Martin (1992: 580-1):
What are the implications of Hasan and Bernstein's work for the interpretation of ideology as system? This is a question which is in some respects premature. Work on mapping out the fashions of meaning constituting a culture at the level of ideology has only just begun (most of Hasan's own work in this area remains unpublished as of 1989). …
All of this is compounded by the fact that fashions of meaning and the more abstract notion of coding orientation need always to be interpreted in context — that is, with respect to the genre and register through which they are manifested. Given our present understanding of these planes, this is a challenging task; and certainly not one for which even a provisional network of oppositions can be provided at this time.
 The work of Hasan and Bernstein has no implications for the interpretation of ideology as system. The reason for this is that neither work is concerned with ideology. One way to define 'ideology' is as a system of ideas and ideals, especially one which forms the basis of economic or political theory and policy. In contrast, from the perspective of SFL theory, Bernstein's work on codes is concerned with how social structures affect the semantics of registers, and Hasan's work is concerned with that semantic variation.
 The implication here is that Martin cannot devise an ideology system until Hasan has provided one that he can alter.
 This is misleading. It misrepresents the work of Hasan and colleagues on semantic variation as framed within Martin's model of ideology.
 Given that Martin locates fashions of meaning and coding orientations on a contextual plane of ideology, this misconstrues lower levels in the stratification hierarchy as the context of higher levels. This is the opposite of what the hierarchy represents.