This article is part of a series of articles whose goal is to explicate traditional voice-leading practices in Western music. This particular study was intended to determine whether polyphonic practice conforms to known properties of the human auditory system.
When the number of parts or voices in a polyphonic texture increases, it becomes increasingly more difficult for listeners to follow or track the individual parts. ( Huron, 1989). In particular, it becomes more difficult to hear ("track") inner voices.
The ideal population would be the totality of all polyphonic music. For practical purposes, the population for this study is the totality of polyphonic works by J.S. Bach.
The sample consisted of 75 fugues, including the complete 48 fuges from the Well-tempered Clavier, 15 fugues from the Art of Fugue, the 2 Ricerari from Bach's Musical Offering, and 10 randomly selected organ fugues. This is not a random sample. Note that the sample is biased towards fugues rather than polyphonic works in general.
Actually, the population and sample consist of voice entries rather than the works themselves.
"Voice Entry." What do we mean by an "entering voice?" How do we operationally define this?
"Inner Voice." How do we define an "inner voice?"
In formulating an operational definition for "Voice Entry" the paper attempts to find a mechanical procedure whose results are very similar to the author's intuitions of what a voice-entry is. Why not use the author's own intuitions? The researcher wants to avoid any possible criticism of possible experimenter bias.