How philosophy gets done
Brian Weatherson, responding to a query from Matthew Yglesias (the link to his query no longer works), writes on doing philosophy, and in particular on applying “old ideas to new fields”. Yglesias had written:
One of the things that’s dawned on me as I approach graduation is that for all the hours I’ve put into listening to lectures and participating in seminars on philosophy, I’ve never really had anyone speak to me on the topic of how, in practice, philosophy is done.
I’m generally of the opinion that expecting students to just “pick things up” is a bad idea; it tends to perpetuate the inequalities already existing among them when they enter graduate school. In order to pick something up, you have to know first of all that there is something you ought to pick up, and then figure out, from the data available to you, what that thing is. 
Some people are very good at picking up, for example, social norms; others are indifferent, by temperament or by class origin, to bourgeois niceties, but will have, if gently advised, no difficulty in becoming observant (in both senses).
Source: The Glasgow Story
The same goes for style in writing: some people are capable of figuring out how to write academic prose just by reading a lot of it; others, equally capable of writing decently, respond readily to advice but are not, at least initially, very good at the induction of norms just by exposure. It just may never have dawned on you, for example, that there is such a thing as a standard bibliographic style, not because you’re too dense to notice or too inept to reproduce it, but because you have never learned that such matters are to be attended to.
So I agree that making explicit our tacit knowledge in such matters is a service to the community. Weatherson shows very nicely how to do the kind of philosophy that consists in Problems, Arguments, and Examples (some of which are Puzzles).
- A Problem is a standard question for which Arguments supply putative answers. Answers to Problems are called Positions.
- Arguments are bits of reasoning. Some arguments have had proper names given to them: the Causal Exclusion Argument, the Third Man.
- Examples are situations which one can describe in a few sentences and offer up to intuition as illustrations of or counterexamples to the Arguments. Puzzles are examples that produce conflicting intuitions among philosophers, thereby creating new Problems and Positions for which new Arguments will be invented.
Doing philosophy, or some part of doing philosophy, consists in having ready to hand a stock of Problems, Positions, Arguments, and Examples, and in developing a knack for
- (i) varying the Arguments slightly so as to apply them to new cases, and
- (ii) generating new Examples from everyday experience or (in “philosophies of”) the discourse of some other discipline.
A competent philosopher is one who is well primed to pick out relevant analogies on the basis of which Arguments can be transferred from one domain to another, and to notice Puzzles that hook up with current Problems.
This is, mutatis mutandis, a decent description of “normal science” or its equivalent in any discipline—or, for that matter, almost any productive activity. Weatherson gives some examples.
▷▷
A lot of what many of us (at least many of my peers) do in philosophical research is apply old ideas to new fields. The danger of this is that a lot of work ends up sounding like the caricature one hears of Hollywood movie pitches. […] The upside is that when it works we get really interesting new results.
The camera continues to track until it is once again shooting through the window of Mill’s office, where he is hearing a pitch from two women writers. His phone rings and when he answers it, he demands to know how Adam Simon got on the lot, and then hangs up and continues to listen to the pitch, for a film in which a TV star goes to Africa and becomes worshipped by a tribe of little people. “Kind of like The Gods Must be Crazy except the Coke bottle is now an actress,” he suggests. One of the writers counters with, “Sort of like Out of Africa meets Pretty Woman.”
The classic real pitch was jotted down, or so the story goes, by Brandon Tartikoff, who was then the president of NBC, on the back of a cocktail napkin: “MTV Cops”. The result was Miami Vice.
Normal philosophy
The “normal” in “normal science” or “normal philosophy” is not pejorative. Even the genius, once she has arrived at her breakthrough, typically reverts to Normality thereafter, a new Normality, no doubt, but normal all the same. I’m often struck, when I read the late Scholastics, by the resemblances between Scholasticism and analytic philosophy (once their heroic periods ended—for Scholasticism, this was about a century or so after the first great commentaries by Albert and Thomas).
Scholasticism was the philosophy of the universities of Europe from 1250 until well into the 17th century. Its basis was the teaching of Aristotle’s works, which supplied the bulk of the undergraduate curriculum (↓). 
Out of that teaching, and spurred by the need to put reason at the service of faith, standard quæstiones arose. These were, in concrete form, the Problems of medieval philosophy: the existence and nature of matter and form, for example, or the relations of God to creatures and their operations. Eventually a collection of standard Positions and Arguments furnished much of the content of commentaries and collections of questions on Aristotle.
Like the School philosophy, analytic philosophy is a philosophy of the classroom. Its practitioners’ publications serve to advance their authors’ standing within the profession. With some exceptions, they are little read by the profane, and are instead addressed primarily to an audience of fellow experts.
There are differences, of course: analytic philosophers grant no single person authority comparable to that of Aristotle among the Scholastics; there is less agreement on what methods to use and on what the core Problems are supposed to be; with few exceptions, philosophers no longer have to concern themselves with the opinions of theologians (though many philosophers take the findings of current science to be authoritative, scientists have no direct say in the administration of philosophy faculties). Now, moreover, philosophers place a premium on novelty that is largely absent from the Schools, who, even when they do introduce genuine novelties, continue to profess allegiance to their authorities. (This holds of all the humanities, and owes something to the metamorphosis of universities, hitherto not noted for innovation, into research institutions in the late 19th century.) To some of the work passed off as new, the epigram borrowed by Wittgenstein from Nestroy no doubt applies. But if claims of progress are sometimes exaggerated, still it does seem that a shared imperative to move beyond one’s predecessors, and, more generally, competitiveness among experts (especially the untenured) lead to an intense exploration of whichever Problems and Arguments are taken seriously at a given time (↓)
Being normal
Doing normal philosophy is not a rule-governed activity. “Be clear”isn’t a rule. It’s a desideratum. If, presented with that advice, a student responds with “Yes, but what’s clarity? I thought I was being clear”, I think that he or she has a point. The same goes for relevance, which in my experience is not easy to define except by reference to cases.
Having said that, I offer some pointers.
- Arguments have structure—repeatable structure that you can pick up and transfer to other contexts. The Scholastic disputatio is exemplary: every single question in Thomas’s Summa consists of a question, a section (introduced by the words videtur quod—“it seems that”) stating some arguments for the opinion to be refuted, a brief statement of the true opinion (introduced by sed contra [hanc opinionem dicendum] est—“but against [that opinion] it is [to be said]”), followed by arguments for that opinion (introduced by respondeo—“I answer”), and finally a refutation of the arguments in the first section (ad primum, etc.). If you look at a year or so’s worth of papers in the Pacific Philosophical Quarterly, as I did once, you will see that the Scholastic structure is still frequently used, with one exception: most often the sed contra and the respondeo come first, perhaps after a very brief summary of someone else’s, or the received, opinion. The videtur quod and its refutation come later.
This is an excellent starting-point, comparable to the school-fugue or life study. Master it and you will cease to digress; you will be less likely to beg the question or to fall prey to confirmation bias. Larger arguments can then be composed out of repeated disputationes.Suppose you want to argue that artworks are rather to be thought of as being discovered than as being created. Someone who thinks that artworks are abstract structures, existing therefore timelessly, and so incapable of being created, will likely want to make such an argument. The point of the distinction between discovery and creation is that discovery is of a thing that exists prior to and independently of the act of discovery; but in creation the thing does not exist before the act. So your question turns on whether (and in what sense) an artwork can be said to exist before it is first produced. Proceeding scholastically, we have as our first task to find or invent arguments purporting to show that artworks don’t exist prior to their first production. The next task is to find arguments that (not purportedly, but actually) show that they do; and finally, to find refutations of the first arguments.
- It’s at this point that you are most likely to falter. You will have seen lots of arguments; you can analyze them, and perhaps criticize them; but coming up with a new one—that may seem like a task of a different order.
Coming up with a genuinely new Argument is indeed not easy. But as Weatherson notes, and as one can see in Thomas, that is not required. The arguments in the videtur quod, first of all, can be taken from others. That is one reason to survey the literature. The refutations of those arguments may not occur to you right away, but there you have at least the advantage of working from something that already exists. It is in the respondeo that the problem of invention is most likely to present itself.
- But here too “invention” consists mostly not in creating an argument from scratch, but in varying an existing argument. X has offered an argument for a Position (whether X’s contribution is original need not concern you for the moment). Y has published an objection. All you need to do is to “show that X’s argument can overcome Y’s objection”. The options include:
- (i) Y has misunderstood all or part of X’s argument, and the objection lapses. Descartes says to Gassendi in the Fifth Replies that Gassendi, in objecting to the First Meditation that the machinery of the evil genius is superfluous, has misunderstood the nature and aim of radical doubt. He therefore dismisses Gassendi’s objection.
- (ii) Y’s objection does tell against X’s argument as it stands.
In case (ii), you have the choice of- (a) making a distinction on X’s behalf, so that Y’s objection stands only if the distinction is taken in one way (the harmless way), but not if it is taken the other way (the one that matters). So in defending the claim that artworks are discovered against the objection that artworks, unlike abstract objects, have a history, you might distinguish between tokens and types. Tokens exist in historical time, types do not. The artist, in producing a token, reveals a type.
- (b) showing that Y’s objection relies on an assumption that X didn’t make or doesn’t need. In the 5th Objections to the Meditations, Gassendi says to Descartes that to know what the res cogitans is “you must work on yourself, as if by a kind of chemical operation, in such a way that you could discover for us and make known the interior of your substance”. Only then would Descartes be entitled to say he knows his mind better than his body. But he hasn’t performed such an analysis, and is therefore not entitled. Descartes answers that “I have never thought to render a substance manifest it was necessary to do other than to discover its various attributes” (5Resp, ad 2, no 9). He rejects the presupposition that knowing a thing requires an analysis into constituents; Gassendi was in effect taking that to be an assumption on the part of Descartes. It wasn’t. Another example: Gassendi wonders how the mind can act on the body or vice versa when the two have, according to Descartes, nothing in common; Descartes recognizes in that query an application of the commonly accepted principle that “like causes like”; he rejects the principle.
- (c) if Y’s objection rests on an analogy, you show that the analogy overlooks relevant features and is therefore beside the point. Objection: discovering America was hard work; “discovering” a new musical object is as easy as hitting a few keys on a piano, or even (if music is “organized sound”) just turning on your air-conditioner. Answer: it’s true that to make available to perception (and therefore to “discover”) a musical object does not require anything more than producing sounds having the structure of that object. The analogy takes “discovery” in a more full-blooded sense than was intended, and so it fails as an objection (on the other hand, it does raise a question about the metaphor of discovery as applied to abstracta of the sort that musical objects are supposed to be).
These items by no means exhaust the possibilities. Any good philosopher will have tacit knowledge of them and others, picked up mostly by imitation (like Coleridge, I would distinguish imitation—“creative appropriation of existing works regarded as models”—from copying, which for Coleridge consists in the mere mechanical juxtaposition of elements from existing works, as in “Pretty Woman meets Out of Africa”) (↓). I say “picked up”: no more than most people do I explicitly teach strategies of argument. Perhaps we should.
Teaching strategies of informal argument was once the province of logic (before the rise of formal logic) or of rhetoric. At one point in my graduate career I went through quite a few introductory books on argument (I was working in a philosophy library at the time, so I could just grab things off the shelves). Here’s a list of the ones I found useful, together with a few others encountered before or after:
Kenneth Burke. Grammar of motives and Rhetoric of motives. —Presentations of Burke’s “dramatism”, but interesting also simply for his analysis of the workings of language.
Richard Whately, Bp. Rhetoric (ante 1841, many editions) and Logic (1826, many subsequent editions; originally published in the Encyclopedia metropolitana). —Whately is very good on notions like burden of proof and presumption.
John Wilson. Thinking with concepts. Cambridge, 1963; several subsequent editions to 2004. —Slightly old-fashioned linguistic analysis; but of more than period interest.
John F. Genung. The working principles of rhetoric, examined in their literary relations and illustrated with examples. Ginn 1900. —Like Whately, a good guide to informal principles of argument.
Richard Watson. Writing philosophy: a guide to professional writing and publishing. Carbondale, Illinois: Southern Illinois, 1992. —As its subtitle implies, this is about writing for the journals. In one chapter Watson dissects one of his own published papers.
Robert Graves and Alan Hodge. The Reader over your shoulder. A handbook for writers of English prose. London: Jonathan Cape, 1943; at least one further edition and many reprintings. —Graves and Hodge start with a collection of rules that I think beat Strunk & White’s all to hell. But the payoff comes when they take passages from well-known writers (Russell, Hemingway, Eliot) and give them a rude going-over, as if they were freshman comp students. They even catch the Archbishop of Canterbury plagiarizing Milton. A classic.
William Empson. The structure of complex words. London: Chatto & Windus, 1951, etc. —Read the chapter on ‘quite’ and you won’t be as naïve as philosophers often are in interpreting authors like Hume.
(↑) I don’t like to make up examples if I can find authentic ones. The reason I don’t like made-up examples is that I don’t trust my intuitions much. In particular, when I do history, I don’t trust my intuitions concerning the interpretation of old texts. If you find yourself saying, “Here’s what X could (or might) have meant”, you’re likely to be whistling in the dark unless you have read enough material, not only from X but X’s contemporaries, to be a reliable “projector” of intentions for people of X’s period and type. Even when the example is supposed to be contemporary, I’d rather start (as Weatherson does with his music review) not from something I imagine someone might say but from what someone really did say.
(↑) In the faculty of theology the Sentences of Peter Lombard (ca. 1150) was the teaching text. As with Aristotle, there came to be a series of more or less standard questions passed on from commentator to commentator. Sentences commentaries could, however, range quite widely; their authors would attach questions on topics of interest to them to some passage in Lombard pertaining to the topic.
(↑) On the structure and social significance of the professions:
Krause, Elliott A. Death of the guilds: professions, states, and the advance of capitalism, 1930 to the present. Yale, 1999.· 0300078668
Abbott, Andrew. The system of professions : an essay on the division of expert labor. Chicago, 1988.· 0226000699
(↑) Tom Payne nicely summarizes the clichés of book reviewing (“Circle of clichés”, London Telegraph 8 Aug 2004). One of them is ‘x meets y’:
Publishers have to tell journalists, shopkeepers and readers what a book is like as quickly as possible, so find themselves using an immediately recognisable language. There is no counting the books that have subtitles beginning with the words “the extraordinary true story of”, or the times when the story is untold and the truth shocking. One publisher told me that a book was a “lie-in-the-bath-with-a-glass-of-wine” kind of book; another that a work was “Alan Bennett meets Victoria Wood”. (I wish I’d stopped myself from suggesting that they might have met already.)
Of course anything can be reduced to such terms, where by ‘can’ I mean that indeed something about the work in question can be conveyed. A Coleridgean “copy” is a work for which clichés tell the whole story.
(↑) Added 4 Mar: The actor who provided Bambi’s voice, Donald Dunagan, was seven-year-old child actor (Son of Frankenstein, 1939) when Bambi was made. His family was conned out of the money he had earned: “We were dirt poor before the movies and dirt poor after the movies”. He went to Alabama on a football scholarship, and became a Marine Corps drill instructor during the Korean War. He now lives in Texas. Having lost his life savings when Enron went kaput, he tutors children to make money (Shelley Gabert, “Bambi breaks his silence”, FilmStew 1 Mar 2005; courtesy of Emmanuel.net).
March 2, 2005 in History of Philosophy
· Philosophy of Philosophy
