One of the most persuasive objections to the identity thesis—the claim that mental states are numerically identical to certain brain states —is the argument from complete knowledge. It goes like this:

1. If mental states were identical to physical brain states, then a complete physical description of a brain state would also be a complete description of the identical mental state.
2. Any physical description of a brain state must fail to completely describe a mental state.
3. Therefore, mental states are not identical to brain states.

The argument for premise (2) usually involves an appeal to subjective experience, or the felt quality ("qualia") of an experience. For example, a color-blind neuroscientist could know all the facts about the brain state of someone who is having a sensation of red, but having never seen red, the neuroscientist still would not know something about the sensation, namely, what red looks like.1 Likewise, the most complete knowledge of the workings of a bat's brain could never allow us to know what the bat's experience is like.2

The standard response to this objection on behalf of the identity thesis is to claim that the objection commits the fallacy of equivocation. One response harks back to Quine's distinction between two types of knowing: knowing-that and knowing-how. Knowing the physical facts of the brain's activity is an example of the former, goes the argument, while being able to recognize a sensation as red is an example of knowing-how.3 Another version calls the two types of knowledge "propositional" and "prelinguistic", respectively.4 The point is that there are two types of knowledge of one thing being pointed to, not knowledge of two different things, and that the objection thus fails to refute the identity thesis.

This response has always seemed forced and unconvincing to me, especially when an alternative that is much simpler but at least as effective is so ready to hand. It is this: the demand that a complete physical description of a brain state be a complete description of a mental state is equivalent to the (unreasonable) demand that a person describing something become the thing described.

To understand this, all that is required is the acknowledgment that it is one thing to describe X, and something else to be X. Suppose I were to give a complete physical description of the pen in my hand, in all the detail that anyone could desire. Someone might say, "Your physical description may well be complete, but you still have left something out, namely, what it is to be the pen, and therefore the pen must be more than just a physical thing." Such a claim would be regarded as so absurd as to scarcely merit a response.

Yet this is precisely the type of claim that the knowledge argument makes. The reason this is not immediately apparent in the case of brain states and mental states is that the describer and the thing described are both the same type of thing, i.e. brains. The absurdity in the knowledge argument is not apparent, simply because in the case of brains, unlike pens and everything else, we do understand what it means to be one. Yet it is just as much the case with brains as with pens that it is one thing to describe it, and something else to be it. And it is just as unreasonable in the case of brains as with pens to demand that a description of something somehow equate to being the thing.

To say that the color-blind neuroscientist can have complete knowledge of the physical facts of the brain state of seeing red, yet not know what it is like to see red, is to say nothing more than that a brain's knowing all the facts of another brain's state does not entail being in that state itself. Likewise, it should be perfectly obvious that a human brain, in knowing all the facts about a bat's brain, does not thereby become the bat's brain. In neither case is there a reason to conclude that there is something nonphysical about that other brain's experience.

The standard response to the complete-knowledge argument is on the right track. But rather than distinguish between two ways of knowing one thing, it seems much simpler, and much more to the point, to distinguish between knowing something and being it.

© 2003 James T. Anderson

---

1.Frank Jackson, "Epiphenomenal Qualia", Philosophical Quarterly vol. 32 (1982), pp. 127-132.

2.Thomas Nagel, "What Is It Like to Be a Bat?", Philosophical Review 83, no. 4 (October 1974), pp. 435-50.

3.Peter Carruthers, Introducing Persons (London: Croom Helm, 1986), ch. 5.

4.Paul Churchland, Matter and Consciousness (Cambridge, MA: MIT Press, 1984), pp. 33-40.