John Cowan: > And Rosta wrote: > > > Among Lx teachers I know, I'm > > relatively unusual in inclining towards the latter strategy, > > believing that it is better to know that you don't know something > > than to think you know something (but which is in fact wrong). > > Then if you were a physics teacher, you would refuse to teach the > Newtonian approximation to physics, or the ball-and-stick model of > chemistry? These things are *wrong*, but still useful for many > purposes, and knowing them only is not really comparable with > "thinking you know something which is in fact wrong". > > In particular, the phoneme may be dead, but is there any replacement > for it when teaching the rudiments? "Language" and "word" are > similarly woolly constructions, but can we avoid them? I would distinguish between two kinds of simplification. The first, which is the one I had in mind when I said I avoid it, is motivated by purely pedagogical reasons, or at least not rationalizable in terms of the subject matter itself, and has to be unlearnt and discarded when it comes to acquiring a more sophisticated understanding of the subject. The second is rationalizable in terms of the subject matter itself, and, like Newtonian physics, can be seen as an optimal model of a simplified body of data; the acquisition of a more sophisticated understanding can proceed gradually by taking into considerationa additional data and providing reasons for the gradual revision of the initial simplified model. To take your example, I view the phoneme as an example of the former approach, and I teach phonology in terms of the phoneme; the complexities of a more phonologically adequate substitute model are not necessary for the students' (very largely descriptive) purposes, and anyway it would be relatively straightforward to proceed to present the inadequacies of the phoneme and the merits of a revised model. But I view introductions to phonetics and phonology that make clear and principled distinction between them as an example of the latter approach. The apparent simplifications the blurring creates must inevitably lead to hopeless confusion until the pedagocial ill effects have been expunged. Furthermore, if you simply say "Last Semester you were taught that p is the case. Now forget that p is the case. Now learn that q, not p, is the case", the students get pissed off, for various more or less good reasons. I would add though, that, perhaps unlike Science students, the vast majority of the students I encounter are comfortable with the notion that some answers/analyses are true and some false, and with the notion that some answers/analyses are neither true nor false and only subjectively better or worse than one another, but they are deeply uncomfortable with the notion that some answers/analyses are false or of unknown truthvalue but nonetheless are objectively better and worse than one another. This makes it quite difficult to be upfront about the epistemological status of what one is teaching them. --And.