Western culture has been shaped for the past several thousand years by its use of writing as a vehicle for thought, making a written support indispensable for any academic study. Music is no exception to this rule, and would seem extremely difficult to analyse in depth unless first reduced to the form of a written score, i.e., a transcription in the case of music from an oral tradition. The essential transience of music requires that its movement through time be fixed in writing as a substantive ‘reference text’ for the living reality. This is what the ethnomusicologist's attempts at transcription aim to provide, whatever the geographical or ethnic source of his material. If this is true of monodic music, it is even more so in the case of polyphony, where the simultaneity of events results in a much more complex musical lattice. Transcription is thus all the more necessary, though commensurately harder to achieve.

In his Theory and Method in Ethnomusicology (1964), Bruno Nettl highlights both the need for transcription and all the intrinsic difficulties attending it. He distinguishes two main approaches to the description of music: the first consists of analysing and describing what one hears, while the second involves writing down what one hears and then describing the audible phenomena, relying on the observations contained in the transcription.