This past Monday I used this site to provide a “status report” on the performances of some of the leading soloists in the current San Francisco Opera season. For a variety of reasons, accuracy of pitch was very much on my mind while writing this report. This led to my taking what many would have thought was an inappropriate digression:
While attending a master class at the San Francisco Conservatory of Music, I once heard a violinist attribute an observation about pitch to Dorothy DeLay, one of the most important violinists to have taught at the Juilliard School. DeLay supposedly once said that no violinist ever hits pitch exactly; the good ones know how to adjust faster than the audience can hear the difference.
I immediately tried to justify that digression by asserting that “what holds for string players holds just as much for vocalists.”
Nevertheless, any stickler for logic would have quickly observed that my claim was a product of both two layers of hearsay (my reporting on a teacher reporting on what his teacher said) then subjected to speculative hypothesis. In addition, the whole chain of reasoning was rooted taking a scientific approach to the psychological dimension of the acts of both making music and listening to it. I am therefore happy to report that, in my efforts to understand these matters better through a generous amount of background reading, I am now in a position to clear up some loose ends. However, as is often the case in scientific research, tying up one set of loose ends often introduces a whole new set of them.
In 1982 Diana Deutsch, a Professor of Psychology at the University of California in San Diego responsible for several pioneering studies of a variety of activities related to making music, published The Psychology of Music, a collection of eighteen papers, each reviewing a different aspect of the discipline. One of the papers was entitled “Perception of Singing.” The author was Johan Sundberg from the Department of Speech, Communications and Music Acoustics at the Royal Institute of Technology in Stockholm, Sweden. Back when I was doing my own research in this area more seriously, I knew of Sundberg’s work and tended both to respect and to enjoy his publications.
With that as context, I wish to reproduce one paragraph from his contribution to Deutsch’s book:
In a review of a number of investigations, Seashore (1938) included a wealthy documentation of fundamental frequency recordings of professional performances of various songs. The trend is that long notes are sung with an average fundamental frequency that coincides with the theoretically correct value. This is in agreement with the experimental findings reported previously. On the other hand, they often “begin slightly flat (about 90 cent[s] on the average [an interval slightly less than a semitone]) and are gradually corrected during the initial 200 msec of the tone.” Moreover, a great many of the long tones were observed to change their average frequency in various ways during the course of the tone. Bjørklund (1961) found that such deviations were typical for professional singers as opposed to nonprofessional singers. One possible interpretation of this is that pitch is used as a means of musical expression.
Now from a back-of-the-envelope point of view, I am willing to accept an interval of 200 milliseconds as “faster than the audience can hear the difference.” What I found just as interesting, however, is that this result dated back to experiments run by Carl Seashore that may have originally appeared in his book The Measurement of Musical Talent, which was published in 1915. The 1938 book was entitled Psychology of Music; and, unless I am mistaken, Seashore wrote it for the general public, rather than just for his academic peer group. As a result, the book was reprinted several times with enough distribution that it would not surprise me that it would come into the hands of a teacher at Juilliard who spent time browsing in new and used bookstores.
Nevertheless, I still must confess that nothing in the above reasoning involves hard data. Indeed, I shall still have to dig up Seashore’s book to find out just what his methods were. On the other hand, no matter how sound they were at the beginning of the twentieth century, we now have far better measuring technology, not to mention far less noisy data sources. (Just try to imagine what a commercial recording of a professional performance must have sounded like 100 years ago!)
As a result, while my argument may not have ascended to the heights of “rigorously scientific,” I feel that I have taken some bold steps in advancing from “speculation on hearsay” to “plausible reasoning.” My next step, therefore, should be to advance from a highly reputable but dated survey paper to a more recent effort to reproduce the results with more advanced equipment. Nevertheless, I cannot keep from being amazed at Seashore’s method. At a time when most psychologists tended to restrict their experiments to the confines of their own laboratories, Seashore appreciated that commercial recordings could be just as valuable serving as data sources as providing a novel (at that time) source of entertainment.