Four Argentinean researchers recently published what might be described as a replication of Socrates' own form of the Socratic method. Andrea P. Goldin, Laura Pezzatti, Antonio M. Battro, and Mariano Sigman, From Ancient Greece to Modern Education: Universality and Lack of Generalization of the Socratic Dialogue, 5 Mind, Brain, and Education 180 (2011). 2400 years ago, as they observe (I admit I haven't gone back to check Plato's account myself), Socrates taught Meno's slave how to create a square with twice the area of another square, by asking the slave a series of questions designed to help the slave come to recognize the method that needed to be used. The authors report that many of the 58 contemporary Argentineans to whom they presented the same problem, and the same questions, made the same error as Meno's slave did (the mistake was to think that doubling the length of each side of the square would double the total area; in fact, doing that quadruples the square's area). The persistence of this mistake is a particularly clear proof of the continuing frailty of human reasoning, and that's interesting in itself, notably for the doubt it raises about "the efficacy of the modern educational system" (at 183).
But the authors also report that "to our great surprise," when the experimental subjects were asked to double the size of a new square at the end of the Socratic dialogue, "more than half of the adolescents (57.1%) and almost one third of the adults (32.4%) ... failed to respond correctly." (182) The authors says that "[o]ur observation of a lack of generalization in a strict Socratic dialogue extends a broad literature which has questioned the efficacy of unguided education, despite their broad popularity and intuitive appeal." (183)
The idea that the Socratic method, so central to traditional law teaching, might be demonstrably ineffective has a certain ironic appeal. But actually I think the lesson of the piece is rather different.
It certainly seems fair to say that one exposure to the Socratic method often is not enough to enable a student to acquire a skill and be able to apply it to new problems (even very similar ones). But why should it be? Meno's slave, and the students in this experiment, had no opportunity to practice what they had learned. Socrates could have returned to the task of teaching his student the following day, with a new square as the subject, and could have reminded the student then of the insights of their previous discussion. That's the sort of approach that seems likely to help someone digest a lesson more deeply and learn the special skill of "transferring" that lesson to new contexts.
Meanwhile it is perfectly possible that Socrates' approach, with the benefit of enough repetitions, will have another effect -- namely that it will help the student to really understand why (as the lesson shows) one can double a square's volume by creating a square with sides equal in length to the original square's diagonal. That insight might mean that the student is better able to go on to master deeper and more intricate problems, because his or her knowledge is based on more than simply memorization of prescribed rules.
I don't want to overstate the case for the Socratic method. It might very well be that for many students a bit of straightforward instruction, followed by drills, would eventually produce a mastery just as firm and just as comprehending as the approach I've just imagined. All that I think is clear is that the Argentinean experiment by itself doesn't prove the superiority of any such alternative. (Nor do the authors assert that it does.)
It is intriguing, meanwhile, to see that the authors view the Socratic dialogue as "probably one of the most emblematic examples of education in a minimally guided environment, in which learners must discover or construct essential information for themselves." (183) I wouldn't describe Socratic dialogue as "minimally guided," but the authors' implicit comparison to many other forms of education is instructive. Socratic dialogue, at times an elaborate performance starring the professor (as my friend Bob Dinerstein has recently noted), can also be a process in which students take responsibility for their own learning, far more than they do in a class where the prescribed lessons are all spelled out. The Socratic method in law school has even been described as a form of skills teaching, in which students learn the skill of legal reasoning precisely by performing it themselves. Clinical teaching -- sometimes seen as the humane and practical response to Socratic tyranny and abstraction -- itself is often quite Socratic, as professors supervising cases may firmly resist giving students instruction and instead ask question after question.
So does the Socratic method pass its test? I think the answer is that we have more questions to ask.