A further note to Jay's helpful clarification of relations between descriptive
categories.
Halliday distinguishes between relations of instantiation and those of
realization. In his initial use of the weather/climate analogy he described
relations between the context of a culture and of situation types within the
culture as an instantial relation, and similarly the relation between a
linguistic system and a specific (linguistic) text. However, he distinguished
this relation from that between a culture and language, and that between a
situation and a text, which he describes as one of realization. On this view
the linguistic system is a realization of the culture, and a text is a
realization of the situation. (This view of the relation does not, of course,
exclude realization of meaning through other semiotic systems.)
There are many implications, but to take up just one: if a text is an
instantiation of the language system then a detailed linguistic analysis of a
text necessitates some account of the systemic potential which the text
instantiates. Similarly, description of texts as instantiations of system
often/usually require analysts to elaborate their working account of the
system. This is, I think, how many people working in an SFL framework
interpret parole/langue relations.
The reference to Halliday's discussion is a conference paper "The notion of
'context' in language education" given to the International Conference on
Language Education, Vietnam, 1991. I will try to have a part of the paper
made available through our department's home page next week since it contains
a useful diagram of these distinctions.
Geoff