Reading Phil draft

In what sense is Anscombe』s exploration grammatical rather than ontological?

Anscombe』s philosophical explorations focus on the intentionality of perception and action. Her basic assumption is that intentional action/perception is necessarily reflective of the agents』 practical reason and will, that is to say, the agent should be conscious of her intentional experience; there』s no sense in saying 「she unconsciously intend to...」. So that a certain sort of first-personal involvement is a necessary item for an action/perceptual experience to be called 「intentional」. Any action/perceptual experience could be articulated as a formal unity consisting of the intentional subject and the intentional object. (to 「intend」 is precisely to direct an intentional object to its subject) The grammatical nature of Anscombe』s approach lies precisely in her theory that: intentional subject and intentional object are entirely grammatical/logical concepts bearing no direct relation(that is, relations such as causality, reference and identity) to other non-grammatical object; instead of classificatory concepts (like proper names and demonstratives) which are signs of other non-grammatical entities.(e.g. the proper name 「stone」 is used to donate that round, solid stuff made out of calcium carbonate) The functions of grammatical/logical concepts, as suggested by Russell, are exhausted by: 「two things to grasp: the kind of use, and what to apply to them from time to time.」() Interestingly (and perhaps unsurprisingly), the methods Anscombe employs for the two pieces of construals for intentional objects and intentional subjects both largely involve reductio ad absurdum: the classificatory use of these two concepts is denied by revealing the absurdity and logical impossibility that result from its adaptation. The positive claims regarding intentional subjects/objects』 grammatical function, on the other hand, are given quite speculatively with slight arguments. That is probably because Anscombe pre-assumes that labels 「classificatory」 and 「grammatical」 are sufficient to give an exhaustive account of the possible natures of signs, so that the rejection of one would automatically bring out the other.

It should not be a surprise that in the case of 「intentional object」 Anscombe chooses direct object as the subject of her argument. Direct object is defined by Anscombe as the object of the verb of a sentence, e.g. , the direct object of the sentence: 「John sent Marry a book」 would be 「a book」. For Anscombe intentional object(the object of intentional verb in a sentence) is a grammatical subclass of direct object, so that linguistically they are of the same nature. Anscombe』s argument runs as follows: if direct object is a classificatory concept which is a sign for other non-grammatical entity, it could neither be a bit of language nor what the language stands for. Because if it is a bit of language, that will lead to an obviously absurd assumption that the objects of our dealings are merely some phrases; but apparently we intend for these words to contain additional properties and functions other than mere signs themselves,otherwise languages would lose their meanings. And if they are what the language stands for, that would imply that the phrase would have the entire set of properties (or at least usages) as the material entity itself; and that』s improbable because, for example, we cannot sensibly answer the question 「which book?」 regarding the direct object 「a book」 of the sentence presented above which is itself hypothetical. (that is to say, that sentence is not necessarily an observational sentence, every direct object of every randomly constructed sentence is entitled to enquiry) Nor could this word stand for the 「impression」 or 「idea」 of a book, since: 「as they (impression and idea) enter into epistemology they will be rightly regarded as grammatical notions」 (Anscombe 1981 p.9) Here Anscombe seems to suggest that the introduction of 「impression」 and 「idea」 sheds no additional light on the above epistemological difficulty purely regarding the usage of the word; so it』s only a far-fetched ad hoc to explain away it, whose implausibility could be revealed simply by our intuition: 「if a sentence says John sent Marry a book, what it immediately and directly says he sent her was a direct object, and only in some indirect fashion, via this immediate object, does it say he sent her a book.」(Anscombe 1981 p.9)--- that process is evidently absent from the mind of every reasonable individual, and so that it is impossible that the usage of the word is derived from such kind of consideration.

The exploration regarding the 「intentional subject」 starts with Anscombe』s dissection of first-personal reports. It is her claim that the first person pronoun 「I」 is not a proper name for the entity we pick out as the subject for various 「I」 sentences whenever we utter them. Rather, as already said, it is a grammatical concept whose functions are exhausted by its usage for constructing and organizing conversations. Her argument runs as: If we imagine a scenario in which the use of 「I」 is eliminated and replaced completely by the word 「A」, and 「A」 is a proper name used to refer to that entity which we call 「self」, there would be two logical implausibilities conceivable for the execution of such an approach: 1. 「A」 could be misused while 「I」 would not. Because when someone gives a first-personal report using the name 「A」 to refer to herself, another person, hearing these words while not aware of the presence of the person around her who just said them, may well be startled and confused, and think that she has just spoken to herself unconsciously. (「Oh no! Did I(A) just talk in sleep? Or did that piece of consciousness of me(A) talking just slide away?」) Yet the possibility of such kind of case, insists Anscombe, is excluded among 「I」 users since it』s impossible for one person (in normal situation) to completely mistake her identity for another. 2. The referent of the first-personal pronoun could hardly be a definite entity; it cannot be the physical body or a collection of memory, because first-personal reports can still be given by a completely paralyzed person whose physical body is undetectable to her; so is the case for a person suffering from serious amnesia who has no trace of her previous memories whatsoever. Now, since the use of 「I」 seems to be compatible with the absences of these two items, they could not be 「I」』s intended referents---a Cartesian Ego seems to be the only choice left. But this is absurd---the existence of a Cartesian Ego is not plausible because of 「the intolerable difficulty of requiring an identification of the same referent in different 『I』-thoughts.」 ()---the collections of the perceptions and the mental states under which every piece of 「I」-thought is produced are more or less discrepant, so it』s extremely improbable for a distinct and unchangeable ego that one can grasp from time to time to exist.

Since intentionality is necessarily reflective of the agent』s practical reason and will, it is a plain constural that only through first-personal reports---or at least the agent』s ability to produce them, is the presence of intentional items---intentional subjects and objects guaranteed. So that the presence of intentional subject and object entails two sorts of consciousness: other-consciousness and self-consciousness, the former allows us to have 「unmediated, agent-or-patient conceptions of actions, happening and states」---resembling what is called 「conscious qualia」 in modern philosophy; and the latter brings out the particular patient that receives that former experience---the 「I」 user, i.e., that person here. In Anscombe』s opinion, the function presented above gives the full account of the roles/usages intentional subject and object serve in conversation. This also explains Anscombe』s reading of the Platonic line: 「He who sees must see something」---that something, instead of being a concrete entity, is an intentional object---a grammatical concept whose existence is ensured by the agent』s capability to generate a relevant first-personal report as 「what I see is_」--- the blank must be filled by something, for it would be unintelligible for a man to consciously report to have a perceptual (intentional) experience that lacks content.

In conclusion, it is clearly illustrated that part of Anscombe』s project is to dismiss the intentional concepts』 direct relations with ontology by showing that they are grammatical concepts reflecting first-personal properties of intentional actions rather than classificatory concepts reaching towards non-linguistic realm. And also, in Anscombe』s project, our usage of the language is the 「given」---the exact subject of exploration. That it is never her task to question about our intuitive ontological understanding that lurks behind our use of language marks, precisely, the distinction between the undertakings of a philosopher of language and of a metaphysician in the traditional sense.

推薦閱讀：