Wednesday, 24 December 2025

The Grammar of Possibility: 1 Language Is Not a System of Representation

It is still common—both in philosophy and in the sciences—to treat language as a system for representing an independently structured world. Words, on this view, stand for things; sentences describe states of affairs; meaning succeeds when the description matches what is already there.

This picture is so familiar that it often goes unnoticed. It feels like common sense.

And yet, it is precisely this picture that makes language appear perpetually deficient: too vague, too ambiguous, too context-sensitive, too metaphorical. Compared to mathematics or logic, natural language seems messy—an unreliable medium that must be disciplined, formalised, or corrected if we are to think clearly.

This series begins by rejecting that diagnosis at its root.

Language is not a failed representational system.

It is not trying—and failing—to mirror the world.


The Representational Temptation

The temptation to treat language as representational is understandable. Formal systems encourage it. In mathematics, symbols are defined precisely; in logic, well-formed expressions map cleanly onto truth conditions. When these systems work, they give the impression of a transparent correspondence between symbol and structure.

But this impression arises after severe constraints have already been imposed. Formal systems do not begin with meaning; they carve out narrow regions where meaning has been stabilised in advance.

Natural language begins elsewhere.

When we speak, we do not start with a fully articulated world and then select labels for its parts. We begin with relations—situations, interactions, perspectives—and we construe them. Language does not copy structure; it brings structure forth by making certain relations salient, negotiable, and repeatable.

To mistake this activity for representation is to confuse outcome with function.


Meaning as Construal, Not Mirror

If language were primarily representational, meaning would be assessed by accuracy: how closely an expression matches an external arrangement of things.

But this is not how meaning actually works.

Meaning is first a matter of how a situation is construed: what counts as relevant, what is foregrounded or backgrounded, what roles are made available, what kinds of continuity are assumed. These are not optional decorations added after representation; they are the very conditions under which anything can be said to be “the same situation” at all.

Consider a simple utterance:

The meeting was difficult.

Nothing here mirrors a ready-made object called difficulty. The utterance configures a field of relations: participants, expectations, tensions, trajectories. Different construals—emotional, procedural, political—remain available, not because language is imprecise, but because the situation itself is relationally open.

Language does not fail to pin the world down. It enables the world to be taken up in particular ways.


Breakdown as Diagnosis, Not Failure

The usual evidence against natural language is well known: ambiguity, vagueness, metaphor, indexicality. These are treated as flaws to be eliminated if meaning is to be made rigorous.

But what if these so-called breakdowns are not defects at all?

Ambiguity arises because language allows multiple construals to coexist until further relations stabilise one of them. Vagueness persists because many situations do not demand sharp boundaries in order to function meaningfully. Metaphor works because meaning is not tied to literal correspondence, but to relational projection across domains.

In each case, what appears as imprecision is actually flexibility. Language remains responsive to context precisely because it is not locked into a single representational grid.

Where formal systems must exclude ambiguity to operate, language exploits it to remain live.


The Priority of Relation

Once representation is abandoned as the master metaphor, a different picture emerges.

Language is a grammar of possible relations. It offers patterned resources for construing experience: ways of organising processes, participants, qualities, evaluations, and perspectives. Objects are not primitive; they emerge as relatively stable cuts within ongoing relational activity.

This is why attempts to ground meaning in reference to discrete things repeatedly run aground. Reference is not a mapping between words and objects; it is a relational achievement, sustained across contexts through shared construals.

Meaning does not point outward toward a finished world.

It works inward and sideways, configuring the very space in which a world can show up as meaningful at all.


Reversing the Hierarchy

Once this is recognised, the familiar hierarchy between language and formal systems quietly inverts.

Logic and mathematics no longer appear as purer, more accurate versions of meaning. They appear instead as local extractions—disciplined regions where construal has been constrained tightly enough to support invariance and calculation.

Their success is real. Their precision is extraordinary.

But it is local.

Language does not aspire to that kind of precision because it serves a different function: not the measurement or constraint of relations, but their enactment and negotiation.


Nothing to Repair

Seen this way, nothing is wrong with natural language. It does not need to be repaired, purified, or replaced.

Its openness, flexibility, and context-sensitivity are not obstacles to meaning; they are its conditions of possibility.

Language does not represent a pre-given world.

It is the medium through which worlds become available to us in the first place.

In the next post, we will look more closely at grammar itself—not as structure imposed on meaning, but as constraint: the quiet machinery that makes meaning doable without fixing it in advance.

No comments:

Post a Comment