Fuller’s seven-chapter book of ‘puzzles in poetry’ is charmingly entitled Who is Ozymandias? I do hope that he wasn’t thinking of that old donnish joke about ‘Ozzy Manders, Dean of Kings’. If so, it can only have been a passing quirk, in a critical study whose quirkiness is intelligent and enlightening about such different matters as odd self-editing changes, the ouija board as poetic source, slithery titling practices, Browning’s innocent misuse of ‘twats’, arrangement of poems by number, and why The Waste Land was so often misinscribed as The Wasteland.

It transpires that Browning thought a twat was a wimple, but I shan’t pursue this error, incipiently comic though it is. What Fuller is after in each of his chapters is a plethora of ways in which poems – especially well-known ones – are overdetermined: in the poets’ intentions, in their forgettings, and in the course of mechanical production. These artefacts are richly hybrid texts, which have sought formal, aesthetic coherence. As Helen Vendler has written,

While it is true that we are initially drawn to poems by their passions, their questions, and their tonal urgencies, we are convinced by them, finally, insofar as they can invent formal means for their impelling motives.

A wry poet himself, Fuller can be concisely funny, in deadpan: for instance, when he writes, ‘Imagine Dante’s great work to be called simply The Cantos’; or when he notes in passing that ‘it was even Auden who had a vast behind’, like the invoked Sphinx. He is also nimbly ahead of our readerly annoyance when he writes at the start, ‘It need hardly be said that in the text the generalized male pronoun stands firmly for the female one as well.’

Some larger themes appear again and again, even if they are only peering over the shoulder of our critic–detective. They include creative intention, carelessness, reader response, and downright muddle. I do like his saying ‘Suppose a poet means one thing, and another is actually printed. Who, then, is doing the meaning?’

If Fuller is wittily alert, he is also dogged in his Sherlock Holmesian forays. Thus, he hunts down the inconvenient implications of ‘wasteland’ at some length, and gives a tireless close reading of that wonderful early poem of Auden’s, ‘O Love, the interest itself in thoughtless heaven’. Another, too, on Tennyson’s ‘Mariana’ and its relation to Shakespeare’s creaky play.

This draws us into Fuller’s psychological methods. If a poem is overdetermined, woven into so many things, then it will call forth some kind of analysis, even of a psychoanalytic kind. To take an extreme sample, his reading of ‘The Hunting of the Snark’ and ‘Jabberwocky’ is so coarsely sexual as to be reductive. One should ask here why he can eroticise most of the proper names but not Jabberwocky itself; and his reading Snark backwards to get the German noun for wreath or garland strikes me as altogether too tricksy.

He is a free agent, John Fuller, a lone cowboy with lots of ideas. In Who is Ozymandias? (yes, that bloke both is and isn’t Rameses II), I found myself delighting in his examinations of ‘This Lime-Tree Bower My Prison’, ‘A Cooking Egg’, and ‘Falling Asleep over the Aeneid’, three very different objects. Loose threads or wayward antecedents have been picked up all over the place.

Let me confess that I have been another young reader who was led away by the opening metaphor of ‘Time, you old gipsy man’, and a little later beguiled by Yeats’s misprint in ‘soldier Aristotle’, which cast a shadow over that philosopher for some years. The book is full of local sugarplums like these.

One of Fuller’s own collections is beguilingly entitled The Beautiful Inventions (1983): surely this is what successful poems of lyric scale should be; and partly invented by the attendant Muse, hovering over the page. Sagacious Margaret Atwood has remarked in passing that ‘fiction requires will power; poetry requires the abdication of will power’. That sounds pretty neat.

But how does a poem begin? With linear thinking? Something more webbed, I’m sure, and that’s what the Muses were for; but if we’ve mislaid them we must press on into secular arguments. Physically, one usually begins for me as I’m taking an urban or suburban walk; or else on a plane, or more terrestrially on a tram: the body is distracted by rhythmical travel, letting imagination do its job and generate the first few lines of what will be a new work. It’s the melody of this clutch of lines that will discover the poem. Mind you, this emergent poem might even be governed by sounds and rhythms: it might be an essentially auditory construct. As it goes on, the lines should build up and acquire a solidity, the work of art acquiring its shape harmonically. Even the sounds and echoes can tell a poet what the current poem is saying.

Goethe once complained about the limits of merely oral utterance, remarking that ‘The more I think about it, there is something futile, mediocre, even (I am tempted to say) foppish about speech.’ Well, the old fox in lion’s clothing yielded to that temptation, giving us his account of that division of verbal practice that is definitely not poetry. Talk is not poems. Poems are not only aesthetic productions: they are charged with a whole array of meanings, some of which the poet in question actually intended.

This creative borderline or shadowland is what Charles Lamb was evoking long ago when he wrote that ‘The poet dreams being awake. He is not possessed by his subject but has dominion over it.’ Such dominion includes the abdication of will power, no doubt, but also encompasses the odd counties and puzzles that Fuller has here explored for us, so engagingly.

This is a book about puzzles of a higher aesthetic family than, for example, crossword puzzles. It is highly readable, and also dippable. And a dippable book is a delightful possession. Who is Ozymandias? would be just the thing to pop on your bedside table for a little night reading.