Halliday

Works

One grammar, one social theory, and the study that founded grammatical stylistics.

• An Introduction to Functional Grammar (1985, and in later editions with Matthiessen) is the descriptive apparatus the field works from.
• Language as Social Semiotic (1978) sets out the social theory: register, context of situation, and language as behaviour potential.
• Cohesion in English (1976, with Ruqaiya Hasan) supplies the account of how a text holds together across sentences.
• The 1971 transitivity study of a novel's language showed that a consistent pattern of grammatical selection can be described precisely and shown to do literary work — the demonstration that made grammatical stylistics possible.

Central ideas

Four, and the first is the one everything else hangs from.

• The three metafunctions: every clause simultaneously construes experience (ideational), enacts a social relation (interpersonal), and organises itself as a message (textual). They are strands in one wording, not separable channels or separate sentences.
• Meaning potential: a system network of choices, where what was said signifies against what could have been said and was not. The paradigmatic axis, not the syntagmatic, carries the meaning.
• Transitivity: processes are typed — material, mental, relational, verbal, behavioural, existential — each with its own participant roles. Saying that someone did something and saying that someone is something are different selections with different consequences, and the grammar makes the difference visible.
• Register as the configuration of field, tenor and mode: what is going on, who is involved and on what terms, and what part the language is playing. Tenor is where the interpersonal metafunction meets the situation.

Distinctive vocabulary

Three terms with precise senses that ordinary usage erodes.

• Register: the configuration of field, tenor and mode in a situation type. Not a level of formality.
• Interpersonal metafunction: the clause as an act between speaker and addressee — mood, modality, evaluation. Not a topic and not tone.
• Grammatical metaphor: the reconstrual of a process as a thing, most visibly in nominalisation. Not figurative language.

Debates and disagreements

One disciplinary split and one problem the theory has not solved.

• The long standoff with formal, innatist linguistics: a functional account of language as social behaviour against a formal account of internal competence. The two programmes have rarely engaged each other's evidence, and the split has cost both.
• Whether the descriptive categories, and process types in particular, are motivated by the data or stipulated. The boundaries are genuinely fuzzy and analysts code the same clause differently.
• Whether a descriptive grammar can bear the critical conclusions later traditions draw from it — an argument conducted mostly by others, about his apparatus rather than his claims.

Intellectual relationships

A London-School inheritance turned into a systemic theory.

• J. R. Firth was his teacher, and context of situation and the system-structure approach come from that lineage.
• Malinowski supplies context of situation and context of culture from the anthropological side.
• The Prague School supplies theme and rheme, and with them the textual metafunction.
• Basil Bernstein's sociology of language is the parallel programme connecting social structure to linguistic selection.
• Fairclough and Wodak, both in this lens lineage, built critical programmes on his grammar; appraisal theory is the internal development of the interpersonal metafunction.

How it changes this reading

Halliday is named in the Discourse-analytic lens lineage and in two through-line readings.

• For the lens he supplies the grammar that makes a claim about a wording checkable — and supplies it without a critical programme, which is the distinction the lens depends on and the one most often collapsed.
• For the reading about calibration language, transitivity is the exact instrument: naming what someone did selects a material process with a participant, and naming what someone is selects a relational process attributing a quality. The second is not answerable to a next instance and the first is, which is the through-line's point stated as a grammatical fact rather than an ethical preference.
• For the reading about pre-naming, the borrowing is the priority of the interpersonal over the ideational: publishing an architecture in advance is primarily an act establishing the terms on which later exchanges will be read, and reading it as information transfer misses what the utterance is doing.
• They are not evidence about anyone's wording, intent, or awareness.

Useful comparisons

The lens names three figures at descending distance from politics.

• Halliday supplies the functional grammar itself, with no critical programme attached.
• Norman Fairclough builds Critical Discourse Analysis on that grammar, with an explicit political purpose.
• Ruth Wodak works on discourse in historical context, with attention to prejudice and to institutional discourse over time.

Where the ideas meet

All three refuse the same view of form.

• None treats grammatical form as neutral packaging for content.
• All three hold that a selection among available forms carries social meaning.
• Each anchors its claims in identifiable features of actual texts.

Where they part

His presence in this lineage is not a critical endorsement.

• Halliday describes the resource; Fairclough draws political conclusions from selections within it. Citing Halliday in support of an ideological reading borrows the apparatus without the argument that would license it.
• Wodak's historical grounding answers part of the circularity objection that a purely descriptive grammar never raises and a purely textual critique leaves exposed.
• The system was built for description and for language education. Its use in critique is an application others made, and the entry keeps that attribution straight in both directions.

Limits

• A grammatical analysis establishes what a clause does, not what a speaker intended or believed in producing it.
• Process-type coding is contested at the boundaries, so two analysts can describe the same clause differently and both be defensible.
• The metafunctions are analytic strands in one wording, so no reading may treat one of them as the utterance's real content.

Criticisms

The objections are about reliability and reach.

• Inter-analyst reliability is a standing weakness: the same text yields different transitivity and appraisal codings in different hands, which limits quantitative claims built on them.
• The terminology is heavy enough to reduce uptake and to make the analyses hard for non-specialists to check, which matters for a method whose value is checkability.
• Formal linguists object that a functional account cannot explain the syntactic constraints their programme addresses.
• The general claim that grammar is shaped by function is difficult to falsify at the level at which it is usually asserted.

Common misreadings

Five uses these placements do not license.

• Treating the metafunctions as separate channels rather than simultaneous strands of one clause.
• Treating nominalisation or the agentless passive as inherently evasive; the grammar identifies the selection, and whether it obscures depends on context.
• Citing his grammar as though it carried Critical Discourse Analysis's critical conclusions.
• Using register for a level of formality rather than for the configuration of field, tenor and mode.
• Reading a transitivity coding as evidence about what a writer thought.

What remains outside this idea

• It cannot establish what any speaker intended, believed, or noticed.
• It cannot convert a grammatical description into a fact about a person.