Cohesion, or parataxis and hypotaxis
Eric Mottram stated, a long time ago, that parataxis is central to the new British poetry: "Instead of an item in a school of rhetoric, the poem could have variety of articulations, continuity and discontinuity, sentence and parataxis, and an awareness of the imaginative possibilities of relationship between particles". Parataxis is the way in which clauses are linked when they are co-ordinate; parataxis is the opposite of hypotaxis, which is when one clause is subordinate to another. "The cat sat on the mat. The mat was yellow" is parataxis; "the cat sat on the mat, which was yellow" is hypotaxis. What Eric says (in that programme-essay which launched the term "the British Poetry Revival") must mean that hypotaxis was an important feature of the old poetry.
Basil Bernstein described (in Class, Codes, and Control) two styles of speech (restricted code; elaborated code) associated with two social groups; one of them is associated with academic success, and is also used by the professionals in the education system. The other type is also called implicit code – using conjunctions to articulate logical relationships is just what it does not do. It is clear in B’s description of restricted code speech that hypotaxis is weak and monotonous there; and that hypotaxis is a feature of elaborated code. So the preference for parataxis in a certain segment of modern poetry could be an expression of protest against, or anxiety about, the elaborated code. Bernstein has observed “a preference for things as against processes”; this equates very neatly with WC Williams’s “no ideas but in things”.
Mottram has elsewhere described verbal poetry as linear; in a paper promoting visual and performance poetry. The appeal of these diffuse or allover types of signalling is partly that they get away from elaborated code speech, and so from the anxiety associated with class differences. (We could also phrase this as: audiovisual media are classless and egalitarian.) The linear quality sheds light on what syntax does: in a succession of linguistic moments, each receiving exact but momentary attention, it adds pointers which indicate relations between the transient moments. The reader has often found a paratactic style difficult because the pointers are missing – it is either without highlights or else it is all highlights.
A list of paratactic poets of the 1960s would include Raworth, Asa Benveniste, Edwin Morgan, Ken Smith, John James, Lee Harwood, Jeff Nuttall, Barry MacSweeney. When Tom Raworth writes
a breeze rustled curtains
detached from night
surrounding the stadium
colours ran in all directions
slowly past them
shaded cool and inviting
inward as usual
a segmented scar
on the edge of the divan
(from: Blue Screen)
this style has, to ignore artistic qualities, zero hypotaxis. Indeed, it is hard to see how hypotaxis could be fitted into Raworth's verbal organization. The logical structure of the text is deliberately unspecified; the degree of control (almost, authority) needed to make a situation predictable enough for syntax (with its ordered causal relationship of phenomena, its repetition implied by the precision with which events are ordered in their incidence, its implication that other views are wrong) to be applicable. We seem almost to be distinguishing a formal situation (where behaviour is regulated by a code of some kind, and so is right or wrong) and an informal one (where people are free to react as their instinct tells them). So social framing, the labelling of a situation which implies what behaviour is appropriate in it, may be part of the unconscious structure of a statement; and the style is telltale of an awkwardly formal, tight, situation, or of a free and easy one. Look at this poem by Ulli McCarthy:
foils of silver spires biting flesh
tug O cord clots
over fur long hymns of sunwind curl
my breath whipped on thorns of glass
before with the cattle O eyes milked
before carrying down lamb bursting from mountain
my touch dwells smoke
O bog quakes
say nothing night steep over face
echo spreads through tree alight roots
torn frock mother black no bonnet wear
(from Sonnet Brushes, 1972)
Whatever else one observes about this, hypotaxis is certainly missing. To take a more recent poet, here is one by Kelvin Corcoran:
Why don't you lift the piano lid to play that tune?
it's not a piano, it's a poor metaphor;
see my fingers presentive,
presentive fingers playing a metaphor
like polar bears dive from glaciers to fish
or my foot arch curls with secret warmth;
the final continents beyond analogy
here are the poor, they have number,
the adults with worn hands of too much care.
('The sound at the end of waiting', from The Red and Yellow Book).
This breaks the rules a bit, by a clear hypotaxis signalled by the word like at the juncture of two clauses.
horizontal on rough grass
For an hour or more
Sleepless I dreamt you here across the spaces the
Silent presence of another discourse
Impenetrable as the limitless
Sky the sky's lights in meaningless pattern the
Whole structure swivels on an axis
Inserted in the earth somewhere
In these dark fields nightwind whispers of
It whispers of you
('Lines in Wasdale Head', from History Labour Night, by John Seed)
Again, there is an underlying rational argument here, and one would like to argue that the participles in apposition are latent hypotaxis, achieving half of the status of subordinate clauses. "(A)s" is a subordinating conjunction. It occurs to me that the evolution of English over the last thousand years or so has been towards signalling syntactic relations by word order alone, discarding the markers of concord in verbs, participles, and relative pronouns; so that promoting juncture to the dominant device of verse construction (which seems to be the Raworth style) is an over-fulfilment of tendencies active in the common language.
The origins of this parataxis in poetry are several. Let's list:
1. the Surrealist method of juxtaposing two disparate images which could not be linked by any logic.
2. the montage of different pieces of film as part of the stock repertoire of television. mechanical techniques of acquiring data allow and encourage mechanical splicing as a way of forming meanings. But splicing is always paratactic.
2.1 the dominance of photography in general, so that poets were imagining static pictures, full of data but very poor on analytical or logical relationships.
3. the theories of Charles Olson, going back to Pound's Cantos and ultimately to Imagism and the model of Japanese and Chinese poetry.
4. the cultural dominance of pop songs, where the need to be continuously present acts to break down any complex relationships between lines; each line has to work on its own. At most narrative is possible, not reasoning.
So much for the history of art. But I am afraid there are many other reasons. For example, if activity is not motivated by obedience to rational codes, it might depend on mood and emotion (which is "irrational" in the conventional conception). So, banally enough, dismantling hypotaxis might be a vindication of subjectivity, and an encouragement to the reader to be subjective. The stepping out of classification, as a way of associating phenomena, might be making way for subjective ("free") association, motivated perhaps by the unconscious.
I believe parataxis was just a feature of a global opposition between the new poetry of the 60s, with its “continuous present”, and the over-syntactic poetry of the 50s, identified (now) as tediously moralistic, drab, and smug. It seems that, in the 50s, poets were excessively concerned with competition to be the most moral. The certainty in their poems was a copy of the certainty of Scripture: the authority of the priest was being taken over by the secular writer, as the authority of the priesthood declined. Dialectically, this meant that the distrust which the Church had attracted would be directed against the poets and their opinions –an inevitable result which the Cold War heads of household apparently hadn’t foreseen. Because they had to win the competition, the circumstances of the race had to be unambiguous – so that the situations in the poems had to be made unambiguous, and moral judgment made simple. This much damaged the artistic value of the poems. This rigidity made long-distance arguments tenable – for example that a certain style was immoral because it inevitably led to laxity of personal behaviour which inevitably led to acts which inevitably had bad results. The goal of parataxis was simply to break these rigid chains of reasoning and restore uncertainty. It seems a lot of 50s poets were content to write poems which were boring, in a sanctimonious tone, because they thought the style showed them to be good people. The belief that a “moral” verbal style shows you to be a moral person (and is, even, a moral act) is one of the long-term prejudices in English poetry; and it just isn’t so. The rigidity may have had to do with a very homogeneous audience for poetry; having fixed shared values was at first comforting for people being “upwardly mobile” through education (because it made them feel secure), but later had the reverse effect – it seemed ridiculous to the new cohorts.
To make the chronology clearer, I should point out that the well-known poets who debuted in the 50s were, of course, still publishing in the 80s. In the 1960s, the conventional young poets, who liked authority, didn’t react against the 50s style in poetry. For them it was also the 1960s style. It’s quite easy to find anthologies of the 1960s which include, chiefly or exclusively, staid, moralising, preachy 50s-style poems without parataxis. Has a new paradigm arrived since 1980? I’m not sure. There was a tactical recursion to the values of the 1950s, visible in the books which Blake Morrison and Andrew Motion wrote about The Movement and Philip Larkin – not unaccompanied by a triumphalist denial that the rebellious 60s and 70s had ever happened. There is an amount of bafflement and frustration among the readership (the ex-readership, commonly enough!) of modern poetry, and I feel that the unsignalled nature of these transitions is a contributory factor here.
The reader is invited to take a poem and remove all the conjunctions, so that the relations between clauses are unexpressed. (Another step would be to remove classifying words.) This will shed light on what happens when you take syntax away. I hope it confirms my remarks.
Investigation suggests that classifying words work like subordinating syntax, in expressing causal relations. For example, if I describe a poem as paratactic, saying this implies that being paratactic is diagnostic of something in the poem, that there is a stable set of relations being illuminated by the adjective. To use the classifier implies that there is a structure to be revealed. The structure is made up of causality.
The moral poem of the 1950s had the drive of classifying people into moral and immoral, and making every form of behaviour into telltale signs of these states. This is when style was solemnly seen as a test of morality. The shift of the 1960s was to stop asking this question all the time. The logical relations which it suspended were specifically those dictated by a legalistic programme of crime, detection, and judgment. This was the body of knowledge being revealed – and poetry became paratactic as a result of losing faith in it. Actually, the paratactic revolution was part of the secession from Anglicanism as the grammar within which all poetry was written and read. One process of the 60s was the seduction of university English departments away from regarding Christianity as necessary to poetry – the success of a secular tone. Of course, the personnel who had been writing about poetry and morality in 1959 hadn’t gone away in 1969. They didn’t surrender.
Sorley MacLean uses a prevailing parataxis that has nothing to do with the 1960s. Eòin MacNeill remarks "The Irish metres required each line to give nearly complete sense. In the couplet, the approach to completion of sense had to be nearer still. In the quatrain it had to be perfect."('Introduction' to Duanaire Finn, p. xxiii) This was the rule in classical Gaelic poetry; it is related to oral delivery, where there is a need to be continuously interesting, and to minimize carry-over of sense; it scarcely allows hypotaxis. Song, and sung poetry, are predominantly paratactic - apparently, all song, in all languages. The ways in which songs achieve contrast, symmetry, consequence, etc., repay study. They use different methods from prose. We should be slow, I think, to insist that poetry retreat to song. If poets can argue in verse - so much the better. The poems of Donne, or the soliloquies in Hamlet, involve complex logical structures.
As Mottram suggests, parataxis is only one of the devices available to radical poets, and a counting approach might well be useful in describing the stylistics of a whole poem in terms of co-ordination and subordination; it might differentiate different poets, and different poems by the same poet, but also different passages within a poem. Certainly I think it's no longer legitimate to consider the implicit relationships, weakened syntactic structure, and grammatical simplicity of this sort of poetry without referring to Bernstein's description of "implicit code", with its links, worked out by him since the 1950s, to broader issues of social reproduction, educational practices, and the new middle class.
Reasoning is hypotactic - its propositions are linked by logical
"therefores" even when these are not expressed. And hypotaxis most often expresses causal relationships. But we do not acquire data by reasoning only, we acquire it by perceiving. Causal relations are not on the surface of reality, but somewhere else; they are as it were the syntax of nature. "Therefore" often does not express reasoning, but simply the classification structure of language: this is a rodent, therefore it is a mammal. Frequently it expresses an arbitrary code of social rules; and, correlatively, these often reflect social authority: he is the manager therefore we do the work when he says. The information is not derived from perception because it has come from outside, from a superordinate authority. So actually we need to ask what kind of hypotaxis is in question. Some language may be elaborate because it is describing something invisible and non-consensual, not because it is more accurate.
Parataxis is related to the “continuous present”, a style in 60s poetry which deliberately threw off the weight of experience and schemas. Suppressing logical connections belongs with a multiplanar view of the world where connections are not inevitable. The rejection of past experience and of organised knowledge was in line with the social attitudes of the 1960s. Montage can achieve a thousand effects without making the relations explicit. Replacing old experience with new, multiple connections demanded the extension of the text into the long poem - the great feature of the 70s.
A check of my own writing over the last 10 years shows at least a dozen different interpretations of the motives for returning to parataxis. Ultimately, I think it’s too large-scale an event, and too deeply part of the fabric of the poems concerned, to be nailed down. I think we should adopt the embracing word “juncture” to describe the linking events which are either paratactic or hypotactic; because of the frequency of joins in any poem, the frequency of hypotaxis is a good thing to count as a start of analysing the poem. But the joins are the poem – not only does any page show dozens of joins, but these probably fall into dozens of types. Again, we should consider component structures of different sizes - where poems are split into sentences, stanzas, and larger structures, there is a join between each one (and a different kind of join).
A linguist influenced by Bernstein was Halliday, whose work can be considered as an exploration of the opposition between explicit and implicit codes. He wrote (with Ihab Hassan) a work on cohesion – where we can see cohesion as a category which includes hypotaxis and parataxis. He further worked on linguistic power and expertise. Pupils of his were Günter Kress and Robert Hodge, who produced the book Language as Ideology. All of these deepen the theme of the relationship between social power, social roles, control of knowledge, and linguistic structure. Work on paratactic poetry will continue to draw on this technical analysis to explain poetic choice as a variation on choices within the language as a whole.
The mapping of literary poetry
Among the underground poets of the 60s and 70s, we have to consider a school of “super-syntactic” poetry with Prynne as its pre-eminent exponent. It is inaccurate, then, to speak of parataxis as a feature which is common to all the underground. It would be tempting to think of hypotaxis as a diagnostic feature of the Ferry Press/Grosseteste school, with the influence of Empson and Davie as part of the Cambridge background. If we return to A Various Art, it includes both paratactic and non-paratactic poems. It is arguable that this is a significant stylistic divide, and that the paratactic style (of Ralph Hawkins, David Chaloner, and John James, notably) is quite different to the logical style (of Prynne and Crozier, notably). However, the audience perceive these poets as a group, united by one stylistic and chronological aura. We could rather simply divide modern poetry between what is based on Metaphysical poetry (with argument, paradox, reflexivity, fine distinctions) and what isn’t. This is also the distinction between university taste and everyone else.
We can address orality indirectly, by considering its rival and paredros; literacy, or literature. In history, the rise of private reading, and books, coincides with the rise of privatisation as a social principle. Nobles get rid of their minstrels but start collecting books. Their power becomes less face to face and more abstract and based on bureaucracy and law - bodies of knowledge which overflow into literature. Ministerialis means minstrel at one point and administrator at another.
The clusters model has an obvious flaw when we look, not at 1995, but at the prehistory, say around 1920-40. Look at the poets included in Agenda. Agenda was mapped, in the model I proposed, as conservative + low because of 9 overlaps with Outsiders, a book of low attainment. But it includes also poets such as Pound, Williams, Zukofsky, MacDiarmid - quite certainly the ancestors of the poets we have called ‘intellectual’ and ‘autonomous’, in A Various Art for example. This connection - quite undeniable - gives us a headache: why do the younger poets in Agenda not appear in A Various Art, and why do the poets from AVA never appear in Agenda?
This prehistoric link obliges us to place A Various Art and its congeners in the same half of the conceptual space as Agenda, scoring high on an axis which we must call literary. An affinity emerges therefore between the underground and innovatory work and the old-fashioned work: namely that both are displaced from the colloquial, and draw on ambitious literary structures, a life of ideas - far, too, from simple realism. Just now, I also associated complex syntax with the moralising and academic poetry of the 1950s - perhaps another hint that the highly organised poetry of recent decades has continuities, even if at the unconscious level, with an earlier generation. We could tell the story as follows: a generation of intellectual modernists flourishing between the wars found little favour with the British reading public. A generation of academics, emerging in the 1950s, had great influence on the poetry world, was the shared voice of the educated public, but had misgivings about the modernist line, seen as technically too difficult or else as too strange. A generation of intellectuals emerging in the 1960s, with a certainly large dose of academic culture, reacted against their teachers (the group just mentioned) and labelled modernism as the criterion of success. In the period of the anthologies chosen, the 1950s group were not featuring much in anthologies (although their grip on many aspects of reviewing and academic Eng Lit was still powerful).
This does not really explain Agenda, which was led by a bizarre sensibility. It does explain why the poets I have identified as group I (for example Geoffrey Hill, Peter Levi, Thwaite) do not feature in Iain Sinclair’s pantheon.
We are tempted by the idea of a picture, perhaps a series of moralising prints called the Three Sins of the Literato, drawing a sinister analogy between the excessive authoritarianism, rigidity, and conservatism of the elder generation of educated people and the excessive complexity, obscurity, and experimentalism of a younger generation - straggling for decades through the derelict Utopia of the Counter-Culture. The furious mutual rejection of these groups was (perhaps) a phase of competition which betrayed a shared wish to inherit specific assets - amounting to affinity of blood.
Whereas the spontaneous strand of Conductors has affinities with Pop and performance poetry of a rather low intellectual standard, the reflexive line which also appears has affinities - visible even though slight - with an older literate poetry. There is a vertex where RF Langley, Kelvin Corcoran, and Peter Levi meet. One can imagine Levi being in A Various Art -and in fact he should have been.
If we group together the ‘literary’ anthologies and search for a pattern connecting them, the hypothesis suggests itself that there is a further chronological transition, much as follows.
(born in 1920-40) Christian/ academic/ Cold war/ existentialist – conservative, formal poetry
(born 1940-55) generation of 68 – Marxist and libertarian – experimental poetry – expansive in form
(born 1955-70) merger of Pop and mainstream currents; disdain for experiment or expansiveness; ‘conservative postmodernist’ line combining whimsy with neatness; ‘ludic’ tone
This would suggest a relationship between the poets of A Various Art and The New Poetry whereby the latter structurally replace the former, occupying the same position in the cultural geometry but with distinctive poetic systems.
The succession would be roughly: Christian – Marxist – floating or confused. It’s quite pleasant to think of an English Faculty somewhere where the elder figure wears tweed and writes rhymed poetry in a Christian existentialist manner, a middle-aged lecturer wears lots of denim and writes conceptual poetry reflecting a purely theoretical Marxism, and a 25 year old fellow writes shallow but glittering poems in a ‘conservative-postmodern’ style’ while wearing designer knitwear, and a novel might well represent things in that way. Actually the link between academic life and modern poetry is too multiple and important to be written off in such a light comic way.
Having checked this out by looking briefly at parts of the cohorts concerned, I think the conclusion is that (a) this hypothesis is not wrong (b) however it is a bad fit to the data and it is clearly inferior to the thesis that the cultural field of (say) 1980 on is polymorphous and resists fitting onto any one ‘spine’ yielding a dominant type. We can speak of one model enjoying a monopoly of prestige in the 1950s, but this rapidly collapsed in the 1960s and the modern set-up is highly ‘clusterized’ in a way which most participants probably find benign. In any case, there is no genetic link from A Various Art to The New Poetry. Experimental poetry did not stop or slow down; the ‘post-modern’ type is weak as a norm. The ‘underground’ has its separate publishing and dissemination structures, and these are outstandingly robust. The structural location of the ’Nincompoop’ lot is quite different. It is unreasonable to think of the wave of poets appearing in TNP, published almost exclusively by commercial publishers, and rejected by the underground audience, as having taken the place of the underground. The serial pattern A-B-C is open to too many criticisms.
While I am suggesting that looking at the incidence of subordinate clauses is a way of locating a poem in the cultural field, I have problems grasping the development of this opposition over time. One would expect the paratactic wave of the 1960s to have broken, after a certain time, and for the streams to merge, gradually, giving way to a new landscape in which a new opposition has assumed the leading role - as challenge, crime, and gesture of loyalty. I am not sure this is what has happened. A major development was the long poem of the 1970s – in the 60s, there were few long poems, but in the following decade there were at least 40 significant ones (with Crow and Mercian Hymns leading the way). The long poem replaced the conventional knowledge of the Movement poem with knowledge which was unconventional, and specified inside the poem. This did represent a line of advance for British poetry – followed by a marked fallback, in the 80s, and probably for economic reasons (the reluctance of publishers to risk this kind of work). If it was for aesthetic reasons – well, somehow I’ve never heard anyone mention or explain what these were. Possibly, interest shifted towards montage – which is really an aspect of parataxis.
This dates from about 1998 and is related to the essay comparing many anthologies and looking for groups. It got moved out of various books because of lack of space. It is also related to the essay on 'Legitimacy. intimacy and role detachment' on this website, which has more discussion of Basil Bernstein. This whole line of argument is quite technical but also important to many parts of 'Affluence'.