Miscategorising Text Types

Martin (1992: 518):

Similarly reconstruction may be either an account of what someone was told to do or what they did, and generalising texts may be either macro-proposals (e.g. assembly manuals) or macro-propositions (e.g accounts of how a product works).

Blogger Comments:

[1] This purports to cross-classify texts according to (Martin's) mode and mood: imperative vs indicative.  However, an account of what someone was told to do — just like an account of what someone did — is "indicative" in these terms, not "imperative"; the "imperative" lies in the 'telling what to do', not in the accounting of it.

[2] This is inconsistent with the meaning of the word 'generalise': to make a general or broad statement by inferring from specific cases.  That is, generalisations are propositions, not proposals, and, moreover, neither 'assembly manuals' nor 'accounts of how a product works' constitute text types that can be accurately termed "generalising".

The practice of insisting that a word means whatever one wishes is termed Humpty Dumptyism.

∞

The fundamental confusion here is that the discussion is presented as theorising mode, the systems of the textual metafunction at the level of context.  Here, instead, the discussion is concerned with the interpersonal dimension of linguistic content — semantics (speech function) and lexicogrammar (mood) — of registers.  The confusion is thus simultaneously along three theoretical dimensions: stratification, metafunction and instantiation.