguish a cow from a horse. We can integrate new knowledge with old knowledge. We can access our memory system or even past mental images. We can report internal states and past perceptions. We can control our behaviour through cognition as when we build a house. Though, alas, why are any of these mental functions accompanied by experience?

And so we have ventured on to the hard problem of consciousness.

What is the relation between these physical functions in the brain and experience? Such a relation would need to abide by ‘natural principles’ if it is to stand the test of materialism or physicalism. What is it about physical processes that bring about states of consciousness? Why do they do so in the first place?