Monday, 26 September 2016

The Test of Excellence

This week we were required to read an interesting article on math pedagogy entitled “Relational Understanding and Instrumental Understanding” by Richard Skemp. In the article, Skemp makes his case that too much math pedagogy is based on teaching calculation techniques and procedures (what he calls “Instrumental” understanding) as opposed to instilling a deeper understanding of mathematics, from which it should be apparent to the student why those techniques (“Relational” understanding). His thesis, if I may so condense it, is that students are rarely being taught actual mathematics but instead of simply memorizing computational rules and the situations in which they apply. Students then can correctly do specific tasks very well, but if presented with something they haven’t seen before, will be at a loss on how to proceed.  In principle, I very much agree with his concerns and have seen the equivalent problem in the first-year physics students I used to TA, who often viewed physics as little more than applying a succession of mysterious formulas of dubious origin. However, I am less enthusiastic when I imagine what would be required in order to implement Skemp’s ideas effectively, and what the outcome would be if the attempt failed.

There is something dangerously optimistic about viewing prospective public policy through the lenses of the best-case scenario. I am sure Skemp himself would be very effective at teaching his own methods, not least because he is an actual mathematician. However, a teacher cannot teach what he does not know, and I wonder what would happen if a teacher who had little to no background mathematics was made to teach in this manner, as though he were an expert. Speaking for myself, I believe I could teach math and physics in this way (and I intend to), but I am not sure that I could do it with chemistry. This is not because I don’t know anything about chemistry, but because I don’t have nearly the same level of expertise in it that I have in physics. I have observed this pedagogical approach go awry when I tutored high school students in physics. I had one student who was struggling quite a bit with relative motion. The student showed me diagrams the teacher was having them draw in order to describe different kinds of reference frames. It was a very “relational understanding” approach, and perhaps could have been effective. However, from the student’s course notes, copied from the board, I could see that proper use of the diagram had been explained incorrectly and that by making these diagrams, the student had been reinforcing a serious misunderstanding of the material. This would take a lot of work to fix because the error ran so deep in the student’s understanding. Thinking of experiences like this, the question that comes to my mind when I read this kind of pedagogical theory is: used well, this method promises certain obvious benefits, but what is the damage if used badly?

I am naturally resistant to a wide and enforced implementation of ideas, even ideas I agree with. This is because ideas have unintended consequences, and if some of the consequences are bad, a broad implementation will make them widespread instead of local and more easily addressed. Additionally, it is often the weakest and poorest members of society who are least equipped to cope with the consequences of bad policy. I recently read a very sobering article by the British psychiatrist Theodore Dalrymple, who worked for most of his career as a doctor in an English prison and in a neighboring slum hospital. He has produced a huge body of very depressing essays and articles describing the disastrous social consequences of ideas that begin at the top, trickle down, take root, and then run amok. In his article “We Don’t Want No Education”, he described some of his countless observations on the current educated state of his teenaged patients, part of which he blames on experimental educational practices:

… I now test the basic literacy of nearly every such youth I meet, in case illiteracy should prove to be one of the causes of his misery. (I had a patient recently whose brother committed suicide rather than face the humiliation of public exposure in the social security office of his inability to read the forms he was required to fill in.) One can tell merely by the way these youths handle a pen or a book that they are unfamiliar with these instruments … I cannot recall meeting a sixteen-year old from the public housing estates near my hospital who could multiply nine by seven (I do not exaggerate). Even three by seven often defeats them. One boy of seventeen told me, “We didn’t get that far.” This after twelve years of compulsory education (or should I say, attendance at school).

I don’t think situations like this could flow immediately, or even naturally from a single idea or theory, but I do think that a good idea for a particular situation can become a very bad idea when pulled and stretched into a broad catch-all policy. Skemp’s theory, and theories like it I feel are abused when they used to downplay the importance of calculation. The question is often raised, why bother focusing on hand written calculation when we have calculators? Why not just focus on “understanding”. What strikes me most about this question (as well as the related, “why bother with spelling when we have spell-check”), is the way in which these skills are so carelessly disposed of, as though the mental development that comes from mastering these basic skills isn’t reason enough to teach them, as well as the enrichment of our lives that can only come from filling our minds with meaningful things. Why learn facts about history if you can just look it up when you need it? Why memorize any great sayings from poetry or philosophy when we have Wikiquote? This philosophy, though easy and convenient, seems to me to lead to a cold and stifling materialism. It is in this respect that I fear that we as a society are becoming a bit like Lady Bracknell, who disapproved of anything that interferes with natural ignorance. Setting aside the question of the intrinsic value of knowledge, at the heart of the pro-calculator stance there sits an assumption I’d like to address with respect to the purpose of calculation.

I occasionally observe an attitude towards the importance of calculation that I think misses the point, namely, that it is an end in and of itself. In any level of physics or mathematics, do we give students problem sets so that they can practice calculations, or rather so that by doing the calculations, they will further cement the foundations they are being taught? To give a simple example, a common beginner physics problem is to describe the motion of a box sliding down a ramp. Typically some variables will be given and the student has to use Newton’s Laws to calculate the speed of the box, or the coefficient of friction, or the angle the ramp makes with the ground etc. A student in high school or in the first year of university will see many problems like this, all alike but slightly different. At the same time it can be shown algebraically that these problems are mathematically identical and that, having solved it once, you’ve actually solved every permutation of it as well. What then is the point of doing the same thing over and over again? It is to force the student to re-derive the algebraic result, but each time within slightly different context, or from a different direction, hopefully deepening their understanding in the process. The students can be shown (and I do try to show them) that they are in fact doing the same thing over and over again, but it takes a lot of thinking and a lot of practice in order to see it. Once you reach that point, I would say that you truly understand it. In this case, the arithmetic involved is secondary, and it would not defeat the purpose of the exercise to use a calculator. (As an aside, I’ve directly observed many students whose ability to do these problems is impeded by their reliance on calculators, because instead of working through the problem algebraically and then putting in the numbers, they just number crunch the entire time. If they get the wrong answer, it’s impossible to see where they went wrong because their work up to that point is just a meaningless list of numbers.)

This kind of physics problem builds on top of skills that should already have been learned. In particular, arithmetic, solving sets of linear equations, and rearranging equations. To learn each of these skills, you would have to go through the same process described above, but at a lower level. You would learn the theory (hopefully taught meaningfully, as Skemp would advocate), and then practice that skill over and over, from different angles, cementing your learning and fixing your misconceptions. For learning arithmetic, using a calculator would completely defeat the purpose. If the point was to simply do the calculation, it could make sense to use a calculator since it may be faster and less error-prone (though for really simple problems, I’m not so sure). But we do not learn arithmetic to do simple calculations, we do calculations to assess and develop our arithmetic, and to ensure that the foundation is in place in order that we may proceed to more advanced topics.

I’ve seen this kind of thinking before, albeit in a very different context. One of the most important pedagogical works in the history of music is Johann Fux’s Gradus ad Parnassum from 1725. This book was used to instruct many of histories greatest composers including Mozart, Haydn, and Beethoven. J.S Bach, perhaps the greatest musician of all time, strongly approved of it. I would like to write about it in more depth in a later post but for now, I will note that the structure is written as an actual dialogue between a teacher and a student. The teacher explains the principle and the student does a series of short exercises. After each exercise, the student explains what he did and why, and the teacher approves or corrects them while explaining the reason for his mistakes. Each chapter builds on the one that preceded it, until by the end, you are writing in full four-part harmony. It’s a remarkable approach, but what strikes me about it is the balance Fux achieves in combining theory and conceptual understanding with practice. Sometimes I think we feel that if the understanding is strong enough, correct usage will follow, but in my own experience as a student, this was rarely the case. They feed and correct each other. Fux states this explicitly in his introduction:

You will notice dear reader, that I have given very little space in this book to theory and much more to practice, since action being the test of excellence, this was the greater need.


Similarly, to take a more contemporary example, J. J. Sakurai in the foreword to his graduate level textbook on quantum mechanics states, “the reader who has read the book but cannot do the problems has learned nothing”.  I think most people would agree with this sentiment when applied to advanced science, but why is it a point of contention when applied to elementary math? I would maintain that the student who has been taught arithmetic but needs a calculator has learned nothing since, as Fux said, action is the test of excellence.

No comments:

Post a Comment