Sunday, 14 August 2016

Misinterpreting Hasan And Proposing Theoretical Inconsistencies

Martin (1992: 551):
For Hasan, text structures are derived from generic structure potentials conditioned by choices in field, tenor and mode — with most of the optionality apparently determined by tenor and mode.  This suggests that systemic relationships among different text structures are equivalent to relationships among field, mode and tenor options; and the question of systemic relationships among generic structure potentials does not arise.  Challenging the first of these suggestions, and redressing the second, Martin (1985) suggests reformulating generic structure potentials as system networks and realisation rules as with Ventola's (1987: 15) reformulation of Mitchell above, proposing a speculative network and realisation rules for service encounters by way of illustrating how this might be done (Martin 1985: 253-4; Fig. 7.21 below).

Blogger Comments:

[1] To be clear, for Hasan (1985/9: 64), a generic structure potential is the structural potential of a text — as the largest unit of the semantic stratum — of a given genre (text type/register).  For Hasan, genres that share the same contextual configuration of field, tenor and mode features share the same structural potential at the level of semantics.  In SFL theory, the relation between context and semantics is realisation: context is a higher level of symbolic abstraction than semantics.

[2] This false inference derives from Martin's earlier false claim (see here) that Hasan associates obligatory elements of text structure with field (p546). For Hasan (1985/9: 62), the obligatory elements of text structure are the elements that define the genre (text type).

[3] This is a false inference in that it blurs important distinctions between the theoretical dimensions of axis, stratification and instantiation:
  • The relation between structure (syntagmatic axis) and system (paradigmatic axis) is realisation; semantic structure realises semantic system.
  • The relation between semantic stratum systems and context stratum systems is realisation, not equivalence.
  • Different text structures are a matter of registerial (text type/genre) variation at the level of semantics.
[4] Martin (1985) is thus challenging his own misinterpretation of Hasan.

[5] The proposal here is to model registerial (text type/generic) variation in semantic structure as a system network at the level of context.  The confusion here is thus along two dimensions simultaneously:
  • the cline of instantiation: system (potential) vs register (subpotential);
  • stratification: context vs semantics.
[6] Ventola's network models registerial variation (service encounters) as a system network with semantic structure specified by realisation rules activated by the selection of features.  That is, it models subpotentials (registers) as potential (system) and specifies semantic structure as its realisation.

The network is thus inconsistent with the architecture of SFL theory in that it posits a midway point on the cline of instantiation (subpotential) as the systemic potential that specifies semantic structure.  According to the architecture of SFL theory, the systems that specify semantic structure are the systems of the semantic stratum (axially) and context (stratally).