by Dave Maier
In his recent book Intuition Pumps and Other Tools for Thinking, Daniel Dennett relates that he likes to put to his fellow philosophers the following dilemma: which of the following would you rather accomplish?
(A) You solve the major philosophical problem of your choice so conclusively that there is nothing left to say (thanks to you, part of the field closes down forever, and you get a footnote in history).
(B) You write a book of such tantalizing perplexity and controversy that it stays on the required reading list for centuries to come.
The wording of the alternatives suggests a common conception of the distinction between analytic and continental philosophers. On this view, the former are “problem-solvers”, engaged, much like scientists, in a collective search for truth. Most of the time they proceed by focusing on a particular well-defined problem in isolation, in the hope of chipping off a modestly-sized piece of truth and placing it reverently in its honored place in the Repository of Established Philosophical Truths. The latter, on the other hand, have waved off the search for truth as a hopelessly naive fantasy, and instead offer provocative readings of a series of canonical texts. If these new texts are sufficiently scintillating, they themselves join the canon to be interpreted by others, part of a continuing conversation with no end in sight.
Although, or perhaps because, it is manifestly unfair to both sides, this account of the analytic/continental divide has proved remarkably durable. Dennett himself has no apparent love for continental philosophy. (From the Introduction: “Continental rhetoric, larded with literary ornament and intimations of profundity, does philosophy no favors […] If I had to choose, I'd take the hard-bitten analytic logic-chopper over the deep purple sage every time.”) However, he is not using this dilemma to illuminate the analytic/continental divide (i.e. such that it is the virtuous former who choose (A) and the self-serving latter who choose (B)). Instead, as he tells it, it is scientists who universally choose (A), “shak[ing] their heads in wonder (or disgust?) when they learn that this is a hard choice for many philosophers, some of whom opt, somewhat sheepishly, for (B)”. He compares these philosophers, not without sympathy, to “composers, poets, novelists, and other creators in the arts, [who] want their work to be experienced, over and over, by millions”.
Lest they miss the point, which of course is not simply that philosphers, unlike scientists, (sometimes) prefer fame over discovery, Dennett then turns the tables on scientists, putting to them an analogous dilemma which he thinks they will find more challenging. Here our choices are:
(A) You win the race (and the accompanying Nobel Prize) for pinning down a discovery that becomes the basis for a huge expansion of scientific knowledge, but that in retrospect […] belong[s] to no one in particular [that is, it is recognized that you were simply the first to the wire with a discovery that was clearly inevitable – if you hadn't done it, someone else would have, and soon; which, Dennett plausibly implies, takes some of the glory out of it].
(B) You propose a theory so original, so utterly unimagined before your work, that your surname enters the language, but your theory turns out to be mostly wrong, though it continues to generate years—even centuries—of valuable controversy. [Dennett's examples here are Cartesian dualism, Lamarckian evolution, Skinnerian behaviorism, and Freudian everything.]
The thought here, if we see (B) as appealing enough to compete with (A), is that “we honor scientists who are wrong in useful ways” – and conversely, I presume, that we forget entirely the poor guys who finished second in the race for the (as it turns out inevitable) discovery. That is, until historians of science dig them up, whereupon the winner is revealed as not so much the pioneering genius as, perhaps, the one who ran the lab from hell, riding his grad students mercilessly in order to reach the finish line first – or instead as the merely lucky beneficiary of some random chance.
Let's leave the scientists to worry about that and get back to the philosophers. Naturally I tried this dilemma on myself. I too have a “philosophical problem” I'm working on, and it's fair to ask what sort of success I might hope for (okay, fantasize about). And I grant Dennett's point, made in more explicit detail earlier in the book (General Thinking Tool #1 = Making Mistakes), that theories which turn out to be utterly wrong can still be very valuable; but even so, I ran into trouble right away. Briefly, it seems to me that not all philosophical problems allow (A) vs. (B) as an intelligible choice. As we'll see, (A) in particular seems to assume too much about what philosophical problems look like.
(A) does at first seem to be the way to go. Like that of the scientists in Dennett's story, my immediate reaction to (B) is one of dismay. The heck with “tantalizing perplexity” – if my readers don't get what I'm saying, I haven't succeeded. Of course (B) isn't all bad; if they're “tantalized” then at least they see something in there that they like. And maybe the “controversy” is due simply to the fact that while some readers do indeed get it, and are converted to the Right Way, there are plenty of unbelievers and heretics around as well; but that doesn't seem to be Dennett's intention in offering (B), which dangles centuries of fame before us as opposed to humble discovery. In any case, it sounds like a door prize compared to (A).
At least until we take another look at (A), anyway. That “footnote to history” bit, for example, sounds like the discovery in question is a typical “problem-solver” conundrum, of the sort Dennett warns us against in another selection in the book, called “Higher Order Truths of Chmess”. You may indeed earn a footnote to history if you discover that “Smith's (2002) claim that Jones's (1989) proof is flawed presupposes the truth of Brown's lemma (1975), which [… and so on]” – and this achievement may indeed, as Dennett notes, “demonstrate considerable brilliance”. But this makes it sound like A. Elk's famous theory (which is hers, and belongs to her) about brontosauruses: undeniable (although I think that that designation is no longer used for that particular dinosaur) but profoundly uninteresting, precisely because undeniable, not a “major philosophical theory.”
So let's stipulate that it is indeed major; but this doesn't help as much as you'd think. If anyone proved beyond doubt, such that (supposedly) no further discussion was necessary about its truth, that (say) the value of artworks depends on their moral value rather than their cognitive content, we would not simply note that fact and move on; we would (as in (B)) put the work on the eternal syllabus – not simply to admire its genius, but to get clear on what it says: what is artistic value? moral value? the cognitive content of artworks? The argument isn't there simply to establish the conclusion as true beyond doubt; it's there to explain it.
In other words, the assumption (A) makes here is that we can detach the fact discovered from the process by which we discovered it. And in science, it seems that this is (or at least can) be true. We don't really need to know exactly how the experiments went, or the actual deriviation of the mathematical elements in a theory. We just add the result to our picture of the world. We know that the phlogiston theory of chemistry has been definitively refuted; but who except historians know the blow-by-blow of how this happened? To do chemistry, you don't need to know who Priestley or Davy or even Michael Faraday is; all you need is the periodic table.
The problems go deeper. (A) stipulates that what I solve is not simply a “major philosophical problem”, but the problem “of [my] choice”. What goes missing here is the possibility that the preferability, or even intelligibility, of (A) may depend on what that choice turns out to be. Given what it is that I am trying to understand, I can't imagine that even “complete” success will leave us with “nothing more to say” on the subject. The idea that a “discovery” about these matters could have everyone saying “well, that's that sorted then, let's all go do something else now” simply gets wrong what success here could possibly amount to.
We tend to assume that solutions to philosophical problems look like scientific ones, only with airtight a priori arguments instead of fallible empirical ones. Whether our answer be positive or negative (i.e., that there is or can be no such thing as, say free will), successful problem-solving results in a discovery that something is the case, and characteristically philosophical knowledge of why this is.
I'm not so sure about this assumption. But what else could a “problem” be? Well, Houston, we have a “problem” when something is wrong; and we don't have to think of it as “we lack knowledge of _______”. It could instead be, “we get confused when we try to theorize (or even talk) about _________.” That is, we talk past each other, we go around in circles, we go down blind alleys, and so on. Wittgenstein, for example, repeatedly states that he is not trying to tell us anything, but instead to get us to do something – or better, not to do something (or, more temperately, to see, before doing it, whether we really need to do it or are simply assuming that we have to do it, due to our previous theoretical commitments). He also tells us straight out that “a philosophical problem has the form: 'I don't know my way around'”. Okay, one more: he also suggests that for some problems at least, their solution is not a matter of moving from premises to conclusion (or, as in science, breaking it down into small, individually solvable bits), but instead that “light dawns slowly over the whole.”
Wittgenstein may seem to be a dodgy example here, as Philosophical Investigations is as pure an example of (B) – tantalizing perplexity and controversy, in spades – as one could hope for. But Dennett too, his naturalistic bent notwithstanding, leans in this direction. After all, his book is not a treatise, but a toolbox; and tools help us with problems just as much as knowledge does. Now as any pragmatist can tell you, whenever you invoke the tool image to suggest an alternative to a priori theoretical inquiry, you risk accusations that you don't care about truth but are merely an “instrumentalist” and/or “antirealist”. (And indeed, even Dennett isn't realist enough for some people.) But the tool image is itself quite useful, if (like all tools) we use it properly.
It might seem that to avoid instrumentalism we will need to vindicate our tools by their success in reaching the truth, so reinstating truth as our proper goal, and tools as secondary. But one virtue of shifting our emphasis from doctrines to tools is the context-dependent nature of tool use. We can't establish for all time (as in (A)) that this or that is the Correct Tool. We choose and use tools as new, unexpected situations come up, and their design reflects this. Still, the danger of instrumentalism is real. Dennett is able to sidestep this issue to some extent, because he can always put on his scientist hat and say, see, there's where “truth” is to be found. Pragmatists (and Wittgensteinians) have more work to do. The latter, for example, tend to reject wholesale the idea of philosophical doctrine, under the banner of “quietism” or “therapy”. I see the point of this, but I also think that there's nothing wrong with discoveries or doctrines if, like “tools,” we conceive of them in the right way. After all, if I have convinced you that one tool is better than another for this or for that, haven't you learned something true?
(A)'s image of “nothing more to say” implies the sort (that is, the degree and type) of success typical of mathematical proofs and other knockdown arguments. There is indeed nothing more to say (although I suppose there are probably cranks out there who deny this) about the irrationality of the square root of 2: the proof is so simple I could explain it to a grade-schooler. Obviously to the extent that we have “solved” a problem of any kind, there is “nothing more to say” about something, i.e. whatever ground we have just gone over, such that we now see the problem as indeed having been solved. But what (A) suggests is that not only is the process of discovery detachable from the result, but also that the problem solved is conceptually detachable from the all of the other ones: that it is conceptually possible, for example, for philosophers of mind to solve completely the “mind-body problem” (such that, again, there is “nothing more to say” about it) without also solving, or at least severely constraining, the supposedly distinct problems of epistemological skepticism or semantic meaning.
This conceptual atomism looks as fishy to me as does the idea of “nothing more to say.” Which raises an interesting counter-dilemma: if I am successful beyond all objection in solving the problem of how to conceive of philosophical progress without falling into either dogmatism or instrumentalism, realism or anti-realism, then all of my objections to (A) have been made not simply clear but also compelling. So does that satisfy the conditions for (A) or not? If it does, then its content undermines the assumptions behind (A); but if it doesn't, then how is it “successful beyond all objection”?