Charles Rosen has cited a well-known but probably apocryphal story about Schoenberg's reaction to the “Eroica” analysis of Heinrich Schenker: “Schoenberg once looked at Schenker's graph of the Eroica, and said, ‘But where are my favorite passages? Ah, there they are, in those tiny notes.’” Rosen uses this story to point up what he regards as weaknesses in Schenker's approach: a disregard of proportions and a tendency to minimize the salient and explicit features of a work by putting them into “tiny notes.” But Schenker's followers often make a similar point from an opposite perspective; it is almost a truism among them that interest and individuality reside precisely in those “tiny notes” of the middleground and foreground, the fundamental structure acting as guarantor of coherence. Thus Victor Zuckerkandl writes, “The main interest is not in the background itself but in how background and foreground are connected, i.e., the middle ground.” According to Allen Forte and Steven E. Gilbert, “The closer we get to the background, the more similar any two pieces are likely to appear.” And here is Felix Salzer explaining melodic analysis: “The structural tones are the spine of a melody; they establish its basic direction. What makes a structural line live, however, are the many different types of prolongation, since they provide the character, rhythmic interest, and color of a melody.”

Without in the least disputing these statements, I should like to suggest a slightly different way of viewing background structure.