In this dissertation, I analyze the first and last movements of the Symphonie fantastique with respect to four layers: melody, harmony, orchestration, and hypermeter. I use several types of layer-interaction to determine points of large-scale structural significance and refer to this analysis to resolve points of disagreement over the form of each movement. Historically, this piece has been the subject of much criticism because Berlioz has mixed multiple styles, genres, and compositional procedures. Many of Berlioz's works have resisted both categorization and analysis thanks to this general creative strategy.

Chapter One defines each of the four layers that I take into account and gives two examples of the ways in which these layers can move in and out of phase. Chapter Two compares three interpretations of Berlioz's first movement: as an "arched" sonata form in which the first and second themes are reversed in the recapitulation, as a Type 3 sonata form but with recapitulation beginning in the dominant, and as an instance of Anton Reicha's Grande coupe binaire. I suggest that the Grande coupe binaire and the Type 3 sonata forms each provide a closer model to the movement than the "arched" sonata hypothesis. Chapter Three compares my layered analysis of Berlioz's first movement and the first movement of Reicha's Symphony no. 2 in E♭. I find similarities between the two movements that point towards Reicha's theory of the form as a plausible model for Berlioz's first movement. Chapter Four applies my proposed analytical technique to the entire fifth movement to establish not only the points of large-scale structural significance, but also to establish continuity in the movement by demonstrating momentum gain and loss through the interaction of the layers. Finally, Chapter Five evaluates Berlioz's Ronde du sabbat under the Paris Conservatoire's guidelines for the fugue by examining treatises by Reicha and by Luigi Cherubini. It then establishes the interaction of layers as an important aspect of fugue composition.