Investigating Functionalism Functionalism is a theory about the nature of mental states. According to functionalism, mental states are functional states that are defined by their functional roles (causal relations to other things) rather than their intrinsic qualities. To illustrate the difference between the two ways of defining something, entities such as clocks (to display time) and calculators (to manipulate numbers) are defined functionally, whereas entities such as water (H20) and gold (Au) are defined by their physical structures.
In the case of mental states, what it is to have a particular mental state is to have a state that plays a particular functional role. To put it in another way, functionalism compares mental states to black boxes connected to other black boxes and to various inputs and outputs – each black box distinguishes itself from the others by its distinct relations to other black boxes, perceptual inputs, and behavioral outputs; what is inside the black boxes does not define what they are.
For example, for someone to believe that it is raining is for the person to be in a state such that the state is caused by the perception of rain (perceptual input), and, along with the desire to stay dry (another black box), causes the person to carry an umbrella (behavioral output).
While the functionalist theory of mental states clearly explains mental states’ connections to other things, some philosophers would fret over what exactly is inside the black boxes: many have argued that functionalism cannot account for the subjective feel of conscious experience such as “the hurtfulness of pains, the itchiness of itches,” and “the smell of a rose” (Jackson, p. 311). Gilbert Harman, in “The Intrinsic Quality of Experience,” considers and rebuts three such arguments against functionalism.
In this paper, I will specifically explain (a) the second anti-functionalism argument about color perception, (b) how Harman exploits the functionalist theory of mental representations and concepts (not just mental states) to refute the argument, and (c) why Harman’s rebuttal is not convincing enough. The Anti-Functionalism Argument Consider a person blind from birth.
The anti-functionalism argument goes like this: (P1) The person knows all about the functional facts of color perception (how he is able to achieve that is irrelevant here). P2) The person does not know what it is like to see something red. Therefore, (C) what it is like to see something red cannot be explained in purely functional terms. If we assume that the premises are true, it is easy to see that the conclusion follows from the two premises. Suppose for the sake of contradiction that (C) is false. That is to say, what it is like to see something red can be explained in purely functional terms, and is therefore a purely functional fact of color perception.
Then it follows from (P1) that the person knows what it is like to see something red. But according to (P2), the person actually does not know what it is like to see something red, which is a contradiction. Therefore, (C) is true provided that both premises are true. Since the inference is correct, to invalidate this antifunctionalism argument, then, we will need to disprove at least one of the two premises. Harman accepts (P2) and attempts to disprove (P1).
Harman’s Reply Harman argues that a person blind from birth can never know all about the functional facts of color perception. Specifically, (1) a person blind from birth cannot know one important functional fact F about color perception, namely, that “there is a concept R such that when a normal perceiver sees something red in good lighting conditions, the perceiver has visual experience with a representational structure containing this concept R” (p. 651).
In short, Fis a fact about “what happens when a normal perceiver sees something red” (p. 651), or “how the concept R functions with respect to the perception of things that are red” (p. 652). If (1) is indeed correct, then (P1) is disproved and the antifunctionalism argument fails. From Harman’s paper, we can extract the following arguments for (1): (2) A person blind from birth does not have the full concept of something’s being red. (3) To be able to represent a fact to oneself requires having the relevant concept(s).
A key concept needed to represent the fact E mentioned in (C2) is the concept of something’s being red. Therefore (from 2, 3): (4) A person blind from birth is not able to represent the fact Fto him or herself. (5) To know a fact is to be able to represent that fact to oneself. Therefore (from 4, 5): (1) A person blind from birth cannot know the fact F. Harman suggests that according to the functionalist theory of the content of mental representations and concepts, (3) and (5) are true.
If (2) is true, then (4) logically follows from (2) and (3), and (1) in turn follows from (4) and (5). Now the only task left for Harman to prove (1) is to prove (2), which is not an easy task. To see how Harman proves (2), we need to take a closer look at the functionalist theory of mental representations and concepts. Similar to the functionalist theory of mental states, the functionalist theory of concepts supposes that what it is to have a particular concept is to have a concept that plays a particular functional role.
Someone has the full concept of something’s being red if and only if the person has available a concept that functions in the appropriate way, which requires that the person must be disposed to use that concept in forming visual representations as the natural and immediate consequence of seeing something red. Harman assumes that a person blind from birth has “no concept of something’s being red that could be immediately brought into service in visual representations of the environment if the person were suddenly to acquire sight” (p. 651).
If that is the case, then the requirement for the person to have the full concept of something’s being red is not satisfied, and thus (2) is true. However, in the next section, I will show that Harman’s line of thought in proving (2) is selfcontradictory. Harman’s Flaw Consider a similar fact F* that, according to Harman himself, the blind person can know: “there is a concept R such that, when a normal perceiver sees something that reflects light of such and such a frequency, the perceiver has visual experience with a representational structure containing this concept R” (p. 51).
Observe that the only difference between Fand F* is what the normal perceiver sees: instead of “something red,” * is concerned with “something that reflects light of such and such a frequency. ” If we run the same argument for Fon F*, we would conclude that the blind person cannot know F*, which contradicts Harman’s own conclusion, unless at least one of the premises fails. The only premise that could possibly go wrong in F*’s case is (2*): a person blind from birth does not have the full concept of something’s reflecting light of such and such a frequency.
To disprove (2*), we need to disprove a similar assumption that is used to prove (2), specifically, that a person blind from birth has no concept of something’s reflecting light of such and such a frequency that could be immediately brought into service in visual representations of the environment if the person were suddenly to acquire sight. But this assumption is equally true as the assumption used to prove (2), because a blind person from birth could no more visualize something reflecting light of certain frequency than he or she could visualize something red.
This shows that the way Harman proves (2) is somehow flawed, which hints at either problems with the assumption or deeper issues with the validity of the functionalist theory of mental representations and concepts. Conclusion Many philosophers doubt whether the functionalist theory of mental states can account for the subjective feel of conscious experience such as what it is like to see something red.
Referring to the functionalist theory of mental representations nd concepts, Harman specifically seeks to disprove the antifunctionalism argument that what it is like to see something red cannot be explained in purely functional terms. Even though Harman’s rebuttal is flawed, he may still be able to revise one of his assumptions to reestablish his argument. But if that is not the case, Harman will need to reconsider the extent to which one can generalize functionalism. Either way, it is still unclear whether functionalism can fully explain the subjective feel of conscious experience or not.
