Raymond Williams was presumably not the first person to notice that the terms “working class” and “middle class” belong to different schemas of categorisation; but notice it he did:

[T]he most significant effect of this complicated history was that there were now [circa 1840] two common terms, increasingly used for comparison, distinction or contrast, which had been formed within quite different models. On the one hand, middle implied hierarchy and therefore implied lower class: not only theoretically but in repeated practice. On the other hand working implied productive or useful activity, which would leave all who were not working class unproductive and useless (easy enough for an aristocracy, but hardly accepted by a productive middle class). To this day the confusion reverberates.

Williams, R., Keywords (London:Fontana 1976), pp. 64-65.

The source of the “confusion” is, in Williams’s account, the “complicated history” preceding it, which is first and foremost a history of usage. The disjunct terminologies are formed within disjunct social realities, in which terms acquire their implications “not only theoretically but in repeated practice”. We have here the seeds of a theory of performativity - of legally binding usage, which materialises the predicates of class (as, in some theories, of gender) by ordering them into being. But that isn’t the way Williams chooses to take things, preferring in general to view language as an arena of conflict rather than an instrument of oppression.

The terms in Keywords are presented as polysemic. They are shown to have a finite, enumerable range of possible senses: a structure (as in the title of Empson’s “The Structure of Complex Words”). They also form a structure between themselves, as Williams notes in his introduction: a “vocabulary” is here “a shared body of words and meanings in our most general discussions”. This structure does not eliminate conflict, but it does allow it to be placed under the purview of a synopsis.

As with the other terms included in his “vocabulary of culture and society”, Williams’s historical sketch of the development of “class” qua “keyword” therefore projects an underlying confidence in the ontological consistency of the field of linguistic usage. That is, he is confident that such confusions of sense as may arise within this field will invariably reveal themselves, on closer inspection, to have traceable historical roots: to be confusions of what once was distinct and unconfused.

The proper force of a term like “class”, however, does not only derive from the fact that its incompatible predications (“working”/”middle”, or indeed, “middle”/”ruling”; note that “working”/”ruling” encounters no such difficulties) emerge from socio-historically distinct milieux, but from the fact that it functions as a sign, in the present, of the ontological inconsistancy of the social field. This inconsistancy in turn threatens the consistency of Williams’s field of study, since a “vocabulary of culture and society” can only be convoked if “culture” and “society”, terms that Williams himself highlights as subject to significant conflicts of interpretation, are stable enough to secure it.

What Merle Brown’s Double Lyric termed divisiveness in poetry is the production of a conflict of interpretation that destroys any possibility of synopsis. One cannot describe the roots of the conflict, show how it has arisen and for what causes, or describe the current relationship of the antagonists in any complete and neutral way (one can give multiple partial and incompatible views of the matter, but these cannot be formed into a single coherent account).

Such incoherence is the sublime of literary criticism, the normal function of which is to form coherent accounts of interpretative procedures. The sublime moment occurs when criticism is able to indicate, by way of something presented in the poem, that which is unpresentable by means of criticism’s normal function: the point at which the ontological consistency of the critical worldview fails, and the poem unfolds into a separate, counterfactual world of its own. Literary criticism is not however menaced by this unpresentable inconsistency: on the contrary, it continually apostrophises it, as if its life depended on it - which it does. “Yale” deconstructionism was neither the beginning nor the end of literary criticism’s taste for this sublime (Brown’s hero, in his account of “divisiveness”, is F. R. Leavis, improbably enough).

“Class” is similarly the sublime of “symbolic” politics, inasmuch as the latter is a form of literary criticism with a wider social remit (and not that much wider). If the tensor of class indicates the Real of political antagonism, it does so specifically from the standpoint of a social theory committed to an aesthetic vision of political coherence.