Some musicians will spend forever tuning their instruments before launching into their recital. Yet according to research presented in the
Edward Burns of the University of Washington in Seattle, USA, and Adrianus Houtsma of the Eindhoven University of Technology, came to this conclusion after setting out to establish whether sensitivity to relative musical pitch is innate or acquired. The musical scales of most cultures are made up of pitch intervals for which the frequencies are related in some simple ratio. The interval of a fifth, for instance, corresponds to a frequency ratio between the root and fifth note of 2∶3.
One might therefore imagine that musical scales have fixed upon on these intervals because, in their mathematical simplicity, they 'sound' right even to an untrained ear. In that case, non-musicians should be able to identify an out-of-tune fifth, say, even if they could not name it as such. Then all musicians are doing is learning labels for innate predispositions towards a perception of harmony.
But is this really so? Burns and Houtsma set out to explore the question by repeating two earlier experiments – one reported in 1976, the other in 1994. In the first study, listeners were asked to compare two tones and to say whether one (the comparison tone) was tuned higher or lower than the other (the reference tone). But the reference tone had no fundamental or 'root' frequency: it was made up only of harmonics. It is a well-known acoustic phenomenon that in such cases the listener can nevertheless 'hear' the missing fundamental, sometimes even if just a single harmonic is provided – the auditory system somehow inserts the missing frequency, rather as the visual system fills in the 'blind spot' in our visual field.
In the later study, a harmonic was removed from a complex tone, and was presented to listeners after they had heard the remainder of the tone. But the missing harmonic was sometimes slightly detuned, and the listeners were tested for their ability to identify this.
These tests sound complex, but are both really just ways of identifying whether listeners can discern detuning in the 'natural' intervals of the musical scale. When these experiments were first conducted, they seemed to indicate that listeners have a good and rather universal ability to detect detuning. But the 1994 experiments used only three subjects, all of whom had a musical training, while no indication was given in the 1976 experiments of whether the subjects were musically adept or not.
Burns and Houtsma found on repeating both experiments that the abilities of their subjects were very mixed. In particular, they did far worse than the subjects of the 1976 experiments. But most strikingly, the ability to detect detuning was considerably better for trained musicians who were already accomplished at identifying relative musical pitches and intervals. In other words, there seemed to be a stronger element of learning than of 'innateness' in the ability to tell whether sequential tones are harmonically related. The more you play, the study suggests, the better your ear for music.