Monday, February 05, 2007

Wikipedia Redux: Don't Know What a Sliderule is For

My post on Wikipedia generated an interesting set of comments and an email from USC law professor Mary Dudziak, pointing me to her own entries (here, here and here ) on Legal History Blog. Dudziak notes that colleges and universities discourage undergrads from relying on Wikipedia and asks why judges can't be held to at least the same standard. That's a fair question, assuming that on any given subject there is a more reliable source than Wikipedia and assuming as well that a generalist judge and her law clerks can identify with confidence which book on, say, the history of firearms regulation in colonial America is by the well-regarded scholar and which book on that subject is by someone with an ideological axe to grind (on whatever side of the issue). To repeat what I should have made clearer in my earlier post, an objection to Wikipedia based on its unreliability is valid insofar as Wikipedia is (relatively) unreliable. But the way to investigate its reliability is by periodic empirical testing, not by a priori assumption. (I'm not attributing a priori reasoning to Dudziak, just considering a possible line of argument).

That said, even if it turned out that Wikipedia were, on average, as reliable as or more reliable than other sources upon which courts or even scholars routinely relied, there might nonetheless be valid reasons to discourage students from citing Wikipedia. The idea goes something like this: Students need to learn standards of evidence, how to work with original sources, etc., and Wikipedia is a short cut that will leave them unprepared or underprepared for making these sorts of judgments. Banning citations of Wikipedia by students, in this view, is a little like the practice of forbidding grade school children from using calculators for simple arithmetic until they have mastered the concepts of addition, subtraction, multiplication, and division, because these are building blocks for other mathematical skills that cannot be performed on a calculator (or cannot be performed on a calculator without some serious thought about what functions one asks the calculator to perform).

That practice strikes me as sound, but it's hard to know whether and to what extent it should apply to Wikipedia. When calculators were first becoming available cheaply, some schools reacted by requiring mastery of not only manual arithmetic but also a sliderule as a prerequisite to using a calculator. Likewise, when I learned legal research as a first-year law student in 1987, we were forbidden from using Lexis and Westlaw--even for Shepardizing--until we had first mastered the laborious process of researching in hard copy compilations. Both of these requirements now seem quaint, not to say foolish. Neither the sliderule nor the book version of Shepard's was a building block of other concepts, and so, in retrospect, there was no good reason to require mastery of either in a world in which calculators and electronic databases were becoming ubiquitous. Whether that will be true of the materials for which Wikipedia substitutes remains to be seen. We can imagine a not-too-distant future in which just about every Wikipedia entry contains links to original source material courtesy of Google Books and other internet databases. In such a future, much of the current technology by which historians and other scholars track down original sources might be as outdated as the sliderule. Or maybe not. We just don't know in advance. My point here and more generally is that it's easy to confuse technologies with which we're familiar with fundamentals about how we understand the world. The question of whether to permit citation of Wikipedia, like the question of whether to permit laptops in classrooms, should be answered with regard to our best estimate of the real-world consequences at any given moment in time.