Total Drek

Or, the thoughts of several frustrated intellectuals on Sociology, Gaming, Science, Politics, Science Fiction, Religion, and whatever the hell else strikes their fancy. There is absolutely no reason why you should read this blog. None. Seriously. Go hit your back button. It's up in the upper left-hand corner of your browser... it says "Back." Don't say we didn't warn you.

Tuesday, March 08, 2005

Damned philosopers and their "intelligent arguments!" Phui!

I have, for quite a long time now, been something of a philosophy-basher. It isn't that I don't think that analogies about people chained in a cave with a fire and shadows are important (You can think of Plato's Cave as a sort of primitive rave, minus the extasy) or that I'm unconcerned with ultimate truth, and real beauty, it's just that... um... well, shit, actually it is because I don't think that's important, and I am unconcerned with ultimate truth. Sorry about that.

In any case, once I grew out of childhood and began to actually explore philosophy a bit, I generally found it to be a quagmire of argument from which no conclusions were ultimately possible. I know some of us here in the blogosphere like philosophers, and I'll admit many of them have compelling reasons, but I've never been among them. For me, science with its firm foundation of empiricism has been a far more appealing option.

Yes, yes, I know science rests on certain philosophical underpinnings. Yes, I know that a proper appreciation of science requires knowledge of these underpinnings. You know what? That's equivalent to saying that my enjoyment of fencing requires an appreciation of oxygen, since oxygen is necessary to the aerobic metabolism humans use while fencing. In other words, it's a more or less meaningless assertion. Can I continue my post now?

It is, therefore, with a heavy heart that I must report having been swayed to the position that philosophers aren't totally useless after all. I know it's big of me to make such an admission, but I'm an expansive guy. Or, at the very least, my ass has become quite expansive since I began spending the majority of each day in front of a computer screen. The philosopher who convinced me that maybe philosophy has a point is none other than the creatively-named Imre Lakatos and I was convinced by his work Falsification and the Methodology of Scientific Research Programmes.

Sounds like a scorcher, eh? Wait til you get to the sex-scene, it's killer.

In this work, Lakatos attempts to answer the question, "What, if anything, distinguishes science from other domains of knowledge?" Such a question is important, given that science is increasingly under attack both inside, and outside, of academia. If we as scientists are to survive and, more importantly, if the Human race is to continue reaping the benefits of scientific knowledge, we must be able to clearly define why science is important. To date, I have primarily relied on Popper's falsification criteria. Specifically, this means that any hypothesis or theory that cannot be subjected to a test that demonstrates its falsity cannot be considered scientific. This derives from the understanding that no system of thought can demonstrate that any principle will always hold, since any given relationship cannot logically be tested every moment from the beginning of the universe to the end. Therefore, given that we cannot prove that something is so, we must content ourselves with demonstrating what is not so- or falsification.

Imre Lakatos, however, makes a useful observation. He points out that many theoretical programs accumulate a substantial amount of falsification over the course of their lives. Initially, many programs receive more falsification than they do support. Yet, honest, decent, hard-working scientists continue to support these programs, in spite of the fact that they have been falsified and therefore should be abandoned.

Following from this point, Lakatos asks what is missing from the Popperian view of science that explains this behavior. Simultaneously, Lakatos wishes to avoid the sterile conclusion of Thomas Kuhn that the abandonment of theories is a result of academic faddishness, rather than progressive objective development. Obviously, this is no small task, but it is one that Lakatos actually accomplishes. First, he asserts that falsification by itself is insufficient to lead to the rejection of a theory. This is, in part, because any test of a theory relies on the assumption that the means used to test it are themselves understood and behaving properly. As we all know, however, all instrumentation malfunctions, all statistical procedures are occasionally hamstrung by violated assumptions, and even direct human perception is sometimes flawed. As a consequence, no single experiment can ever completely falsify a theory, it can only produce results that are consistent with, or not consistent with, a given theory. The source of said inconsistency remains to be determined.

Because of this perpetual uncertainty about the source of inconsistency, falsifying evidence will not succeed in eliminating a theory until a crucial event happens: until a new theory succeeds in surpassing the older theory. In short, falsifying evidence will not be recognized as such, instead of as experimental aberration, until an alternative theory arrives. So far, this seems similar to Kuhn's paradigm shift, but Lakatos adds a distinction. The new theory must accomplish three things in order to surpass the earlier contender.

First, it must account for the predictions and assertions already incorporated into the old theory. In short, it must explain as much of the world. This prevents science from growing regressive. Second, the new theory must make novel predictions about the world. So, it isn't enough for a theory to account for previous phenomena, it must do so while providing new insights into the universe. This allows science to be progressive. Finally, it must provide the building blocks for the construction of new concepts. Therefore, any given theory must prove fertile not only in terms of new predictions, but new lines of theory that themselves make new predictions. Thus, until all these criteria are met, no amount of falsification will condemn a previous theory. Similarly, once these criteria are met, no amount of hedging can save a now-falsified theory. Falsification, in combination with alternative research programs, allows for the forward movement of science.

This argument is rather satisfying for me. It recognizes the empirical reality that scientists do not abandon theories at the first, second, or even third signs of falsification, but yet allows room for intelligent, objective selection of one theory over another. While I certainly think it worth pointing out that a certain amount of faddishness does exist in science, and even that some theories are only abandoned when their last proponents die off (What I like to think of as, "demographic falsification"), Lakatos' conception neatly explains the operation of science and its success in describing the universe in ever-increasing detail. I think it fairly apparent that, in this case, a philosopher did have something meaningful and useful to say about the world in general, and science in particular.

Hey, I'm at least as suprised as you!

The funny part about all this is that what convinced me wasn't the major content of his argument, although that was fairly compelling. Instead, it was his first footnote on page 148 which reads, in part, "This little story bears out, I think, my pet thesis that most scientists tend to understand little more about science than fish about hydrodynamics."

Indeed. I have long argued against detractors that while we all negotiate social life on a daily basis, we do so without necessarily understanding it on a deeper level. Thus, sociologists are needed to discover the underlying principles of social life that we grapple with, and manipulate, without truly understanding. Lakatos makes the parallel assertion that scientists may grapple with the actual process of doing science every day, but may do so without truly grasping the logic and order of what they are doing.

I may not like the conclusion I am led to, but I am honest enough to recognize it. If I am to claim that sociologists can offer knowledge of social life to a world that is saturated by it, then I must also accept that the philosophy of science has something to offer to actual scientists. So, reluctant though I may be to do so, I suppose I must finally be willing to listen to what the philosophers of science have to say.

Fortunately for me, "listening" and "not thinking they're jackasses" are entirely different things.

9 Comments:

Blogger Brayden said...

I think there are other ways that philosophy can be directly useful to those of us in the social sciences. One way is using formal logic to tighten our assumptions. The problem with many social scientific theories, however, is that they are not easily susceptible to the techniques of formal logic. That is, we haven't defined the premises of our own theories rigorously enough to assess them formally. And when it is done, we typically find that our theories have some loose ends to be tied up.

Tuesday, March 08, 2005 9:53:00 AM  
Blogger Tom Bozzo said...

In economics, obviously, we have a lot of mathematical formalisms, and certain philosophers tend to beat us up over them, one way or another. I'm thinking specifically about Brian Leiter, who periodically posts or re-posts anti-economics rants, particularly in the run-up to Nobel Prize season.

Those arguments often seem to have the flavor of:
1. Some philosopher says economic theory has no empirical content to speak of.
2. Leiter agrees.
3. And Richard Posner's defense of economics is lame. That, incidentally, lodges the lameness with economics, rather than Richard Posner's understanding of economics.
4. Therefore, economic theory has no empirical content.

So while I substantially agree with Drek here, I'd emphasize the "totally" in "aren't totally useless." Though I also suppose my puny economist brain can't detect the true rigor of the argument.

Tuesday, March 08, 2005 11:09:00 AM  
Anonymous lago said...

You might pick up Feyerabend as a counterpoint to Lakatos. Among many other good points that Feyerabend makes, he contends that Lakatos' criteria are sort of pointless because the assessment of a program(me) cannot be specified. What if, in the next minute, a degenerate program turned around into a progressive one? What if you dismissed it in that minute before it turned around? Whoops? This is a problem beyond a single case precisely because Lakatos thinks (thought) that you can assess methodologies by looking at history. Feyerabend basically says that the Lakatos explanation for theory change really isn't all that helpful, when you think about how it would have to work. This is one of the grounds on which Feyerabend challenges the importance (or existence) of method in the sciences.

Or you could read Burawoy on Marxism as a science. You might like that better.

Tuesday, March 08, 2005 8:16:00 PM  
Anonymous Anonymous said...

One of the interesting things about general philosophy of science (i.e., the kind that asks "What is it about Science that makes it different?") is that almost no-one does it anymore. The "demarcation problem" (as it's sometimes known) is no longer really a motivating question amongst philosophers of science. In fact, general philosophy of science as such is very uncommon these days at top departments. Instead you get philosophers of physics, and philosophers of biology.

If you're interested in the last rounds of demarcation, try Larry Laudan's work, and also Bas van Fraassen's empiricism. (NB: Empiricism here doesn't mean what you think it means.)

General debates about realism and anti-realism originated in metaphysics before migrating into philosophy of science. But now phil of sci has specialized, and debates about realism continue (in a highly evolved form) in metaphysics. The debate between Hilary Putnam and David Lewis is an example.

Incidentally, if you were convinced by Popper once, you need to learn about Nelson Goodman's new riddle of induction, outlined in Fact, Fiction and Forecast. Directly relevant to Durkheimian theories of culture.

-- Kieran

Tuesday, March 08, 2005 8:38:00 PM  
Blogger Erin said...

This comment has been removed by a blog administrator.

Wednesday, March 09, 2005 8:23:00 AM  
Blogger Erin said...

Sorry, didn't mean to delete my last comment. I was just mentioning Andrew Pickering's work, physicist turned sociologist, The Mangle of Practice: Time, Agency, and Science, simply because I'd be a poor UIUC student if I didn't. Science studies don't come up that often.

Wednesday, March 09, 2005 8:26:00 AM  
Blogger Drek said...

Brayden: You're right about the uses of logic, but I'm not convinced that philosophers are all that much better at it. Some of them clearly are, but many of them seem rather weak on the subject.


Tom: It's easy to win an argument when you're talking to yourself (we bloggers should note that principle, as a side issue). Sounds like what we need is some philosophically-gifted economist to give Leiter the shaft.



Lago: I can only assume that there's more to what Feyerabend says than you report as the issues you raise were addressed by Lakatos in the piece I refer to. It may well be the case that Feyerabend is unsatisfied with Lakatos' conclusions, but I'm not entirely convinced that there is a satisfying solution to a problem phrased as, "What if you mis-predict the future?"



Kieran: What I love about you is that whenever you post a comment, I walk away with an entire fucking bibliography. As if I didn't have enough to read before, not to mention that for me reading philosophy is like eating brussel sprouts. Blech!



Erin: I'm familiar with Pickering and have read some of his work. As soon as he blunders into a logical argument, I'll be happy to read more.

Wednesday, March 09, 2005 9:45:00 AM  
Blogger Tom Bozzo said...

I'll throw in a plug (assuming, of course, that you have time for some more cross-disciplinary reading) for Phil Mirowski's More Heat than Light, which can be read as greatly amplifying Brayden's comment as applied to mathematical microeconomics. It's also a bit lighter reading than Mirowski's more recent Machine Dreams, reviewed by Kieran here.

Wednesday, March 09, 2005 10:06:00 AM  
Blogger lauren said...

You, dear realest, are the shit!

Sunday, August 13, 2006 12:41:00 PM  

Post a Comment

<< Home

Site Meter