What does the relational predicate ‘… analyze…’ mean? To what kind of relation the term ‘analysis’ stands for? These are difficult questions, and a satisfactory response to them would be essential for those who take the analysis as ‘the right philosophical method’.
First: what is the object of philosophical analysis? Well, the philosopher is not really interested in answer a question of the form “What is P?” only by providing a linguistic definition of the term ‘P’. So, the object of analysis, the kind of relation that would be considered as the right kind of answer to that question, is not about verbal expressions at all. To provide definitions of verbal expressions is a lexicographer’s task. It is more proper to say that philosophical analysis, instead of being about verbal expressions, is about the concept associated semantically with these expressions. It all seems perfect here: the analytic philosopher has its proper object, one that differs in nature from the object of the cientist; we have conditions to not confuse their objects of study. But it’s not sufficient just to say which is the object of philosophical analysis: we need to know what is that object. What is the concept? Again, it is not sufficient to explain what is the concept, because we need to know what is the relation between concepts in analysis. It’s a lot of work. Still, we can make some important points on it.
It’s is generally stated that the concept associated with a predicate is a function which, if completed, becomes a proposition with truth-value. It is an intentional object that is expressed by predicates when used in non-oblique contexts, as in
Marcos is a biker,
where ‘… is a biker’ is a propositional function that, when completed with the singular term ‘Marcos’ (and also with other singular terms, I suppose) becomes a true proposition. We can talk of a epistemic analogue of the concept as a function: the concept as an abstract object, a cognitive pattern that fulfills the task of classifying objects. These two senses of talking about concepts do not conflict – they are just applied to different areas of philosophical inquiry. Well, and how can we analyze concepts? What is the necessary and sufficient conditions for a relation of analysis?
There are at least four conditions. And they can be expressed in the following way. The concept analysans and the concept analysandum stands is a relation of analysis if and only if:
(i) the analysans and the analysandum are necessarily coextensive
(ii) the relation between analysans and analysandum referred in (i) is knowable a priori
(iii) the term standing for the analysans is synonymous to the term standing for the analysandum
(iv) the analysandum does not occur in the analysans
The clause (i) is the coextensionality clause, and it says that every object that exemplifies the first concept, exemplifies also the second, and vice-versa. In formal logic we say that there is a biconditional operator between the two concepts, a relation of extensional equivalence. The clause (ii) is the knowability clause: if the relation of equivalence between those concepts were not knowable a priori, it would not constitute a philosophical analysis, but an empirical one. So that the sole undertanding of the concepts is sufficient to believe that they are equivalent. Clause (iii) is the synonymy clause: it says that the concept expressed by the analysans term and the analysandum term is the same. But there is an important point here: this clause means that, in ordinary, non-oblique, contexts those expressions have the same meaning, but if we change the issue to an intensional, oblique, context, we would say just that the expressions ‘the concept analysans‘ and ‘the concept analysandum‘ refers to the same object.
It is this important observation that seems to be missing in general accounts of philosophical analysis. We must remember that the philosophical analysis stands in a non-ordinary context, an properly intensional context. Thats the reason why the philosophy talk differs not only from the common use of words, but from scientific talk in which we can find a lot of definitions and extensional analyses.
Weel, the clause (iv) is just standing for the necessity of the analysis to not be circular. It says that it’s not valid to analyze, for example, the concept of human by saying that human is human and animal, because in this sentence the analysandum is occuring again in the analysans.
I offered an analysis of the concept of analysis, and I think that most of analytic philosophers would agree with it. But it has yet some explanatory problems. Still, we can be secure in using the analytic method with respect to its self-consistency.