Nicholas J. Shea

Abstract: Current and historical methods of metric analysis often assume that the first of two concurrent and equal-duration pulses is stronger than the second. This, however, is not the case in all repertoires. For example, it has been shown that Verdi’s midcentury operas often place emphasis on even-numbered beats (Rothstein 2011). This paper shows this metric trend to be even more prevalent in a corpus of nineteenth-century operatic excerpts (1809–1859).

​

I present a formalized decision tree that classifies phrases according to anacrusis length and prosodic accent, showing where large-scale metric accents fall within a phrase. This model produces three metric types which align with Rothstein’s previous work. Compositional and historical features (e.g., language, premiere date, librettist, etc.) were tracked alongside type to determine whether preferences for certain metric forms were more prevalent in certain contexts. This indeed was the case. For instance, use of even-beat-emphasis meter increases over time, although odd-beat-emphasis meter remains most common. Individual composers also show a significantly distinguishable preference toward each type of meter. These results not only confirm that the highest concentration of even-beat-emphasis meter occurs in Verdi’s midcentury operas, but that Verdi is the primary user of this type overall. Language and composer nationality do not significantly affect an excerpt's metric type; only Verdi shows the most distinction in these areas. With these findings, I argue against using nationalist language to identify metric types (e.g., Franco-Italian meter and German meter) and instead propose a more nuanced understanding of nineteenth-century metric conventions.

​

Some findings:

• German meter is most prominent metric type during this period [SLIDE 3]

• Verdi is the primary user of Franco-Italian meter [SLIDE 4] 

• French-language operas rarely use Franco-Italian meter [SLIDE 5]