Many, if not most, contemporary philosophers of music contend that the musical expression of an emotion does not depend upon the perceiver feeling that emotion.1 That is, while these scholars do not necessarily deny that emotional responses to music are valuable, even inevitable, they do not view such responses as properly part of the emotional content of the work itself, or even properly part of an aesthetic experience of the work (as opposed to a more informal one). Some argue this on the basis that musical emotions lack intentional objects: if a piece expresses sadness, for example, this could not consist in the listener feeling sad for the piece gives him nothing to be sad about.2 Others invoke the problematic paradox of negative emotion: if music really made us feel sad, why would we want to listen to it? (Davies  and Levinson  explore this issue.) Still others observe that pieces often evoke feelings they do not express and vice versa; as Goodman states, “whatever emotion may be excited [by music] is seldom the one expressed” (1976:47). As a consequence, even when the emotions expressed by the music and those felt by the listener happen to coincide, the latter are nonetheless incidental to musical expression. Equally incidental is what composers feel when they compose. As Peter Kivy avers, “It is unthinkable that I should amend my characterization of the opening bars of Mozart’s G–minor Symphony (K. 550) as somber . . . if I were to discover evidence of Mozart’s happiness . . . during its composition” (1989:14–15). While none of these arguments is ironclad, collectively they make a compelling case for the fundamental separation of musical expression and arousal. Departing from this axiomatic distinction, I will attempt to demonstrate precisely how music possesses emotional content, how emotional qualities arise from (if not completely inhere in) musical form and structure. In the process, I hope to demonstrate that musical emotions need not be of the most general sort, such as joy and sorrow (what Kivy designates “garden– variety” emotions) but may at times be subtle, specific, and “cognitively complex.”3 This view runs counter to those of Eduard Hanslick, Peter Kivy, and Susanne Langer, to cite three prominent aestheticians. However, rather than dismiss their ideas, which are extremely valuable in their own right, I hope to assimilate them into what I feel is a more satisfying theory of musical emotion—one that accounts for more nuanced shades of emotion than these theorists allow. In what follows, I shall neither comprehensively survey thepublished positions on musical emotion nor offer an entirely original theory; rather, I shall attempt to synthesize these well–known stances toward musical emotion, incorporating my own, music–analytic approach (which utilizes formal, motivic, Schenkerian, and implication–realization methods). My two related aims are (a) to account for the multiple musical parameters and structural levels that generate emotional content; (b) to explicate the means by which music in certain instances is able to convey specific emotions.