I ended last month’s column with a list of some ideas about memorization, sight-reading, and looking or not looking at the keyboards. This month and next I will focus on the pros and cons of memorization as a learning tool. That is, I want to consider ways in which working on memorization—or not working on memorization—can help the teaching and learning process, and what can be learned from thinking about the phenomenon of memorization, whether a student memorizes music for performance or not. I will also consider the role of sight-reading, or reading in general, in performance, and how reading relates to knowing a piece thoroughly and well. I want to start with a brief account of my own history with memorization. This, of course, affects my thinking about memorization in general, as does the whole range of experiences of my students—and other students whom I have observed—over the years.
I have, to put the punch line first, done very little public performance from memory over the years. I have actually never played a piece in concert or in a recording session from memory. When I was applying to graduate school at Westminster—it was 1983—I had to play my audition partly from memory. This was a requirement for the organ performance program, though not for organists applying to the church music program. I was unaccustomed to memorizing, and I worked very hard at it. In the end, at the audition, I had a brief memory slip or two, from which I recovered fairly well. During my years in that graduate program, I also had to play a jury or two from memory. The experience was similar: that is, I worked very hard on the memorization, had a few brief memory slips, and more or less got through it.
Meanwhile, the rules of the organ performance program at Westminster, at the time I was a student, stated that I would have to play my master’s recital entirely from memory. Entering that program as someone who had done little or no work on memorization prior to my audition, I had no idea how I would manage to cope with that requirement. Either I would work very hard at it and hope that it went well—better, I would have hoped, than the audition or the juries, since noticeable memory slips in concert would have felt quite bad—or I would hope for some sort of miracle. That miracle came when the department decided to change the requirement. We were now allowed to choose either to give one recital from memory, or two playing from the scores. I chose the latter, which, among other things, permitted me to take on the challenge of learning The Art of the Fugue and playing it as one of those recitals. I could not even have considered trying, at that point in my life, to memorize something that long and complex.
Since the last of those juries that I played as a graduate student, I have not played a piece from memory with anyone listening. Clearly this means that I do not believe that memorization is a necessity for good performance: if I did believe that, then either I would have memorized repertoire for all these years or I would have been taking, and would still be taking, an ongoing blow to my self-esteem.
Furthermore, it would be hypocritical of me to believe that we teachers ought to expect—let alone force—our students to memorize. Indeed, after many years of teaching and playing, I cannot see any good reason to expect students or any players to perform repertoire from memory. This is, of course, a fairly extreme statement about a more or less “hot button” topic, and I hold onto it lightly: that is, while I feel quite convinced about this view, I am also open to being persuaded otherwise at some point. I have not been persuaded yet, though, in spite of both generally paying attention to writing and teaching on the subject and having conducted a review of the literature in preparation for writing this column.
The case against memorization
It makes sense to me that, in spite of the very strong tradition of memorization in piano playing and the weaker but persistent tradition of memorization in organ playing, the burden of proof must fall on the side of maintaining that performing from memory is necessary. This is in part because it is usually extremely time consuming. If I am going to ask my students (or myself) to spend a lot of time on anything—time which could be spent, among many other things, on learning and performing more pieces—then there must be a very good reason for it.
However, I have seen the imposition of a need to memorize do actual harm. Literally all of the auditions, juries, and student recitals that I have ever heard that were performed from memory have included memory slips—sometimes small, sometimes large—or passages that were clearly executed in a tight, hesitant way because of fears about memory. This is perhaps a small sample size, but it has been so consistent that it strongly reinforces my belief that if students are required to play from memory, the benefits of doing so must be unambiguous and compelling. I have also seen students do what I would have had to do with The Art of the Fugue if I had been required to play my degree recital from memory: that is, avoid certain pieces that they would really like to play because those pieces seem daunting to memorize. Many students go around in a constant state of tension and anxiety because of concern about memorization. And, worst of all, some people decide that they cannot aspire to be performers at the highest level because they do not—rightly or wrongly—believe that they could confidently perform from memory.
Is there a case for memorization?
Of course, playing music and being a performer is difficult and can be nerve-racking. But is the extra difficulty and tension caused by memorization justified? How good are the reasons for asking students to play from memory?
Some of these reasons are, it seems to me, either essentially stylistic or just practical and arguably rather superficial: that it looks more professional, that it saves the inconvenience of having to use a page turner, that if you use music you will feel like or look like a “student”, that memorization will save you if the music blows off the music desk, that it will enable you to give a recital at a moment’s notice when you are away from your library of printed music, that it will permit you to play at a social occasion at which you were not planning to play. (These specific reasons actually constitute the majority of what I have seen mentioned about the subject in my recent review of Internet-based discussions.) Some people mention that if a piece is fully memorized, it becomes easier to look steadily at the hands and feet and to look to find pistons, stop levers, etc. This is interesting and has more musical/technical substance to it than some, and I will discuss it more later.
However, the main claim for memorization is that only by memorizing a piece can you learn it really thoroughly. This claim takes several forms. The most direct is that it is only through the techniques of memorization that a piece can really be learned—that is, that experience shows that only after doing the kinds of things that lead to a piece’s being memorized will you really know the piece inside and out. Another claim, turning things the other way around, is that if a player engages in the act of learning a piece really thoroughly then he or she will indeed, almost automatically, have memorized it: therefore playing from the score is seen as a sign that the player can’t have learned the music very well. Both of these ideas have been incorporated into the ways that some people talk about learning and playing music. I have seen phrases like “learn the piece inside and out, backward and forward” used as a synonym for “memorize the piece.” I have encountered as a sort of aphorism: “get the music into your head and your head out of the music.” Indeed, in some circles, and in particular at certain times in music history, “learn a piece” has been used as a synonym for “memorize a piece.”
Furthermore, of course, we normally use the language in which we talk about performances or performers to imply, without necessarily having made a considered judgment about it, that playing from memory is playing of a higher order. “She was the first to play the works of so-and-so from memory,” “he had memorized such-and-such repertoire by the time he was 14.” Feats like this are impressive because they are difficult, and there is no reason not to acknowledge the work of people who accomplish them. (By the way, however, they also get more notice than they might otherwise, simply because they can be described objectively. If we try to say that “she was the first to play the works of so-and-so in an absolutely riveting manner” there is no way to establish objectively that this is actually true.) We are still just slipping around the question of whether playing from memory is in any way better—or, for our purposes here, whether asking students to play from memory really helps them to become better players.
Some observers report seeing performers—both students and others—playing pieces with their eyes intently, almost frantically, following the music, clearly needing that music to teach them the notes as they play. In fact, most of us know that this is common, that it always creates bad and insecure performances, and that it is a sign of poor preparation. However, in itself this doesn’t prove or even really suggest that performing from memory is the solution to this problem, although it points to the fact that some people misuse the circumstance of playing from music.
The bottom line of learning
So this all comes back to the same thing: that anyone who wants to play a piece should take on the responsibility of learning that piece extremely thoroughly, and that anyone who wishes to become an accomplished player must get into the habit of studying all pieces thoroughly and well. Much of what I have written about over the last several years—in particular the methods of analyzing and learning counterpoint and the technique of paying attention to elements, small or large, that recur in any piece—has been geared towards helping people to know their pieces very well musically by the time that they have learned the notes. Much of the rest of what I have written—about pedal learning, slow practicing, paying attention to hand choices and more—has been geared towards making sure that the physical side of playing will be secure enough that a player can take advantage of what he or she has learned by getting to know the piece really well, that is, not be distracted from it by physical problems or insecurity.
It seems to me that anyone with good practice habits and good physical technique who has put in the time to study a piece thoroughly will end up being able to play that piece from the score as well as anyone could play it from memory. Therefore my own approach—the bargain I would make with my students, so to speak, is this: that there should be no compromise on studying the music in depth, including taking things apart contrapuntally and motivically, noticing harmonic patterns, recurring rhythms, changes in texture, in what order voices enter, playing hands separately when that seems like a good idea for technical or musical reasons, and so on; but that this intense study should be for its own sake and for the sake of the performance, not for the sake of leading to full memorization.
Those who advocate memorization are right that the greatest source of wrong notes, insecurity, and hesitant, unconvincing playing is not knowing what is coming up next. Too strong a reliance on reading—only half-learning a piece and expecting to fill in the rest by quasi-sight reading in performance—is a trap into which many of us fall, experienced players as well as students. It does not often result in good performances. I would suggest avoiding that trap in the most direct way—by insisting to one’s self and to one’s students that pieces be studied thoroughly and carefully. It is, looking at it one way, overkill and perhaps a distraction to relate that process of thorough study to the act of playing from memory. The opposite of reading a piece that is ill prepared is, I would say, reading a piece that is extremely well prepared.
For some people, the act of studying a piece well will indeed lead naturally and apparently automatically to the musical text of the piece actually being memorized and the printed music’s becoming unnecessary. There is, most obviously, nothing wrong with this. However, there is also nothing wrong with the more common scenario in which even very thorough study of the music does not lead to real, note-perfect memorization. I would encourage teachers and students to be comfortable with that.
Next month I will continue this discussion, talking about some of what I consider to be beneficial ideas that have arisen from the tradition of memorization, such as studying music away from the keyboard, and also discussing the role of sight reading, some of the pitfalls that reading presents, and ways to avoid them.
On a completely different matter: I have recently had a fascinating conversation with several friends on the following question: who was the musician that you have heard live in performance who was born the earliest? This led to quite an interesting and far-ranging discussion about time and history, and the reach of living memory. I would like to open that discussion up to a wider group. I encourage anyone reading this to think about your own answer to that question, and to e-mail it to me at Gavin Black is Director of the Princeton Early Keyboard Center in Princeton, New Jersey, and a recitalist on organ, harpsichord, and clavichord. He can be reached by e-mail at