The phrase jo-ha-kyu describes an ideal dynamic pattern which may be realised in any martial or performance art, the latter category embracing not just music and drama, but also flower arranging, the tea ceremony and linked verse composition. As the name suggests the jo-ha-kyu pattern is essentially tripartite: in its widest sense jo may be given as introduction; ha as scattering; and kyu as finale.

Originating as gagaku, a form of 7th century Chinese court music, the concept of jo-ha-kyu was popularised in Japan by the great Noh master Zeanni who advanced it as the key pacing principle for theatre. Though poets subsequently discussed its potential in respect of linked verse such talk was largely speculative until the Edo period and the advent of the Basho school. In part the impetus which drove Basho to favour the shorter 36 verse Kasen over the traditional 100 verse Hyakuin was the appreciation that jo-ha-kyu might be most effectively employed in a more compact form of sequence.

With the establishment of the Shomon Kasen as the standard vehicle for haikai linked verse the notion of jo-ha-kyu became a core element of renku theory. The remarks which follow treat the dynamic pattern in this more narrow context. The reader may find that the Schematic Guides and the article Common Types of Renku Sequence provide useful supporting information.

The initial movement of a poem is called jo: the preface or introduction. Tonally the preface is expected to be relatively restrained. It is polite and pleasant, perhaps exploratory, never challenging. Such humour as is expressed will be light rather than crude, possibly wry but never sardonic.

There are topical restrictions which further the tonal constraints. Heavy or unfortunate subjects are avoided. In the Edo period this would have included such topics as death, war, illness, impermanence, religion, sex and long distance travel, an indicator of how grueling such a prospect could be. 'Foreign', meaning 'Chinese', reference was also frowned on, as were all person and place names.

These latter injunctions are particularly interesting. The bar on all things Chinese is at first sight an assertion of cultural identity in the face of a looming neighbour, such sentiments as are often encountered in Wales with regard to England, or Canada with regard to the United States. There is some truth in this; the history of Japanese literature is often one of how, where and when elements of Chinese design and diction might most appropriately be used. But to concentrate on this aspect is to miss the larger point, as with the bar on proper nouns, the intention is to limit both extra-textual direction and excessive or arcane erudition.

A reference to a person or a place opens up a broader entelechy than that which is desirable in the preface. For the moment the intention is to keep the readers' experience within the scope of the sentiments and associations directly controlled by the poet. Tu Fu, James Dean, Stonehenge and Area 51 are far too volatile. They risk eliciting interpretations and responses that are too individualised. This kind of emotional probing of fellow author and reader alike is left until later in the poem.

The question of discouraging intellectual flummery is less apparent until one learns that for centuries much linkage in courtly renga had relied on a forensic knowledge of a limited number of core texts such as the imperial poetry collections and the Tale of Genji. Links and references often relied on narrative association, the reworking of some famous passage, or otherwise demanded intimate familiarity with non-essential minutiae. Lexical links were also common, making great play of cognates and homophones.

Renga poets had already decided that these techniques should be toned down in the opening movement of a poem, a practice reinforced by the Basho school who were keen to avoid superficiality at all times. Such considerations hold true, whatever the linguistic or temporal context of composition. The avoidance of proper nouns and foreign diction helps to ensure that the preface is a relatively quiet place where poets may establish their presence without feeling the need to wave icons and bray.

It should be emphasised that the head verse, the hokku is in fact exempt from all such tonal and topical constraint. However, given the expectations surrounding the rest of the movement and the historic practice of using the hokku to extend a coded greeting or augury, it is hardly surprising that very harsh treatments are rare in the first stanza. Renku does not avoid difficult or unpleasant topics, indeed high quality renku seeks to reflect all aspects of human experience, but the less tractable subjects are dealt with in ha, not in jo.

The second movement of a poem is ha: the development or intensification. In the Kasen, and quadripartite sequences derived from it such as the Nijuin, ha is split between folios into two equal halves.

Where the preface - jo - is restrained and circumscribed, in ha just about anything goes. It is rare for poets of the Basho school to be extremely vulgar, crassly insulting or gratuitous in their portrayal of violence; nonetheless sex, sarcasm and civil strife find their way into ha as do high religion and low farce. The widespread occidental misapprehension that the haikai arts treat only nice subjects is never more mistaken than here.

This is not to suggest that ha is composed of ever more strident attempts to shock or that it is uniformly blaring. The heavy reliance on lexical and conceptual trickery which characterised earlier linked verse is gone, and whilst the requirement to shift (q.v. Link and Shift) means that narrative or thematic extension is impossible in renku other than between a pair of verses this does not preclude subtle concatenations of tonal and emotional evocation over a more extended range. The twelve verses of a single face of ha in a Kasen for instance will show peaks and troughs of impact mediated by the steady build of tension that some contemporary theorists refer to as the 'renku wave'. The best quality renku is a sinuous beast, not a swarm of gnats.

The extratextual reference eschewed in the preface may now used, if sparingly by comparison to classical renga. Poets will take more risk with the variability of response to cultural emblems; known quantities are recast; common assumptions combed for fresh insight. There is a tendency too to become more heterodox as the movement progresses, in terms of both content and prosody. The latter portion of ha in some Basho Kasen contain stanzas so structurally exceptional that later editors assumed them to be transcription errors and thought it best to 'correct' them. In contemporary Japanese renku likewise some poets abandon conventional metre in favour of a vers libre approach to a limited number of stanzas.

In short, the development movement is the place for experiment and innovation. But only the crudest writing seeks diversity at all cost, or mistakes the loud for the profound.

Master Zeanni reputedly likened the jo-ha-kyu dynamic to the course of a mountain river: jo is the tributary's gentle rill; ha the powerful cutting back and forth between peaks of the river in spate; and kyu the plunge of a mighty waterfall into a deep and silent pool.

The metaphor is important, for though jo and ha have been reasonably well understood in occidental renku the closing movement - kyu - has tended to be taken as simply a mirror of jo. As a result the 6/12/12/6 verses of a Kasen would be viewed as quiet/loud/loud/quiet, or restrained/open/open/restrained - an interpretation sustained by more than one contemporary Japanese renku theorist.

There is nothing dreadfully wrong with this proposal, but it does not accord with the evidence from Basho's own writing and tends to yield sequences that fizzle out. The 6/12/12/6 of the Kasen are better understood as quiet/varied/intense/rush-and-stop. Kyu is rush-and-stop or, as it is more commonly given, rapid close.

Crudely put we might expect the first three of the closing six verses to be striking, whilst the final trio achieve a happy and tranquil state, abetted by the fact that the closing pair, and often all three, have spring as their background and are therefore conducive to felicitous sentiments.

Whereas the latter part of the development, ha, encourages diversity of content and style the early stage of kyu requires compaction and irresistibility. So whilst the verse content might be quite brash the prosody will now be unchallenging, the metres conventional and the inter-verse linkage relatively tight. Extratextual direction will be limited too; the desire is to once more hold the reader directly within the ambit of the poets' intention. We wish the reader to experience the waterfall, rather than fly off in some random direction.

With the initial part of kyu understood as drum roll and cymbal crash it is not difficult to imagine how the few verses which close the poem can more easily generate a peaceful and lingering resonance, the precise colouration of which is provided by the final verse, ageku. Far from being a succession of anodyne verses kyu is the wave that breaks followed by the hiss of foam and the growl of the undertow.