Myth and Misconception
In the English speaking world it is almost axiomatic that the most commonly repeated assertions concerning renku technique are false. Nowhere is this more apparent than in the question of cut and uncut verses.
On the one hand expert A assures us that the internal verses of a renku sequence may be considered as being like haiku - the ones with three lines that is. On the other, editor B rejects our submission because an internal verse employs a syntactic disjuncture at the end of line one, and it is well known that only haiku should do that, and the hokku is the only verse that may look like a haiku.
Given that these positions appear to be diametrically opposed we can be forgiven for thinking that one of them must be correct. In fact both are mistaken.
From Hokku to Haiku
Even the most cursory history of haiku will reveal its origins as the hokku, the first verse of a renku sequence. Though Matsuo Basho is often referred to as the father of haiku, the practice of writing, collating and publishing collections of single verses predates him. Elsewhere in Renku Reckoner the article Beginnings and Endings describes the performative functions that the initial verse of a formal renku sequence might be expected to discharge. But such considerations relate to content, to the poet's intention. In terms of prosody, of structure, there is absolutely no difference in the way a hokku and a haiku are constituted.
The fundamental aesthetic quality of both hokku and haiku is that it is internally sufficient, independent of context, and will bear consideration as a complete work. A hokku, a haiku, is a poem in itself. It is 'free-standing', a quality described in Japanese renku theory as tateku.
The cut and the cutting word - kire and kireji
It is beyond the scope of this article to examine Japanese haikai prosody in any depth, but it is necessary to understand two terms if we are to engage meaningfully with the question of the 'haikuness' or otherwise of the internal verses of a renku sequence. The terms are kire and kireji. They describe key features of the structure of a typical hokku or haiku.
The word kire might be given more or less literally as 'cut'; it denotes a pause-which-separates. A kireji is a 'cutting word', a kind of verbal punctuation mark which signals the moment of separation and colours the manner in which the juxtaposed elements are related. Such colouration is possible because there are a number of different cutting words.
The long verse of Japanese haikai is written as a single unbroken line or column comprising three metrical feet of five, seven, and five syllables respectively. Strictly speaking these syllables are in fact mora, Japanese poets using the terms on or haku. But for our present purposes the word syllable will suffice.
Within this 5/7/5 phonic constraint a hokku, or a haiku, typically juxtaposes a pair of phrases or image-sets. The natural point at which to conclude one phrase, and begin another, is at the end of a metrical foot. As a result the phrases are asymmetrical; one will occupy a single phonic unit, whilst the other will spread over the remaining pair. The typical distribution is therefore 5~(7+5) or (5+7)~5, the symbol ~ here indicating the kire, the pause. In such examples the cutting word, the kireji, generally appears as a monosyllabic utterance immediately before the pause.
Clearly all notion of the juxtaposition of a pair of phrases or image-sets is only possible if there is a degree of semantic discontinuity between them. The kire therefore, and its attendant kireji, marks the point at which the sense of a hokku or haiku may be said to 'turn'.
It should be cautioned that there are other ways of structuring a hokku or haiku in Japanese which avoid any discernable pause-and-turn and in which the kireji appears at the end of the closing metrical foot, as the final word. However the bipartite structure outlined above is very common and has served as the model, or the justification, for the tercet typical of English language haikai in which either line one or line two is end-stopped.
In the historic evolution of linked verse the question of precisely which utterances might be considered as emotive verbal punctuation marks - as cutting words - has been subject to a degree of change, and in classical renga one will occasionally encounter one such used in an internal verse. However by Basho's time this debate had essentially been settled and in Shomon haikai-no-renga we may say that no internal verse of a sequence will use a cutting word (as ever with the work of Basho, the perspicacious student will find an exception).
Be that as it may, in the case of contemporary renku the question of kireji is straightforward enough: cutting words are the preserve of the hokku or the haiku. They do not appear elsewhere.
Nagekomi - Throwing in
Whilst the internal verses of a renku sequence do not exhibit the distinctly bi-partite structure involving the juxtaposition of otherwise unrelated image-sets that is typical of many a haiku or hokku it is simply incorrect to assert that no internal verse will employ syntactic disjuncture allied to a degree of turn.
Japanese renku theorists borrow the term nagekomi (throwing-in) from the art of ikebana (flower arranging) to describe the way in which space may be teased out between constituent elements the better to show the whole to good effect. The space in question is generated by semantic and syntactic disjuncture, even to the point of occasionally end-stopping a metrical foot with a sentence ending particle (shoujoshi).
At first sight such a verse looks very much like a haiku or hokku other than for the fact that it does not employ a recognised cutting word. However, on examination, the degree of turn is far less marked - one phrase typically giving the setting in which the action of the other takes place. That such a description might adequately describe many verses purporting to be haiku in the English language is unfortunate. But this has more to do with poor artistry amongst some haiku poets than with deficiencies in renku theory.
It should be noted that nagekomi is not employed in the majority of renku verses, nor would it be expected to feature in a succession of verses other than as a deliberate attempt to create a particularly jagged passage. As is discussed elsewhere in Renku Reckoner in the article A Dynamic Pattern, a renku sequence does not comprise verses of uniform intensity and the technique of 'throwing-in' is most likely to be used to create moments of high impact.
Equivalence or equivocation?
A large part of the confusion surrounding the structure of internal verses in renku springs not in fact from renku at all, but from questions of form in English-language haiku - particularly the question of how to replicate the effect of the Japanese cutting word, the kireji.
Since the turn of the century English-language haiku prosody has been little debated, questions of form having been settled more by inertia than by insight. One such default position is that the, or rather all, Japanese cutting words might be emulated by the use of an em dash, or similar, at the end of line one or two. More maximal yet is the suggestion that a break in syntax, allied to a line ending, is sufficient to qualify as a 'cut'. These highly equivocal standards are then projected onto renku, with the result that the simple use of parataxis in a renku verse is mistaken for a Japanese kire (cut) or, stranger yet, for a kireji (cutting word). Thus we are led to the frankly absurd conclusion that all internal verses of an English language renku sequence must display unitary syntax if they are to avoid being labeled as 'haiku'.
Hiraku and haikurashii
Clearly a break in syntax alone cannot be the final arbiter of what is or is not a cut verse in English-language haikai. It is not the pause which creates juxtaposition, it is the nature and degree of the turn. As we have seen, a verse which uses strong juxtaposition, with a marked break-and-turn, is almost certain to be internally sufficient, to be free-standing. Japanese renku theorists describe such a verse as haikurashii - 'haiku-like'. When used in the context of an internal verse it is not a compliment.
The expressive force of a renku sequence lies more in the space between verses than in the content of any given verse. Therefore the verses of a renku sequence should derive their primary resonance from their position in series, not from their internal dynamics. The hokku is of course the exception for, as the head verse, it has no other verse to which it is subordinate and must generate its own effect. Whilst the hokku therefore is described as tateku - 'free-standing' - all other verses of a renku sequence are hiraku - 'open' or 'plain' verses whose relationship, one to another, is more important than their internal structure.
In English, as in Japanese, a hiraku may indeed on occasion employ both a break in syntax and a degree of semantic turn. The question of how frequently such techniques should be employed and how marked their intensity might be is a matter of artistic judgment. The only rule here is that a renku sequence is better understood as a sinuous whole rather than as a succession of individual, autonomous, verses.