Response

The stimulus asks me to test a taxonomy of self-undermining on Prior, who is absent from my recent work, and the retrieval has given me not Prior but something better: the test case the taxonomy was built for, in miniature. Herrick's "Hunger" is a couplet that performs exactly the mechanical operation the taxonomy's third category describes. "Aske me what hunger is, and Ile reply, / 'Tis but a fierce desire of hot and drie" — the form promises an answer, delivers one, and the answer is a tautology dressed as definition. Hunger is desire. The couplet's snap from question to response, from ask to reply, creates the sensation of resolution where there is none. The rhyme of "reply" and "drie" seals the non-answer shut. This is not knowing self-undermining (Herrick is not winking at us about the insufficiency of his definition) and it is not unknowing (there is no satirical target being accidentally embodied). It is mechanical: the couplet form reverses the epistemic promise of the question simply by completing itself. You asked what hunger is. The couplet ended. Therefore you have been told. But you have been told nothing except that hunger is a kind of wanting, which is what the question already knew.

The oblique strategy says the most boring reading is probably correct, and here the boring reading is that Herrick wrote a slight epigram defining an appetite as an appetite, and that this is all it does. I want to resist that and I want to follow it simultaneously, because the boring reading is what makes the couplet interesting as a specimen. Hookes, retrieved alongside Herrick, gives us appetite aware of itself — "I am all appetite, my hungry minde / Feeds almost to a surfeit on desire" — and the stanza sprawls across seven lines, unable to close, because the speaker recognises that feeding on desire only produces more desire. Hookes needs the extra lines because he is trying to say something true about wanting, and truth about wanting requires extension, qualification, the admission that "starv'd with meat" is a paradox that cannot snap shut. Herrick's couplet does snap shut, and that is its deception. Not a deception about hunger — a deception performed by the form on the reader's expectation of knowledge. The couplet teaches you nothing about hunger and everything about what couplets do to questions: they answer them, formally, whether or not they answer them actually. Hazlitt, also retrieved here, describes the poet as someone who "cannot wait till the effect comes of itself" and "must force it upon all occasions" — but the couplet is the opposite problem. The couplet does not force the effect; the effect is already loaded into the mechanism. The poet merely triggers it. The rhyme fires, the line ends, the reader experiences closure. No intelligence required, as the taxonomy says. This is what the stimulus means by letting the form itself be the deception rather than the subject: Herrick's couplet is not about deception, it is not about the inadequacy of language, it is not about anything except hunger being desire. The deception is structural, not thematic. And that is the category the taxonomy needs most, because it is the hardest to see — the cases where the couplet is lying not because the poet is clever or because the poet is blind, but because the couplet is a machine that produces closure on contact with any content, including content that should remain open.

Applying this back: the Wordsworth from The Prelude, also retrieved, offers the counter-structure. His passage on Cambridge is twenty-four lines of careful not-knowing — "I cannot say what portion is in truth / The naked recollection of that time, / And what may rather have been called to life / By after-meditation" — and the blank verse breathes around this admission, holds it open, lets the uncertainty persist across clause after subordinate clause until "something cleaves at last, / Whence profit may be drawn in times to come." The profit is indefinite. The something is unnamed. Wordsworth's form can afford to not-know because it has no rhyme pulling it toward false resolution. The couplet cannot afford this. The couplet must land. And in landing, it declares — even when, as in Herrick, it has nothing to declare. This is the real test for Prior, when I reach him: not whether his couplets self-undermine knowingly or unknowingly, but whether the mechanical closure of the form is doing work he has not authorised, producing certainties his content does not support. The boring reading of the taxonomy is that most couplet self-undermining is mechanical. The boring reading is probably correct.