Expressions and Entities

Of course, Place notes, we don't need to jump straight to the worry that it’s conceptually impossible for a table to be a packing case. That seems absurd. The reason, one might surmise, that we cannot analyze the concept TABLE into PACKING CASE is because it’s not the case that all tables are packing cases. It’s ok if some are, but the concepts will still be logically independent of each other. If it were so that every table were a packing case, then the sentence, ‘His table is a packing case’ would seem analytically true, and the ‘is’ in that case would be the ‘‘is’ of definition'; to be referrerable to by the term 'table' is just to be referrable to by the term 'packing case'. As Place has it, this kind of analyticity is fundamental to language. We'll call it the A-rule, or (A):

Assume that a kind of object/state of affairs has two properties (or two sets of properties) x (x = the property of being red) and y (y = the property of being colored). Where x is unique to the kind of object/state of affairs in question (only the class of red things are red), the expression used to refer to x, namely ‘red’, will always entail the expression used to refer to y, namely ‘colored’.

If (A) were an exception-less rule for language, any expression logically independent of another expression which uniquely characterizes a kind of object (as in 'table' and 'packing case') would necessarily refer to a characteristic which is not associated with the entity in question. But since (A) applies almost universally, we are usually warranted in arguing from the logical independence of concepts and expressions to the ontological independence of referents. Hence the intuition that consciousness is not a brain process; again CONSCIOUSNESS and BRAIN PROCESS appear to be logically independent concepts.

Crucially, Place argues that there are cases in which (A) fails: (A) fails where the verification conditions for the instantiation of two different properties (or sets of properties) can not be satisfied simultaneously. Note that, for instance, CLOUD and MASS OF TINY PARTICLES IN SUSPENSION appear to be conceptually/logically independent. However, this doesn’t provide us with grounds for asserting that a particular cloud and the mass of tiny particles constituting it are ontologically independent entities. There is one thing, a cloud, picked out by logically independent concepts and expressions. This is just because we can’t satisfy the verification conditions for claims like ‘That cloud x is a huge fluffy, white, fleecy-looking mass’ and ‘That cloud x is a mass of tiny particles in suspension’ simultaneously. We observe the cloud as a fluffy white mass from afar, and as a mass of tiny particles from within it. In fact, as Place notes, we have different words/concepts to capture the objects to which we aim to refer. In this latter case, being inside a cloud, we call it a ‘fog’; being outside a fog, we call it a ‘cloud’; but a fog just is a kind of cloud and vice versa.

For this reason, we might say that it is synthetically (by which I mean contingently) true that a particular cloud is a mass of tiny particles (and nothing else). But it is analytically true that, for example, a particular bachelor is unmarried; the former expresses an ‘is’ of composition, and the latter an ‘is’ of definition. Also, to repeat another example, it synthetically true that ‘His table (by composition) is an old packing case’, whereas it is analytically true that ‘A square (by definition) is an equilateral rectangle’.

What else can we say about the identity expressed by the ‘is’ of composition? It’s worth noting, as Place does, that in the case of the cloud, raw visual observation will suffice to verify that some cloud or other is a mass of particles. So there is the obvious continuity between one’s cloud-observations and mass-o'-particles-observations, where the one virtually seamlessly becomes the other as one approaches. But our case is more difficult; the verification conditions for something’s being a conscious state and those for something’s being a brain process appear to be mutually exclusive and totally disparate. That is, the verification conditions for one never verify the other.

Consider, with Place, the example of lightning. A bolt of lighting is an instance of electrical charges in motion. We observe it visually, but we don’t observe the charges. No matter how hard we try, we couldn't; verifying the existence of the charges requires special methods. What warrants our assertion that lightning is electricity in motion? In what sense are the disparate requisite observations for verification of each side of the relation observations of the same phenomenon?

Correlation will not suffice. Two things can be correlated perfectly without being identical. For instance, in a coin-tossing contest, the correlation between my tossing heads and my opponent doing so can have a correlation of 1 if every time I throw heads he does, too. But our tosses are not identical. Instead, we should say that two different observations are observations of the same thing when one explains the other. More specifically, we verify that we are observing the same phenomenon via different methods when our technical observations gel with theory so as to provide an adequate explanation of some phenomenon. So,

Lighting is the motion of electric charges (and nothing else).

might be a true statement if it becomes apparent that the motion of electric charges through the atmosphere causes what the untutored eye perceives to be lightning; and, of course, even the tutored eye will never see the electrical charges, in and of themselves, moving.

Three ‘is’s’: definition, composition, and the ‘is’ of predication.

The initial worry Place alludes to is that statements about conscious states are not analyzable in terms of statements about brain processes. Behind this:

(i) Conscious states are describable independently of our awareness of brain processes.
(ii) The verification conditions for statements about consciousness and brain processes are entirely independent.
(iii) There is nothing self-contradictory in the thought that some subject has a conscious experience without any correlated brain process.

There maybe problems among these, but, as Place notes, even granting (i), (ii), and (iii), it doesn’t follow that the claim ‘consciousness is a brain process’ is necessarily false. Rather, ‘consciousness is a brain process’ is neither self-contradictory nor self-evident.

Consider the following sentences, which have the feature that the subject expression and the object expression appear to be adequate characterizations of the same things/states of affairs:

(D1) A square is an equilateral rectangle.
(D2) Red is a color.
(D3) To understand an instruction is to be able to act appropriately under the appropriate circumstances.

(C1) His table is an old packing case.
(C2) Her hat is a bundle of straw.
(C3) A cloud is a mass of particles in suspension.

Contrast these with the following examples:

(P1) Toby is eighty years old.
(P2) Her hat is red.
(P3) Giraffes are tall.

Place claims that one reason for holding that D-sentences and C-sentences have the feature that the subject term and object term are adequate characterizations of the same thing is that in both cases we can reasonably add the clause ‘and nothing else’ to them. Compare:

(D1*) A square is an equilateral triangle and nothing else.
(C3*) A cloud is a mass of particles in suspension and nothing else.

Compare with a P-sentence:

(P3*) Giraffes are tall and nothing else.

P-sentences don't seem to have the property in question. This is plausibly just because the relation does not seem to be between independently adequate descriptions of the same things.

Aside from the similarity pointed out, D-sentences differ from C-sentences in that the former appear to be necessarily true by definition and the later only contingently true (on condition of verification). For D-sentences, there is a logical/conceptual/analytic relationship such that the subject term picks out something contained in the predicate term. Presumably, this is what Place means by the ‘‘is’ of definition.’

By contrast, for the ‘‘is’ of composition’, there is no such (apparently) analytic relationship between the terms. The two terms involved seem to adequately describe each other (for a given context), but it is only contingently true that they do so. This goes even for (C3), which stands out because it lacks an indexical component.

Why assume that we can deny the claim that 'consciousness is a brain process' on strictly logical grounds in the first place? The argument that Place is criticizing, initially, is this:

(1) If the meanings of two expressions differ, they can’t both provide an adequate characterization of the same object.
(2) There is no contradiction in conceiving of a conscious state which is not underwritten by a brain process.
(3) The meanings of ‘consciousness’ and ‘brain process’ differ (are conceptually independent).
(4) Therefore, ‘consciousness’ and ‘brain process’ cannot be adequate characterizations of the same thing.
(5) Therefore, consciousness is not a brain process.

Place digs in right at the beginning; at premise (1), offering a kind of error theory that might help to motivate the identity theory. It is a mistake to assume that where meanings of expressions differ, that both cannot adequately describe the same thing, in effect, there are true, contingent identity statements. After all, if we were to subscribe to the faulty assumption (says Place), then we might wind up in the weird position of having to assert that it is logically impossible for any table to be an old packing case; since, there is no contradiction in asserting that someone has a table but not a packing case. The same would presumably go for the impossibility of any hat being a bundle of straw, or any cloud being a conglomeration of particles. [t-b-cont’d]

Welcome!

Welcome to the PCRG. Thanks for your continued interest. Post and respond at your convenience.