Ordinary language is all right.
One could divide humanity into two classes:
those who master a metaphor, and those who hold by a formula.
Those with a bent for both are too few, they do not comprise a class.
Daniel Lindquist on relative identity in Heidegger (with reference to Geach and Frege).
'We imagine nothing, nothing can be imagined': that means something like, we have no sense of possibility. In contrast to a skepticism about knowledge, where the conclusion drawn bears on our ability to know—it's not enough, we just can't do it—a skepticism about the imagination would seem to bear more on the imagination's vitality. 'No sense of possibility' means: it is dead in us, has died. But where the skeptic about knowledge looks back at what seemed to be knowledge and infers, 'we never really knew', perhaps nonetheless anticipating a time when what we do know, did know, can know, will be properly established, the skeptic about imagination seems more focused on now: what must be denied is not so much that we have imagined, now that imagination is dead, but that it is dead at all. The skeptic's conclusion, or conviction, is opposed by the insistence: 'it's not really dead'. Or: 'it can't be'. This is a weak opposition. Because the skeptic about imagination's reply is: 'it is'. And if it's dead, it's dead: nothing can be brought back from the dead. Everything living dies. (But: there is a spring. Hence Kora, Spring and All, The Descent of Winter. Hence Thoreau's year of observation.)
Cavell adverts regularly to an alternative skepticism of this sort, associating it with Romanticism, with the imagination, with acceptance or rejection of the world, with deaths and lives. But I want to say that it's not staged like this—not with the imagination aligned in this way with knowledge, as it appears in his recitals. But his recitals of the skeptical argument about knowledge play some important role in connecting the two, for they put the 'I' onstage. And his reinterpretation of skepticism is written as an 'I' (all first-person plural business aside), as are, modulo this or that, many of the marginally canonical 'texts' he regularly appeals to in just the same connection with the alternative form of (imaginative) skepticism. His focus on external-world skepticism can obscure the significance of this, because the proximity of his rehearsal to the Cartesian model or its traditional successors can suggest overmuch that the I's perspective is exhausted in an exploration of the I's capacity to know, and that any salience a philosophical writer's own I has naturally pertains to problems of knowledge. The significance of Cavell's writing as an I is, I think, first of all that he has reinscribed the problems he discusses within, let's call it, the problematic of the I. And thereby reinscribed the I of the philosophers within that broader tradition. (Why 'reinscribed'? Look at something like the Investigations, or even Austin. Considered as texts, they are far less I-ish than Cavell's convincing portrayal of them can suggest. He draws out, reconstructs, I's he sees in them. That's part of his work.)
Many literary modernists pursue solutions to their difficulties by intensifying the thingification of the work of art: Stein, Williams, Joyce. Writing philosophy rather than literature, Cavell chooses personalization of whatever sort rather than thingification. Montaigne's 'consubstantial' book is a closer fit, as a model. But if you consider that in a certain light, even Wittgenstein's writing can seem more thingified than personalized (Larochefoucauld: 'a heap of diverse thoughts'), it's not obvious that the need to write philosophy (in contrast to literature) must lead to Cavell's choice.
Even in Paterson there is still a lyric I, the poet's I. As in Williams' ground-clearing works, in which the prose often assumes a journal form. Cavell's alternative or expanded canon (really a restored one) is populated by philosophical authors who write as I's, I think crucially because they too are concerned with this alternative skepticism, in whatever form it takes in them. But as with Cavell, it is hard to find in them a 'skeptical argument' for this sort of 'conclusion', even if only to treat or reject it. I want to say, as Cavell does of Wittgenstein having 'discovered the problem of others for philosophy', that most of these philosophers are fully aware of the problem, of whatever skepticism about the imagination is. But there's no Descartes for it. Or too many.
In the prologue to Kora in Hell, Williams sounds a traditional note: 'the thing that stands eternally in the way of really good writing is always one: the virtual impossibility of lifting to the imagination those things which lie under the direct scrutiny of the senses, close to the nose. It is this difficulty that sets a value upon all works of art and makes them a necessity. The senses witnessing what is immediately before them in detail see a finality which they cling to in despair, not knowing which way to turn'. —As if what modernism is, is exactly a skepticism in literature: not a failure to know things but a failure to imagine them, to see them imaginatively, with a parallel disaster consequent upon that failure: an anxiety, hardening into a grim conviction—the passage quoted leads on quickly to the idea of despair, sounding like Thoreau, sounding like Schopenhauer, sounding like modern daily life—that we imagine nothing, that nothing can be imagined. Nor (thus) lived.
—As Paterson begins, with the sleeping city-man's dreams (i.e. us), automatons animated by the river, who 'neither know their sources nor the sills of their disappointments', walking about the city, it's their unroused desires which are the ground of some other possibility. Desire which would lift. But lift nevertheless from 'things', those things, the 'no ideas but' (in) things. (Again, paralleling that track of necessity leading to skepticism, undeniable, natural.)
'Yes: not just theirs. Also wrong. That's why they need to replace whim and expression with clarity and rigor.'
Elenchus not as a technique of refutation, but as an art of making expressions of thought serve self-knowledge; of making them perceptible, noticeable.
—Theirs, and, once expressed, not just theirs. Not once it's out in the open.
—Of course, expressions are yours, or mine, or ours—'owned', per 'other minds' jargon. So students expressing their thoughts in writing would have to be able to believe that what they were doing was theirs. Both before and after they submit an assignment for a grade. What does it take to believe that—to come to be a person who can believe that about what they write, say? And, to believe it there, then, after a dozen or so years in that institution, a school?