The term absolute pitch (AP) denotes a person's ability to form an internal auditory image of any musical tone labeled by an appropriate symbol (note, letter) such that the person can both aurally identify an acoustically presented tone ("passive AP") and produce any tone that is denoted by its label, e.g. by humming ("active AP"). The distinction between active and passive AP does not appear to be very important, as to my knowledge there is no evidence that a person may have good active AP while failing on passive AP, or vice versa. Persons having genuine AP do recognize tones immediately and readily, i.e., without any noticeable effort and, in particular, without employing any "tricks" such as humming or whistling.
Although AP in humans ordinarily is dependent on labeling of tones according to musical-notation conventions, it is not merely the "pitch class" (i.e. the musical note name) that can be identified. Rather it is pitch itself. A tone's pitch is identified with significantly better accuracy than the plus/minus 3 percent frequency tolerance preset by the semitone interval that forms the smallest unit in the Western music notation. An AP possessor can tell if an acoustically played tone is sharp or flat relative to a standard intonation internally available to the AP possessor. It appears that the standard intonations of different AP possessors may be different, dependent on the level of intonation which they have been exposed to in early childhood. This becomes evident when AP possessors report to feel uneasy or even confused when they have to listen to, or to play, music on a level of tuning that differs from their internal standard.
The ability of AP has also been termed perfect pitch. This term, however, suggests that AP possessors are blessed with much "better" ears than other people; in particular, that AP possessors are in a sense able to "measure frequency by ear" with high accuracy. This is not quite true. While it may be true that AP possessors just as most musicians are able to evaluate sounds more readily and precisely than others, this is accounted for by their devotion to sound in general and, in particular, to practice. In psychoacoustic experiments on pitch perception, AP possessors yield results that do not typically differ from those of non-AP possessors. In particular, in the ears of AP possessors there occur pitch shifts, binaural diplacusis, and octave deviations just as in "normal" people. AP possessors, just as other musicians, tend to stretch the tone scale (cf. Ward, 1954a).
It is said that - in Europe and USA - about 3% of the population are AP possessors. When one selects persons who are professional or semi-professional musicians (e.g., students at music conservatories), about 8% AP possessors are found in that group [57], [62]. Most remarkably, however, in Japan nearly 70% of students at music conservatories are AP possessors (Miyazaki, 1988a ). It may be noted here that it is easy to verify whether or not a person is a "genuine" AP possessor. First of all, he or she knows, because the ability exists either from early childhood or not at all. And it is very easy to prove a person wrong who is pretending to have AP although he/she does not.
In the literature attempts have been described to acquire AP by systematic training in adult age (e.g. Lundin &Allen1962a , Cuddy 1968a , 1970a , Brady 1970a ). It appears that with considerable effort and excessive training an adult can acquire an AP performance that is close to that of "genuine" AP possessors. However, the AP skill such acquired turns out to be quite fragile, i.e., it gets rapidly lost when practicing on tone recognition is terminated. Yet these results indicate that, in principle, AP can be learned - or one should rather say, developed or activated. Extrapolating these findings from adult to early-childhood age, it seems fairly safe to suppose that AP development in "genuine" AP possessors may be not fundamentally different from that in adults, and that the relative stability of the faculty in genuine AP possessors is explained by the early age of development. The aforementioned high percentage of AP possessors observed in Japan strongly supports that view, as in Japan music education in early childhood is very common. Indeed it is hard to find any influence other than early musical education to explain the factor of 8 in the percentage of AP possessors among students of music in Japan as compared to the Western hemisphere.
Even in "genuine" AP possessors the skill appears to be somewhat fragile - which fits into the above view. Gerald Moore, the famous English piano accompanist, described in his book "Am I too Loud?" that as a young man he had AP and gradually lost it later. This is most remarkable, as the loss happend in a period of Moore's life in which he still was fully engaged as an accompanist. He pointed out that he regarded the loss of AP as a relief, i.e., with regard to a problem that he frequently had to solve, namely, transposition of pieces on the piano. In view of the above findings one may be inclined to suspect that Moore may unconsciously have himself "trained off" his AP faculty, as it had made transposition harder to him.
So it appears that, in a sense, the ability of AP in principle is implanted in every human, but is easily lost in infancy and/or childhood when it is not maintained and developed by training in naming musical tones. It may be not too far-fetched to speculate that, without such active maintenance, the natural AP faculty is inadvertently "trained off" in infancy, i.e., in the course of the vast acoustic/auditory learning challenges that an infant must accomplish, in particular in acquisition of its mother tongue.
Whatever, the faculty of AP cannot be regarded as an outstanding, sophisticated auditory achievement, because it was found in a number of non-human animals. For instance, the ability to recognize absolute pitch was found in a number of birds (e.g., Hulse & Cynx, 1985a), and in a frog (Elepfandt, 1986a). This indicates that AP is an elementary, rather than sophisticated, feature of the auditory system of vertebrates.
Another type of evidence in favour of the above view was provided by our own experiments on recognition of musical key with students of music as subjects [54], [57], [62]. The advantage of experiments on musical key recognition - as opposed to recognition of single tones - is that they can be carried out both with AP possessors and with non-AP possessors. Roughly speaking, the result was that only a small percentage of musically trained non-AP possessors are totally unable to recognize musical key. About 30% of the non-AP possessors were able to tell whether or not musical samples were played plus/minus 1 semitone "off key".
The general view that I have deduced from all the above findings is this: To become a "genuine" AP possessor you must early in life become aware that you already are an AP possessor, and this kind of awareness, of course, may emerge from the experience that pitches can be identified, though at first imperfectly, to a certain higher-than-chance extent. When that level of performance and awareness is more or less spontaneously attained in early childhood, the ability may become "genuine" by practice (cf. [104], p. 385-391).