What makes a tune pentatonic? This is one of those questions, like 'How long is the coastline of Great Britain?', that prove to be much less simple than they appear at first sight. The obvious answer, 'using the five notes of the pentatonic scale, and no others', does not take us very far. 'Auld Lang Syne', for instance, is pentatonic in this straightforward sense, but not all pentatonic melodies are so obligingly regular. All the tunes quoted in the previous chapter are in some sense pentatonic; but one of them, 'Na more utushka' (Ex. 3.12), includes an extra-pentatonic b, and most of the others use fewer than the complete five notes. 'Pentatonic' melody may be, strictly speaking, sub- or super-pentatonic. Moreover, not every sub-pentatonic melody makes an equally pentatonic impression. Triadic tunes like the railway porter's chant (Ex. 3.4) and 'Shack Bully Holler' (Ex. 3.5)—or for that matter any bugle call—seem somehow less pentatonic than those built on the children's chant.
Why should this be? There are several points to consider.
First, melodic hierarchies, like any others, are by their nature ambiguous. The triad is an entity in its own right, but it is also a part of the pentatonic scale. In one sense, it is pentatonic; in another, it is not. At the other extreme, every heptatonic or even chromatic tune will reveal a pentatonic skeleton when its semitones are removed. So deep-seated is the instinct to search for pentatonic underpinnings, that composers of the atonal school had to draw up elaborate rules to frustrate it.
To make matters worse, not all these modal skeletons are equally pentatonic. So it is quite possible to have two tunes, each making full use of all seven notes of the heptatonic scale, that are 'pentatonic' to different degrees.
The next (and most important) point is that it is of no consequence that the pentatonic scale happens to contain five notes; what really matters is the degree of dissonance it permits (or excludes). It is a sort of a musical crystal that grows out of consonance. Interval may be added to interval in any order, and, provided one keeps within a certain limited range of dissonance, the result will always be pentatonic. (This is one reason why attempts to discover the origin of the pentatonic scale will always be futile.)
The last point is that this 'range of dissonance' is deliberately ambiguous. It