Wednesday, 8 April 2026

Systemic Functional Linguistics Under Constraint — 2 Register Is Not a Variable Set

Once context is treated as an environment, a further move follows almost automatically:

register is redefined as a set of variables that influence linguistic choice.

Field, tenor, and mode become:

  • parameters of a situation
  • dimensions along which variation occurs
  • factors that condition what is likely to be said

This appears to preserve the Hallidayan terminology.

It does not preserve the Hallidayan model.


Because the moment register is treated as a variable set, its function changes.

It is no longer:

  • a functional variety of language
  • a subpotential within the semiotic system

Instead, it becomes:

a set of independent factors acting on the semiotic from outside.


The consequences are immediate.


1. Register Becomes External

If register consists of variables, those variables must:

  • exist independently of semantic organisation
  • be identifiable prior to linguistic actualisation
  • and operate as conditions on choice

This places register outside the semiotic.

Even if it is still called “context,” it now functions as:

  • an environment
  • a set of inputs
  • a source of influence

The distinction between context and environment collapses.


2. Realisation Becomes Causal

In the Hallidayan model:

  • semantics realises context

But if register variables act on linguistic choice, this relation reverses.

Now:

  • context influences semantics
  • variables determine probabilities
  • linguistic forms are selected in response

Realisation is no longer a stratified relation.

It becomes a causal process.


3. Choice Becomes Response

Systemic choice is no longer:

  • the structured potential of the semiotic

It becomes:

  • a response to external conditions

The system is reinterpreted as:

  • a decision-making mechanism
  • operating under constraints imposed from outside

Which means:

the semiotic is no longer internally sufficient.


This entire structure depends on a single assumption:

that variation in language must be explained by variables acting on it.


This assumption is rarely stated.

But it is what makes the model appear explanatory.

If we observe that language varies with situation, it seems natural to say:

  • different fields lead to different lexical choices
  • different tenors lead to different interpersonal meanings
  • different modes lead to different textual organisations

From this, it is inferred that:

field, tenor, and mode are variables that influence language.


But this inference does not hold.

Because it depends on transitivity.

To say that a variable influences choice is to say:

  • something passes from the variable to the semiotic
  • altering the distribution of possible selections

Without this passage, “influence” is empty.


But transitivity has already been excluded.

Nothing passes between domains.
No mechanism connects context and semantics.

So the idea of variables acting on language cannot be sustained.


At this point, the account again faces a forced choice.

If register is treated as a variable set, then either:

(1) Variables determine linguistic choice

In which case:

  • the semiotic is grounded in external conditions
  • meaning is produced by something outside itself
  • and internal sufficiency is lost

Or:

(2) Variables are redescribed as semantic features

In which case:

  • field, tenor, and mode become internal components of the semiotic system
  • and the distinction between context and semantics collapses

There is no third option.

Because without transitivity, variables cannot act.

And if they cannot act, they cannot function as variables.


So the concept of register as a variable set fails in the same way as context-as-environment.

It attempts to preserve:

  • distinction

while requiring:

  • a relation that eliminates it.

The alternative is not to discard register.

It is to restore it.


Register is not:

  • a set of variables
  • a collection of external factors
  • a parameterisation of situation

It is:

a functional variety of language—
a subpotential within the semiotic system, viewed from the pole of potential.


This must be read perspectivally.

From the pole of potential:

  • register appears as a patterned subspace of semantic possibility

From the pole of instance:

  • the same variation appears as a text type

There are no variables acting on the system.

There is only:

  • structured variation within the system

Field, tenor, and mode do not:

  • influence language
  • determine choice
  • act as inputs

They specify:

  • the contextual configuration that is realised in semantic actualisation

This restores the correct relation:

  • semantics realises context
  • register is a functional organisation of that relation within the semiotic

No causation.
No influence.
No external variables.


What, then, of variation?

How do we account for the observable fact that language differs across situations?

Not by invoking variables.

But by recognising:

variation is the differentiation of semiotic potential under constraint.


Different registers are not produced by external conditions.

They are:

  • different organisations of meaning
  • actualised under coupling with value
  • and reproducible as patterns of constraint

This removes the need for variables entirely.

There is no:

  • input
  • parameter
  • or conditioning factor

There is only:

  • semiotic potential
  • its patterned differentiation
  • and its constrained actualisation

Closing formulation

Register is not a set of variables acting on language.

It is a functional variety of the semiotic—
a patterned subpotential realised in semantic actualisation.

The moment it is treated as external,
realisation becomes causal,
and the semiotic loses its internal sufficiency.


With this, a second pillar falls.

Because once register is no longer a variable set, the entire model of:

  • language responding to context
  • choice as conditioned selection
  • variation as externally driven

loses its foundation.


Next: the final technical retreat—realisation itself, and the persistent attempt to make it bidirectional.

No comments:

Post a Comment