On Some Computational Models of Music Theory

John Rothgeb's pioneering 1968 Yale dissertation successfully showed that two early eighteenth-century theories about how to realize an unfigured bass were inadequate and incomplete, and suggested that the fault lay in the inadequacy of the general theory of music at that time. The example from that thesis in his present paper, Simulating Musical Skills by Digital Computer (1980), is a good example of a process now being described as constructing an explicit base in a formal language equivalent to an text in a natural language (van Dijk 1977). But most of John Rothgeb's present paper is of a more discursive nature. I shall venture to criticize two methodological positions implicit in his text. Since both positions are subscribed to by many of the most venerable music theorists active in this field, it should be kept in mind that my attack is not so much on positions held by John Rothgeb as on the assumptions of a very reputable community of music theorists.