Web Reviews
Bob Snyder: Music and Memory: An Introduction
The MIT Press, 2000, softcover, ISBN 0-262-69237-6, 291 pages, illustrated, glossary, bibliography/references, index, appendix: listening examples; The MIT Press, 5 Cambridge Center, Cambridge, Massachusetts 02142-1493, USA; telephone (+1) 800-356-0343; electronic mail mitpress-order@mit.edu; World Wide Web mitpress.mit.edu/

Reviewed by Chris Kennett
Harrow, UK

The historical tendency in musicology to eschew the cognitive implications of its varied analytic methodologies has been noted, criticized and, to a certain extent, rectified in the past two decades or so. Several writers from the semiotic tradition have made the act of listening (and, to a lesser extent, the psychoacoustics of the human brain) form a key stage in the process of creating musical meaning through analysis. Indeed, this trend is perhaps most notable in the range of methodologies constructed in the last 15 years with which to analyze pop music. However, whether by professing the primacy of a recording as analytical text and then using not-quite-Schenkerian graphs to analyze a possible way of hearing the piece, or by castigating another writer’s formalist analytical ethos and then replacing formalism with hermeneutical guesswork, an opportunity to link the worlds of cognitive science and musicology has, generally thus far, been lost. This missed opportunity is unsurprising, of course—how many musicians could honestly claim to know as much about the neurology and chemistry of the brain as they do about musical structure? (and, doubtless, vice versa)—and is largely due to a Western academic tradition which trumpets its own interdisciplinarity, and yet ghettoizes music as an art, cognition as a science, and so on, minimizing the chances of the development of genuinely interdisciplinary research.

With this background, Bob Snyder’s breezy, extremely reader-friendly exposition of current research in the psychology of music cognition is as timely as it is significant. Written from the perspective of a tenured assistant professor in composition, frustrated in his unsuccessful attempts to find a pedagogical text offering “at least tentative answers to many of the ‘why’ questions [his non-musically trained] students would ask as they grappled with the mechanics of music”, and then writing such a text himself, Mr. Snyder’s volume neatly sidesteps most, though not all, of the pitfalls of many other musicianly efforts on the subject of the links between cognitive psychology and musical meaning, for two principal reasons. Each of these will be dealt with in more detail below.

As a first example of his fancy footwork, Mr. Snyder clearly understands the science of current research into the human brain’s strategies for cognizing music, and for imbuing that music with meaning-structures. The first half of the book consists of an admirably clear, concise, and current exposition of the ways in which the brain first cognizes sound in “echoic memory” (especially pages 19-30), combines sounds into groups in “short-term memory” (31-58), then creates meaning through a range of synaptic associations and comparisons with other groupings and “schemas”—cognitive-semantic templates constructed from prior listening experience—in “long-term memory” (69-120).

The essentially circular, iterative nature of such processes is never far from the surface, in that almost each stage of cognition is at least partially dependent upon mediation from other stages. After the initial, near-universal process (amongst humans) of cognizing discrete acoustical vibrations repeating at greater than about 20Hz as combining together to form single events—what Mr. Snyder calls the “level of event fusion” (12-13))—there follow the processes of “feature extraction” and “perceptual binding” (19-23), in which common features of perceived events—such as timbral similarity, common register, and so on—enable them to be grouped together before being input into short-term memory (“STM,” 47-68) and long-term memory (“LTM,” 69-106). These extracting and binding filters are, to some extent, mediated by past experience, and each new sonic experience will encourage the formation of new feature-recognizing filters.

STM, and, within it, “working memory” (48-49), the principal center of conscious awareness, is fed by the results of the feature extraction and perceptual binding processes, and also by the process of “chunking” (53-58) , combining events into groups of between five and nine elements, and placing these events in hierarchies defined by other factors such as perceived closure and repetition (pp. 59-68). Needless to say, this process mediates (and is mediated by) the LTM through comparison with, and modification of, pre-existing “schemas” of meaning gained from a lifetime of listening. Thus, the LTM is also the repository of formal and structural meaning-creation, mediated by (and mediating) the temporary chunking processes taking place in the STM, and so on. Much of the second section of the book is taken up with categorizing these larger melodic, rhythmical, metrical, and formal units. Harmonic hierarchies are omitted deliberately, because “the theory and use of harmony, especially as it relates to modulation between different harmonic centers, is primarily a European phenomenon” (xiv), and thus outside the scope of a book which is focused primarily upon near-universal meaning-creating processes.

The second way in which Mr. Snyder shimmies past most of the pitfalls of existing music-cognitive literature is by stopping deliberately short of the hermeneutic abyss that has claimed the souls of several cognitively-inclined analysts before him. Instead of using the processual theory exposed in the first half of the book to inform some specific music-analytical examples of the brain’s use of these processes to construct musical meaning, the second half merely deals with the brain’s ability to construct a hierarchy of meaning-structures by combining events into melody and rhythm, and thus into increasingly large formal structures, in the LTM.

Mr. Snyder’s rationale for this is clear:

Whether there are any such things as psychological musical universals is debatable, and I would be the last to deny that music exists within a cultural context. Nevertheless, it seems clear that the human nervous system is subject to universal cognitive constraints, some of which impinge on possibilities for musical structure. This book takes a position between the extremes of autonomous perception, unaffected by cultural construction, on the one hand, and purely cultural construction of all levels of musical order, on the other. I believe there are bottom-up (perceptual) effects that, while not completely autonomous from top-down (cognitive) influences like culture, are relatively predictable, and that these processes are necessary for the perception of patterns in the first place, especially unfamiliar patterns. (xvi)

However, the ramifications of this stance, mid-way between the perceptual and the cognitive, are at times problematic, and are most sharply defined in terms of the uncertain status of the appendix of listening examples (245-254) which follows the second section.

While the choice of musical pieces which exemplify Mr. Snyder’s concepts is generally sound enough—a good range of Western and non-Western art musics, including some jazz (why no popular musics, though?) from a variety of historical periods—none is discussed in more than the most fleeting detail. For example, Luciano Berio’s Sequenza VIIb (sic—one assumes that Mr. Snyder means Sequenza IXb, here, since VIIb does not exist) for alto saxophone is cited as a fine example of “the axial principle” used (compositionally? cognitively? both?) as a melodic schema (248) in a piece of contemporary music, but no further discussion of the work is offered and nor is the composition referred to in the chapters discussing schemas. Indeed, both the listening examples in general, and the appendix which houses them in particular, are rarely if ever referred to in the main body of the text, not least because of Mr. Snyder’s wish to avoid the pitfalls of analytical examples. This renders their usefulness to the reader as listening examples questionable, at best: in the case of the Berio example cited above, after a single listening, and without guidance from Mr. Snyder, it is hard for this reviewer to hear the piece as an axial melody “around a central pitch” (154, paraphrasing Leonard Meyer)—maybe after several listenings, or with the aid of the score, but certainly not without.

This point highlights the difficulty likely to be encountered by anybody using the purely cognitive—what might be described as “this part of the human brain demonstrably can be said to group these sounds together in this way”—to inflect the perceptual—“this group of sounds means this.” This is certainly not to suggest that any such attempt is doomed to failure, but merely that there is little guarantee that one person’s version of what comes out at the back end of the process of creation of musical meaning will tally with another’s, notwithstanding the much stronger likelihood of a match in terms of what goes in at the front end, in terms of the cognizance of frequencies, durations, and relative dynamics. It is easier to absolutize physiologically verifiable cognition; much less so the personal, often irrational perception.

To be fair, Mr. Snyder acknowledges this difficulty on more than one occasion. In one extremely lucid chapter (107-120) he explores the extent to which metaphors of gravity, centricity, linearity and goal-direction in musical discourse are just that: commonly-agreed, linguistically-derived metaphors, as opposed to verifiable music-cognitive “facts.” Similarly, he admits that “individual knowledge and preferences are certainly a factor in chunking” in STM (223), and that, even at the level of echoic memory, “the grouping a particular person may hear on a particular occasion is not always completely predictable” (37); but, with such potential for the meaning-creation of individual listeners (or at different listenings) not to overlap at all cognitive levels “higher” than the largely physiological level of event fusion, is it any wonder that different listeners would not overlap in their opinions about the highest-level structures created in LTM?

As a consequence, the listening examples strike a note oddly dissonant with the otherwise absolutist, universalist ethos of the book, and might have been better omitted, as might unproveable and, in comparison with the first section of the book, poorly corroborated comments dealing with cognitively-led definitions of various schematic phenomena in the second section. One example will suffice here—the definition of “rhythmic contour:”

The profile of changing timing patterns in a rhythmic grouping, metaphorically thought of as having a “rising and falling outline” formed by changes in time intervals between events. Faster rhythms with shorter time intervals form the high points of a rhythmic contour. The “height” of a rhythmic contour indicates the amount of purely rhythmic (nonmetrical) tension present at any given moment. (161)

As a personal definition, this is unproblematic; but what if some listener (such as somebody who does not hear Sequenza IXb as especially “axial”, say), shrugs his (in this case) shoulders and says, “says who?”

Again, in fairness, on several occasions Mr. Snyder does imply that there may be differences of meaning-structures across different cultures—for example:

Cultures whose music is very rhythmically oriented, such as the music of West Africa, might have smaller, tighter beat categories [scope for “rushing” or “dragging” the beat, or other types of rubato] than music without such an orientation. This would mean that small rhythmic changes we might perceive as expressive nuances could function structurally for listeners brought up in that culture. (167)

But what about individual differences within a culture? Aren’t we living in a world so culturally complex and diverse that it becomes nigh on impossible to predict how any of us would define a “rhythmic contour,” “beat category,” or whatever? Mr. Snyder admits that “the time interval proportion categories I have given here are based on informal experiments on me and my students” (191); but does that mean that there is any more global, absolute element to such locally-defined categories, in our individuated world where musical meaning-creation becomes a sort of idio-ethnomusicological journey into the self, each time we listen? The fact that the number of references, citations, and quotations from other writers tails off dramatically in the second section, as compared to the first, would seem to underline this problem.

Whatever the reader’s response to the above, there are a couple of minor areas of Music and Memory which certainly could be improved. Mr. Snyder’s admirable desire to use clear, simple, and non-elitist language “to make [a potentially impenetrable subject to the lay reader] more comprehensible” (xvi) is all very well, but it is not clear how certain examples of this approach might avoid the arbitrariness of personal taste. Taking Mr. Snyder’s preference for “schemas” over “schemata” as a plural form of “schema” (rationalized on p. xiv), for instance, the wish to avoid the elitism of unnecessary Latinism is laudable, but the words “scheme” and “schemes” would do the job just as well, and would leave the old language, and its nefariously elitist and obscurantist agenda, intacta.

Similarly, the standard of sub-editing is puzzling: the word “grammar”—surely an important word in a cognitive-linguistic context such as Music and Memory—is spelled as “grammer” on several occasions (including one entry in the bibliography (275) where a Stanford University Press publication (originating from “Standford, CA”, apparently) is titled, Foundations of Musical Grammer). Other unfortunate typos include “constituant” (46) and “Noth” for “North” (205), and not forgetting the Sequenza IXb problem noted above. It’s not too elitist to use a spell-checker before publication.

However, the strong points of this book (and they are strong) remain those which are the most clearly corroborated: the hierarchical structures of initial cognition. Thus, for his lucidity of the first section, both of the text and of the range of extremely clear and comprehensible diagrams to clarify the more rebarbative aspects of the initial cognitive process, Mr. Snyder is to be congratulated. A little further exploration of the “highly idiosyncratic” way in which “patterns of associative reminding” (56) affect the way each of us constructs musical meaning structures would enable the reader to rehabilitate the second section as a useful, personal example of the creation of larger cognitive-semantic structures in music, rather than as an doomed attempt to absolutize the unabsolutizeable.