Tuesday, 2 September 2025

Purview Confounded: Duplicating Grammar and Semantics as Context

Doran, Martin & Zappavigna (2025: 78):

This opens space for reconsidering examples such as (49) above along the same lines, replayed here as (51). Externally speaking, considered in terms of whether or not mass extinctions were caused by changes in sea level or disruptions in the food chain, the proposition is aired – the may indicates that the author is not tied to whether or not this is the case. A response could render this proposition with something like No, they weren’t – focusing on whether or not this was the cause, rather than any negotiation of whether it could be the cause. On the other hand, the possibility of them being caused this way is asserted (through the use of a declarative without any tagging). One could focus on the could and render the possibility of them being caused by saying something like No, they couldn’t have. Asserting this suggests that rejecting the position along these lines would be more socially risky. But for academic discourse in particular, the play between the two purviews given through NEGOTIATION (the statement) and ENGAGEMENT (through the modality could) allows an author to have their cake and eat it too – to assert something baldly while not tying themselves to its actuality.


ChatGPT Comments:

In this excerpt, Doran, Martin & Zappavigna (2025: 78) propose a dual-layered system of purview in which external purview (– speaker; + listener) “airs” a proposition, while internal purview (+ speaker; – listener) simultaneously “asserts” its possibility. The example in (51), concerning mass extinctions, is intended to illustrate how an academic author can assert something baldly without committing to its factual actuality.

From a systemic-functional standpoint, however, this theorising is deeply problematic. The authors relocate what are fundamentally clause-level interpersonal distinctions and engagement up into the realm of context (tenor: purview). Grammar and semantics, language, are treated as independent contextual variables, effectively conflating content and context in a manner that SFL explicitly resists. This move is conceptually confused: the internal/external purview distinction does not describe a separate level of social organisation but merely re-describes the interplay of hedging and assertion already encoded in modality.

Further, the dual purview framing introduces unnecessary complexity without explanatory payoff. By splitting a single clause into “external” and “internal” propositions, the system duplicates semantic distinctions at the level of tenor. The distinction between commitment to actuality and commitment to possibility is culturally and register-specific, and cannot reliably support a generalised system of context. The authors’ discussion of “having their cake and eating it too” highlights this: it is less an insight into social interaction than a re-labelling of standard hedging practices in academic English.

In short, the dual purview model is theoretically confused, duplicative, and inconsistent with SFL’s canonical stratification. External and internal purviews do not constitute a new distinction of context; they are a misapplied reinterpretation of what is already realised grammatically and semantically. The excerpt exemplifies the broader issue with purview: by elevating grammar and semantics into tenor, the authors obscure the very mechanisms they claim to describe.

No comments:

Post a Comment