It is now more than fifty years since the death of Heinrich Schenker. The academic debate over the legitimacy of his work is essentially over, and his theories are recognized by those who know them as the most important body of theoretical musical thought of this century. Schenkerian theory has profoundly changed graduate study in the theory of music on this continent in the last twenty-five years, yet it seems we are still very unsure how to incorporate it into our undergraduate curricula. Both theory textbooks and articles on the pedagogy of theory reflect the breadth of current views. Texts range from the staunchly traditional (Kostka & Payne 1984) through the subliminally linear (Aldwell & Schachter 1978) to the aggressively Schenkerian (Lester 1982), and recent articles by leading practitioners of Schenkerian analysis reach radically different conclusions about the undergraduate curriculum (Beach 1983, Rothgeb 1981). Curiously enough, even works owing a heavy debt to Schenker's theories seem almost embarrassed by the Ursatz. Der Freie Satz itself begins with a three-chapter discussion of the background stratum. Schenker considers the background an indispensable prerequisite to a work of art (Schenker 1978:3-4), and gives philosophic and technical justifications for his point of view. Oswald Jonas's presentation of Schenker's theory (Jonas 1982) reverses the order in which background and are discussed, but since his book antedates Schenker's, the difference is certainly understandable. What is curious is that Jonas' plan began a virtually unbroken pedagogical tradition, followed even by the most recent Schenkerian text (Forte & Gilbert 1982). Now, there are necessarily crucial differences between a theoretical treatise - or even an advanced text - and a text for beginning students. The one is aimed at mature musicians, the other at neophytes; the one presents a monistic theoretical disquisition, the other lays a broad foundation of musical understandings; the one can be highly abstract and technical, the other must necessarily be propadeutic. Nonetheless, the delay of a discussion of background may seriously undermine comprehension of the Schenkerian position. If Schenker was right in believing that his theories presented for the first time a genuine theory of language (Schenker 1978 vol.I:9) - that is, a precise way of defining what is, and what is not, - there is no reason for university instructors to continue our longstanding practice of lecturing to our students for two or three years about music without ever attempting to define tonality accurately, either for the students or for ourselves. Too often, we have simply hoped that students will eventually intuit what is meant by tonal harmony or tonal counterpoint from a mass of empirical data and unnecessarily restrictive rules, even when the rules are largely pedagogical conveniences far removed from actual compositional practice. But another approach is possible, if Schenker is right in proposing that the essence of music is not just in chord-tochord progressions, or in principles of voice-leading (most of which in any case derive from pretonal practice), but rather in a compositional coherence which can be achieved only through the Fundamental Structure in the background, and its transformations in the middleground and foreground (Schenker 1978 vol.1:6). A beautiful example of how background helps clarify the foreground, and vice versa is found in the opening measures of Beethoven's Op. 101 (see Example 1). At first glance, the excerpt seems to exemplify tonality without a Schenkerian Ursatz: there is no tonic chord directly supporting a reasonable candidate for Kopf ton. But if we accept the Schenkerian definition of tonality, and if we understand the degree to which the Ursatz is an abstraction, not always to be revealed by simply stripping away passing and neighbouring tones, then we can easily understand that the tonic chord of m. …