SISSON has said: 'any form which finds its way into human speech, however articulated or disjointed, may find a place in poetry.' This sounds very permissive. And yet Sisson as a reviewer isn't permissive at all, but very angry about 'the already overwhelming number of transgressions in verse by people nature never intended for poets.' How can this be? The truth is that Sisson can afford to be so permissive about syntax ('articulate or disjointed') because he is very exacting indeed about something else: about rhythm, and more precisely about inventiveness in rhythm. Neither he nor his admirers have been altogether above-board about this, about how consistently Sisson is an unscannable poet, and how small and dubious a margin his poetic theory leaves for the poet who writes to be scanned, metrically. In his -theory and practice alike he is very much a free-verse poet, though he will legitimately protest at the permissiveness that 'free' seems to connote, and so I use the less misleading term 'unscannable'. Sisson's verse, like most of the nursery-rhymes he so much admires, resists scansion according to any of the standard accentual-syllabic metres. His praise of nursery-rhymes ('the original and most important element in poetic education', he says) was explicit in his review of Donald Wesling's The Chances of Rhyme (TLS 12 September 1980), where he also declared, finely and yet I think unconvincingly: 'To liberate the language, even ever so little, from the shadow of what has become familiar, and walk a few ...