Notes on Howard Robinson’s ‘The irrelevance of intentionality to perception’

Three characteristics of intentional verbs

1. Opacity: “one cannot substitute, salva veritate, in the object position different expressions that refer to the same entity”
2. Indeterminateness: “the characterization provided by the verb’s object can be vague and indeterminate”.
3. Nonexistence: “the object expression can fail to refer and the sentence can nevertheless express a truth”

According to Anscombe, these three characteristics interconnect with each other.

Two factions of the intentionalist camp

A. Sense-content theory: “sensory states are intentional by assimilating sensory awareness to some other, prima facie different mental activity which is indubitably intentional” (Armstrong, Pitcher)

B. non-reductive intentionalism

Anscombe’s argument:

1. The question of the ontological status of direct objects (especially whether they are “in language” or in the world) cannot be raised (p. 164).
2. Intentional objects are a species of direct object. (p. 162)
3. The question of the ontological status of intentional objects cannot be raised (p. 164)
4. The contents of sense-experience are a species of intentional object (p. 169)
5. Therefore, the question of the ontological status of sense-contents cannot be raised.
6. The contents of experience are a species of direct object.
7. The notion of a “direct object” is a purely grammatical notion.
8. The notion of a sense-content is a purely grammatical notion.
9. 1.7 At least some grammatical notions are not purely linguistic notions.

Anscombe’s argument for 1 relies on the answers to the following question

a. She asks whether direct objects are “bits of language” or what the bits of languages stand for.

b. the answer is neither.

c. there are no other “reasonable candidates” for being their ontological category.

— Here is how she dealt with b

‘John sent Mary a book’

a) What is the direct object of the verb in this sentence?

b) What does the sentence say John sent to Mary?

c) What does the phrase which is the answer to these questions communicate to us, i.e., is it being used or mentioned?

d)Is the direct object a bit of language or rather what the bit of language stands for?

Anscombe’s answer: “A book” for a, b, c. “Neither, for this is not a proper question” for d.

1. a book is not a real book because there is no answer to the question “which book?”

— my comments: weird. No answer for “which book” because of no enough information. But I guess Anscombe’s point is that the answer indicates indeterminateness.

2. ‘a book’ is not a phrase because we could not admire Zeus as a phrase.

Robinson: we could give different answer to a-c, Anscombe is confused about objects of verbs and objects of actions. So b is untenable and premise 1 collapses.

My comments: even if Anscombe were right about the answer to those questions, she only established the verb ‘send’ to be intentional, which has nothing to do with perceptual verbs.

Robinson’s positive argument

T’: Whenever S seems to see something F he is aware of the quality F…

Robinson tries to establish T’ by proving

1. That it is correct to say that S is aware of an F quality-object whenever he seems to see something F.
2. That the sense or way in which he is aware and the nature of the object are empirical or concrete in the way required for a sense-contents theory
3. it is vacuous to take ‘aware’ in this formula as intentional.

—1

The possibility of ostensive teaching →1: “talk of “what it is like” in the case of a quality just is a way of referring to the quality itself. Thus one is acquainted with, or aware of, qualities in such experiences.”

My comments: ostensive teaching seems not possible when experience is not veridical at least for some non-veridical cases such as hallucination. Ostensive teaching is compatible with 1' S is aware of an F quality-object whenever he sees something F.

— 2

Objection: When S seems to see something F and is aware of quality F, then this quality is an abstract entity or universal. (I guess intentionalists would say this concerns hallucination)

Response: empirical in the sense that exclusion is possible.

— 3

1. S is aware of an F quality-object.
2. The F quality-object occupies or blocks an area of S’s visual field.
3. The F quality-object does not exist.

Robinson thinks that 3 does not fit well with realism because it means that there is no thing (individual) which is F, which amounts to the suggestion that sense-contents are not particulars.

My comments: can we have token property? Why properties must be universals?

Robinson argues that if ‘intentionally inexistent’ means ‘not physical’, then the thesis becomes uninteresting, philosophical idle. Given 1 and 2, the problem of veil of perception could not be avoided as the traditional sense-data theory.

Indeterminate →intentional, Robinson argues this inference fails.

“The empirical nature of the phenomenon requires a high degree of determinateness.”

Typical intentional activities such as thought, could be very indeterminate. For example, we could think of colour without thinking of a particular colour.

‘Intentional’ is characterised under some descriptions. The giveness of perception radically restricts the range of aspects under which something can be seen. Non-perceptual intentional verb could have a wider descriptions such as thought. Non-substitutability fails because quality-pattern does not possess any aspects of the sort of which someone could be aware but of which S is not aware.