(Cross-posted at M-Phi.)
Greg Frost-Arnold has a nice post on the origins of the phrase ‘analytic philosophy’, in particular on when it began to be used roughly with its current meaning. He has a useful chart showing that already in the 1960s the phrase was being regularly used, whereas ‘continental philosophy’ only became more prominent as a phrase in the 1990s.
But besides the question of uses of the terminology, in comments to the post different people have been presenting insightful remarks on the origins of the very idea of ‘analytic philosophy’ as a particular way of doing philosophy. Greg himself (in comments) offers the following observation as a starting point:
Go back to Europe in 1932. We have the following 3 intellectual groups: the phenomenologists (esp. Husserl and Heidegger), the folks dedicated to (something like) Moorean analysis, and the Vienna Circle and their intellectual allies (perhaps we include here the Lvov-Warsaw school).
He then goes on to argue, and correctly to my mind, that analytic philosophy emerged essentially from the ‘fusion’ of the last two groups, the Mooreans and those who viewed logical analysis as the main philosophical methodology (Vienna Circle etc.). Continental philosophy would have naturally emerged from the phenomenology tradition. (An aside: for a while, I entertained the hypothesis that the analytic/continental divide could be explained by observing which of the Critiques each tradition focused on: mostly the first for analytic philosophy, mostly the third for continental philosophy. But Eric Schiesser told me at some point that this hypothesis doesn’t work, and although I forgot his arguments, I found them compelling then.)
Anyway, I think Greg hit the nail on the head in identifying Moorean ‘common sense’ philosophy and logical positivism (or at least a strong emphasis on the role of logic as a methodology for philosophy) as the two main sources for the development of analytic philosophy as we know it. But this also means that this tradition has always had a mild schizophrenic component, so to speak (and I hope disability philosophers will not take offense here!). In a sense, Moorean common sense philosophy is inimical to the conception of logical analysis in question, which at least to some extent *is* about discovering new facts.
Let me make this more precise. In comments, Greg quotes the ‘statement of policy’ of the very first edition of the journal Analysis, in 1933:
"the contributions to be published will be concerned, as a rule, with the elucidation or explanation of facts, ... the general nature of which is, by common consent, already known; rather than with attempts to establish new kinds of facts" (p.1).
Clearly, it doesn’t get more Moorean than this… And why am I saying that this is in tension with the project of using logical analysis as a key philosophical methodology? After all, one can (and does) also use logical analysis for the explanation of facts which are already known. However, as I see it, this is definitely not the core of the good old Leibnizian idea of using logic to discover new truths, which was to a great extent one of Frege’s main sources of inspiration (although it is true that the very logicist project can be seen as the search for a logical analysis of already known mathematical facts; but it seems to me that this is really just the beginning of the story).
Let me mention a couple of examples. In ‘On Denoting’, when Russell claims that the actual logical form of sentences such as ‘The present king of France is bald’ is not the subject-predicate form but rather a tri-partite claim, he is clearly questioning ‘already known facts’ by common consent. Similarly, the Tractatus is filled with un-commonsensical statements. A little later, in his works on truth and logical consequence, Tarski does take something that he himself refers to as the ‘common conception’ of these concepts as his starting point to formulate criteria of adequacy for his formal theories. But it’s clear that these formal theories are intended to go much beyond than just offering an elucidation of these ‘common conceptions’. In the same vein, Carnapian explication does not in any way rule out the possibility of establishing ‘new kinds of facts’ about its object of analysis (which may well be already known facts, but as a starting point).
It is no secret to anybody that I am no enthusiast of Moorean intuition-based philosophical methodology (as explained here and here), so I am not neutral on any of this. But the genealogy offered by Greg Frost-Arnold seems to me to highlight the fact that this tension between sticking to (and explaining) what is known and discovering new facts (in particular, by means of logical analysis) has always been at the heart of analytic philosophy, insofar as it emerged from the confluence of two rather distinct approaches to philosophy. This explains for example the reactions to Michael Streven’s definition of philosophy in a post at Leiter’s blog: “The point of philosophy is to defy common sense.” (For the record, I couldn’t agree more.) It’s Mooreanism still alive and kicking! But modern counterparts of logical positivism are also alive and kicking, and it is surprising that people are not more aware of the intrinsically conflicting nature of philosophical methodology within analytic philosophy.
(Let me also add a little plug for Mike Beaney's great entry on the concept of analysis at SEP.)
UPDATE: Jeff Bell has a recent post very much in the same spirit of this post, 'In praise of the incredulous stare'.