Friday, November 28, 2008

"MA!" - Do You Want Mom or Your Pony?

Absolute Pitch, Speech, and Tone Language:
Some Experiments and a Proposed Framework

By Diana Deutsch, Trevor Henthorn & Mark Dolson
Music Perception, Spring 2004, Vol. 21, No. 3, 339-356

Review by John Picone

(Note: the response to this study is also informed by discussion of the monthly MIMM meeting at McMaster University, Friday, November 21, 2008)

Although I thoroughly enjoyed preparing for my grade 10 RCM practical piano examination many years ago, there was one aspect of my lessons I still recall with something of a resentment. A good friend of mine at the time had his lesson immediately after me at the old conservatory on James Street South in Hamilton. Often, our teacher, Mrs. Eileen McManamy, would have our lessons overlap for 15 or 20 minutes and conduct our ear-training exercises together. I was always baffled when Paul would identify a chord as follows: “Oh, that’s a diminished chord in the third position.” He would then pause, turn to me with a wry smile, and continue: “In the key of F#!” Although such ability didn’t count for anything on the exam, I always felt Paul was miles ahead of me in his musical talent.

The rare attribute of absolute or perfect pitch is generally defined as the ability to name or produce a note of a particular pitch in the absence of a reference note. In this study, the researchers base their comparative experiments on the fact that absolute pitch involves, of necessity, verbal labeling. That is, one sings the pitch “A” in response to the verbal label “A” or, “please sing A.” Likewise, when hearing the above frequency, the person with absolute pitch verbally labels it: “That’s A.” “The verbal labeling of pitches necessarily involves speech and language… it is tied to linguistic processing” (p. 342). If this is the case, then clearly a person only has to learn 12 labels – the notes within the octave – that accompany the 12 pitches. For this study, the question is not so much why some people possess absolute pitch, but why it is not universal.

The researchers posit that there is a critical period in a child’s development for the acquisition of both speech and language, and absolute pitch. They hypothesize that it is exposure to pitch in language during this critical period that has a significant influence on the development of absolute pitch. Like learning a second language, young people acquire absolute pitch almost automatically, effortlessly, without specific training.

Their experiments compare the speakers of tone languages such as Mandarin or Vietnamese, and intonation languages such as English. While both kinds of languages employ lexical tones that involve pitch contours, it would appear that tone languages also employ pitch heights (registers). While it is not clearly pointed out in the article, it would seem that the pitch “height” is relative to the speaker’s normal range. What is a “high” pitch for one tone language speaker may indeed be of a different frequency than that of another speaker but the same relative to their normal speaking voice pitch range. Another difference (although this, too, was not clearly explained in the study) seems to be one of consistency. In English, an intonation language, for example, let’s consider two responses to the question, “How are you?” While two people may respond with, “Fine, thanks!” in different intonations, it’s clear that they’re really saying they are not fine at all. Intonation in English is more closely aligned with semantic meaning than lexical meaning, that is, the meaning of an actual word. To illustrate, the researchers examine the word, “ma” in Mandarin. They note that, depending on the register and the pitch contour, the word can mean “mother,” “horse,” “hemp,” or a reproach (p. 343). As it happens, one of the musicians in my music class speaks Mandarin and noted that, indeed, there was a fifth use of this word: to designate a question. I had the opportunity to record her saying five sentences with each of these lexical meanings. The pitch contours and heights were clearly different in each case.

"The question then arises as to which features of pitch are critical to conveying lexical meaning in tone language. If these features were purely relational, then the present discussion would be irrelevant to the genesis of absolute pitch. If, however, absolute pitch were employed to signal lexical meaning, then we would have the beginnings of an explanation as to why speakers of intonation languages, such as English, find absolute pitch so difficult to acquire in adulthood. The study reported here was carried out as a test of the hypothesis that absolute pitch is indeed treated by tone language speakers as a critical feature of speech. The hypothesis entails that tone language speakers would evidence absolute pitch in speech processing, and that the memory representations of the pitches of speech sounds would be qualitatively different for speakers of tone and intonation languages" (pp. 344-345).


All three experiments involved subjects reading aloud from word lists. Their voices were recorded and pitches measured and compared for consistency.
In the first experiment, seven Vietnamese speakers read out a list of ten words and repeated this reading on a different day. The study does not indicate how much time had elapsed. The results, according to the researchers, showed little difference in pitch and suggest that “the subjects must therefore have been referring to stable and precise absolute pitch templates in enunciating the list of words” (p. 346).

The second experiment was much the same as the first. In this case, fifteen native Mandarin speakers were the participants. Again, they read lists of words which were recorded and analyzed for pitch variation. The second experiment, however, had the participants read the word list twice in succession on one day and again on the second day. The goal was to compare pitch variances between successive readings with readings on the different days. “Remarkable consistencies were again obtained” (p. 347).

The third experiment was identical to the second except that native English speakers were the participants. The intonation (English) language speakers’ pitch consistency was then compared with that of the tone (Mandarin) language speakers.

"…the Mandarin and English speakers showed roughly the same degree of pitch consistency in enunciating their word lists twice in immediate succession, but the Mandarin speakers were significantly more consistent than the English speakers in enunciating their word lists on different days. Thus the Mandarin and English speakers performed differently on this reading task, both qualitatively and quantitatively, with the English speakers showing less pitch consistency across days" (p. 350).

An interesting comparison made by the researchers in their discussion refers to the neuropsychological literature showing that “whereas pitch patterns are processed for intonation purposes primarily by the nondominant hemisphere, the processing of lexical tone primarily involves the dominant hemisphere” (p. 350). They also hypothesize that “different individuals of the same sex who speak in the same dialect should match up in terms of the absolute itch levels with which they enunciate words” (p. 350).

What about the relationship between absolute pitch in language and absolute pitch in music? While the researchers acknowledge that the present study does not address this, they “surmise that absolute pitch for music is acquired by speakers of tone language as though it were a feature of a second language” (p. 351). They do note, referring to a 1999 survey of students in U.S. conservatory, university and college music programs by Gregersen et al., that “a higher prevalence of absolute pitch was reported among those students who described their ethnic background as Asian” (p. 351).


While this study is fascinating in its comparison of tone and intonation languages, with likewise interesting results from the experiments, it is not anchored in a clearly defined conceptual framework. The term “absolute pitch” is defined in the first sentence of the study as “the ability to name or produce a note of particular pitch in the absence of a reference note” (p. 339). This is the musical definition with which most people are familiar. However, the term is never clearly defined as it refers to language. It is not possible, for example, that the pitch height and contour of the Mandarin word “ma” meaning mother refer to actual musical frequencies all the time. The frequency of the lexical tone used in producing this word would naturally change with age. This reader surmises that the pitches used in producing a word in a tone language are “absolute” in the sense that they are consistency relative to the speaker’s natural range. For example, a young tone language speaker may be compared to an alto saxophone while an older tone language speaker to a baritone saxophone. The tone of the Mandarin word “ma” meaning “horse” is described by the researchers as “low, initially falling and then rising” (p. 343). Let us assume that, musically, this pitch contour starts on C, drops to A and then rises to E. Although in different registers on the two saxophones, would they, indeed, be these actual notes all the time? Is this what the researchers mean when they refer to a “pitch template”?

There also seem to be two obvious inclusions in this study which were not carried out. The first is the comparison between two people of the same sex and age producing words in a tone language. The apparatus was already in place for this. The second, given the reference to neuropsychological literature and hemispheric dominance, was a comparison of brain activity between tone and intonation language speakers when saying the words.

Many people around the MIMM discussion table found the experiments in this study to be unconvincing, noting poor controls and methodology.

Some interesting observations were made by the MIMM participants. Perhaps, given the paucity of expressed emotion in Asian culture, there is a limited pitch range in their speech heightening overall attentiveness and sensitivity to pitch. Is there a genetic component involved? There was the observation that we, as intonation language speakers, associate meaning with pitch in many facets of our lives: the dial tone is always F below middle C, most two-tone door chimes have the interval of a major third below. And everyone recognizes the Toronto Transit Subway tones before the doors close.

As to the greater prevalence of absolute pitch among Asian music students, the discussion participants noted that music lessons are often begun at a much earlier age than in Western cultures. Further, music students are “weeded out” so that only the best continue to the conservatory or university level.

Perhaps the most interesting question brought forth had to do with the possibility of a pitch code in the motor cortex. Is there pitch processing involved in controlling our vocal chords? What about the embouchure of the trumpet player? The most important question was whether or not anyone had done a pitch map of the motor cortex. Has anyone studied this?

“Dibs!” said an eager student.

And the meeting adjourned.

No comments: