A culture begins to drift from realism the moment it treats naming as creating. Once the dictionary becomes the criterion, the territory is already being asked to agree with the map. The older tradition reversed the courtesy: the dictionary is correct only when it conforms to what is the case.
This is not a thesis about words. It is a thesis about what a word is for. A word is a small, ordered service performed by a community on behalf of something the community did not invent. The community can perform the service well or badly. It cannot, by performing the service, make the thing being served into something else.
And yet the contemporary ear, when it hears "what is the case," reaches for scare quotes. Surely "the case" is itself a construction. Surely "is" is just a verb of social agreement. Surely the question of whether the world precedes the description is itself only another description.
The hardest thing for a constructivist to admit is that the case for constructivism would itself need to be the case.
The self-refutation
The hardest thing for a constructivist to admit is that the case for constructivism would itself need to be the case. If every assertion is only a vantage point, the assertion that every assertion is only a vantage point is also only a vantage point — and there is no longer any reason to prefer it to the realist one. The position dissolves the moment it is stated with the seriousness it asks of others.
Aristotle saw the trick fifteen times before lunch. To deny that something is, you must first identify the something you are denying. Identification, in turn, presupposes a being to identify. The denial cannot get out from under the affirmation it is parasitic upon. This is not a clever debate move; it is a structural feature of intelligible speech.
A more honest constructivism would simply admit: "We prefer to talk this way." And then the conversation could begin again — about whether the preference is a true one, in the older, embarrassing sense of true.
The dictionary and the territory
Consider what changes when we treat the dictionary as the criterion. A new word, properly inscribed, redescribes a thing. The redescription, repeated enough, becomes the thing. The thing, having been the thing, becomes a residue — an inconvenient ancestor of the term that has now eclipsed it.
But the residue is the world. And the world keeps producing evidence — quietly, without rhetoric — that the term has not in fact replaced it. A redescribed bridge, badly built, still falls. A redescribed marriage, badly lived, still ends. A redescribed body, badly understood, still bleeds. The dictionary cannot inherit the cost of being wrong.
The dictionary cannot inherit the cost of being wrong.
This is the realist's only proof, and it is enough: the world does not collect on the term. It collects on the agent. When the agent is mistaken, the term does not soften the consequence. The term floats; the world bills.
Why the older view is also a kinder one
A common reflex objects that realism is harsh — that to say "the world does not care what you think" is to enthrone an indifferent universe over the soft preferences of persons. The reflex is wrong twice. First, the indifference is liberating: it is precisely because the world does not bend to thought that thought can correct itself. A world that conformed to belief would be a world in which error was impossible — and so a world in which knowledge was also impossible. The realist is not enthroning indifference; they are conserving the conditions under which knowing has any meaning at all.
Second, the realist tradition has always been the tradition that takes the soft preferences of persons most seriously. Persons cannot be helped if their pain is treated as a description rather than a fact. The realist physician inspects the wound; the constructivist physician revises the chart. One of these is medicine.
It is sometimes said that the realist position is metaphysically extravagant — that it commits us to "natures" or "essences" or some other shadow population beyond what science requires. But the position is more modest than that. It commits us only to this: that some claims are made true by the way the world is, and that the way the world is is not produced by the claim. The further questions — what kinds of things there are, how they are individuated, how their identities are stable — can be answered with restraint. They begin, however, only after the dictionary has been put back on the shelf.
A modest closing test
When you next encounter a sentence you suspect of slipping the leash — a sentence that uses "discourse" or "construction" or "narrative" where the older speech would have used "is" — try the simple substitution. Replace the contemporary verb with the realist one. If the sentence still says something the speaker would defend, the speaker is a closet realist and only the vocabulary is wandering. If the sentence collapses into evident nonsense, the speaker has been describing a dictionary all along, and mistaking the description for the world.
The first room of any serious thinking is here: in the recognition that the world is not, and never was, ours to spell. We did not write it. We are reading it. And the reading goes well only when the page is allowed to be older than the eye.