Since Ferguson's initial Diglossia paper of 1959, an enormous amount has been written on the phenomenon. Surprisingly, very little has been published on the situation in Indonesian, given that, first, it conforms very closely to the concept as originally proposed by Ferguson and that, second, it is the sole national and official language of the fourth most populous nation in the world.

This paper takes Ferguson's article as its starting point and compares the ways in which Indonesian conforms to his model and the areas in which it appears to be different. It is shown that Indonesian, at least in the speech of educated Jakartans, conforms to a remarkable degree to the concept as described by Ferguson in his examination of four 'defining' languages. It does, however, differ in one important way: rather than two distinct forms of Indonesian with a clear boundary there is a continuum between the two extremes. As the social situation becomes more formal L features are gradually replaced by H features, though not at a consistent rate. Some characteristics of the L variety are replaced in semi-formal speech, while others persist even in quite formal situations. Moreover, there is considerable variation in usage between individuals of even quite homogeneous background.

The paper considers the speech of educated people in Jakarta and examines a number of variables, showing how L variants are gradually replaced by H variants as the social situation becomes more formal.

Although some writers state that strict functional compartmentalisation of the two varieties is an essential characteristic of diglossia, it is argued here that Indonesian should be recognised as being in a diglossic state; the concept is too useful for analysing social aspects of the language to be discarded. This position receives support from Ferguson's recognition in a later paper of a continuum between L and H in his four defining languages, and this may be a general characteristic of diglossic languages.