Version classiqueVersion mobile

Letters of Blood and other works in English

Göran Printz-Påhlson

The Words of the Tribe: Primitivism, Reductionism, and Materialism in Modern Poetics

Part One. Linguistic Primitivism in Modernism and Romanticism

Texte intégral


1In a little-known lecture of 1942, ‘Poetry as Primitive Language,’ John Crowe Ransom presents an important program for poetry and poetics. The context of Ransom’s remarks—the Avery Hopwood Lecture at the University of Michigan—made it necessary for him to express himself in quite simple terms, as he was talking mainly to undergraduates. It had been just one year after that seminal book of momentous title, The New Criticism, appeared, and Ransom set out in his lecture some of the fundamental tenets of the New Critical doctrine, but did so in much plainer language than in his more ambitious essays.

2The definition of primitivism offered is not a very helpful one from our perspective, but still of importance for the understanding of a complete body of thought which has had a profound, and often misunderstood, influence on modern poetics. He writes:

By primitivism I mean an antique or outmoded cast of thought, so that the poetry is likely to seem heroic as compared with contemporary thought, or to seem pastoral, agrarian, medieval, Pre-Raphaelitish, or merely old-fashioned and quaint. After some progress of civilization comes a movement of regress, with poets in charge of it. But I have generally laboured this point in large or philosophical terms, with the result that I seemed to myself profound, but not very pointed, and academically correct, but as a student of poetry, not really close to the topic. (74)

3He goes on, adding more disclaimers: ‘Today, in your honour, I will talk about the primitive quality that appears in poetry as language. This version of critical theory is brand-new for me, and experimental, since I have not worked it out, but it seems more streamlined and presentable than any other I have hit upon’ (74).

4Ransom’s definition of primitivism as an ‘antique or outmoded cast of thought’ leaves a lot to be desired in clarity, and has obviously not benefited from comparison with, say, George Boas and Arthur O. Lovejoy’s fundamental distinctions in their 1935 study Primitivism and Related Ideas in Antiquity; nor does Ransom seem to take into consideration the quite different insights from Yvor Winters’s Primitivism and Decadence of 1937. Boas and Lovejoy present a powerful definition of classical primitivism as an enormously influential body of thought, concerned with the historiographical evaluation of primeval times, which are presented as normative for the present-day actions of children and animals, as well as for the pastoral pursuits of innocence, and as forceful models of morality. While the authors do not attempt any definition of ‘linguistic primitivism,’ their conceptual framework is clearly much more precise than that of Ransom.

5Ransom’s attempt at determining the area for the criticism of poetry is as little successful: ‘literary criticism is not identical with philosophy at large, but it occurs to me that it may well be identical with linguistic [sic]’ (74). In the ensuing argument, Ransom defines a primitive language as:

[…] one whose standard discourse, in trying to be conceptual (or rational), is obliged also, and whether or not, to be imaginal (or substantival). That is, in trying to make useful formulations about things, relating them by virtue of some common or class property, it is obliged to refer to the many-propertied or substantial things themselves, the things as wholes. Primitive languages are sometimes called radical languages: they consist almost wholly in root words, each one denoting a whole thing or whole event. In discourse these roots are jumbled together, and it devolves upon the hearer to figure out the properties in which the things named are related, and by elimination to read into the jumble a consecutive argument. Here is the famous ambiguity of language. You still have it in poetic metaphor, for example, and in all unskilful speech. (75)

6This seems to me—and, I think, to any reader with some elementary knowledge of modern linguistics—to be a hopelessly inadequate account of views on so-called primitive languages, an account already outmoded by the time of Ransom’s lecture. After Sapir and Whorf on the complications of temporal and other systems in Amerindian languages; after Lévi-Strauss (whether we agree with his methodology or not) and his investigations of the abstract content of myth; after Franz Boas on Kvakiutl metaphor and Bernhard Karlgren on primitive Chinese phonology, it seems impossible to maintain such views at all. Even Wilhelm von Humboldt, writing early in the 19th century, had reached a much more sophisticated standpoint in his observations on the Kawi language.

7The idea of a radical language, described by Ransom as consisting ‘wholly in root words’ (75), is a fiction going back to the very origins of comparative linguistics. In this form, the idea seems parallel to the Pound/Fenollosa hypothesis that, one must insist, addresses itself only to the written character in Chinese. It was no doubt of paramount importance for the development of the doctrine of imagism and, as such, conducive to the formation of an implicit theory of language of high modernism (of which more later).

8In intimating that poetry in some way is to be, if not identified with, at least derivable from, such a radical primitive language, Ransom has stated one of the most basic tenets of the American New Criticism, while at the same time also revealing the fundamental ambivalence of the position. When relating poetry to the shortcomings of primitive languages—ambiguity, for instance—or all ‘unskilful speech,’ what is implied is a firm distinction between, on one hand, a language that is artless and illogical, and somehow poetic in a pristine way and, on the other hand, a perfectly logical natural language which successively gets rid of its ambiguities. This is a distinction Ransom took over from his natural enemies, the logical positivists and the general semanticists, and also a distinction that is recognised by virtually no one today: the basic ambiguities and vagueness of natural languages being fully accepted as ineluctable, and even to some degree extending into the language of mathematics and logic. Furthermore, Ransom seems to be quite unsure of his own pragmatic principles when referring to ‘useful formulations’ or ‘unskilful speech’ without clarifying to himself or to his audience to whom or for what purpose these formulations are useful, or which skills are required. Such naiveté is indicative of how the later Wittgenstein’s arguments on linguistic use (of course not available to Ransom) have been incorporated into our thinking. But to some extent it is also a question of political change, as we see when Ransom exemplifies his primitive language with joking quotes in the homemade pidgin English of an imagined American Indian chief: ‘Heap big Indian hunting go, heap big paleface firewater come’ (ibid., 75–76). He contrasts this with the language the chief may have used had he benefited from a college composition course, in which case he would have expressed himself in detailed or pedantic officialese. Ransom very cleverly undercuts the supposed superiority of college-level English, saying ‘linguistic precision illuminates the values offered in a bargain’ (ibid., 76). In the decades since Ransom’s lecture, history has illuminated such values in a very different way, so that we now would be more inclined to contrast, let us say, a speech by one of the Iroquois orators with a Watergate tape, and not find ourselves convinced of the latter’s superiority in logic (although we would have no doubt about the latter’s superiority in duplicity). But again, duplicity of intention is not to be identified with excellent logic. Here, events have overtaken primitivism and left it high and dry in the well-intentioned condescension of colonialism.

9It was not my intention to conduct a belated polemic against Ransom, whose work I admire, and whose strengths and shortcomings no less a poet than Geoffrey Hill has already examined with exemplary judiciousness. I have chosen Ransom’s lecture as a suitable point of departure for this discussion not because it is a good example of his critical ability, but because it shows with great clarity one of the inescapable presuppositions behind the multifarious body of thinking we designate as ‘the New Criticism.’

10There is no doubt in my mind that the sharply dualist theory of language originating with I.A. Richards and permeating New Criticism is not only linguistically and philosophically outdated and lacking in empirical evidence, but also wholly inadequate in coping with even the most elementary problems of poetics. Concepts like primitivism and pseudo-reference (in Winters), primitivism and ontology (in Ransom), tension (in Tate), paradox (in Cleanth Brooks), and gesture (in Blackmur) all suffer from their secondary nature with regard to logical systems that cannot accommodate the aesthetic principles they embrace, and which have a completely different genealogy. Nonetheless, these concepts have had enormous importance in designating an area of investigation of permanent value for the genealogy of high modernism—indeed diametrically opposed to the locus of logical positivism—an area, situated roughly between technique and philosophy, of which more recent poetical theories of poetics have almost completely lost sight.


11In the strange meeting of Yeats and Pound in the winter of 1912–1913, the primitivist strain in imagism comes to the fore. The elder and more famous poet was at the receiving end of the younger poet’s criticisms and benefited from them. Indeed, he even seems to have relished them. Yeats tells us about the meeting in a letter to Lady Gregory (3 January 1913), a letter that begins with the very down-to-earth experience of persistent diarrhoea:

My digestion has got rather queer again—a result I think of sitting up late with Ezra and Sturge Moore and some light wine while the talk ran. However the criticism I have got from them has given me new life and I have made that Tara poem a new thing and am writing with a new confidence having got Milton off my back. Ezra is the best critic of the two. He is full of the middle ages and helps me to get back to the definite and the concrete away from modern abstractions. To talk over a poem with him is like getting you to put a sentence in dialect. All becomes clear and natural. Yet in his work he is always uncertain, often very bad though very interesting sometimes. He spoils himself by too many experiments and has more sound principles than taste. (167)

12It is quite clear that Yeats was delighted to get led away from ‘modern abstractions’ to a more authentic experience when working with ancient Irish material for the poem ‘The Two Kings’ (the Tara poem) and that he appreciated Pound mainly as a medievalist, not as a modern poet. When Pound expatiates on his principles two years later in an oft-quoted letter to Harriet Monroe, he stresses a slightly different conception of linguistic primitivism, although the ‘fear of abstraction’ is never far away:

Objectivity and again objectivity, and expression: no hindside-beforeness, no straddled adjectives (as ‘addled mosses dank’), no Tennysonian-ness of speech; nothing—that you couldn’t, in some circumstance, in the stress of some emotion, actually say. Every literaryism, every bookword, fritters away a scrap of the reader’s patience, a scrap of his sense of your sincerity. (49)

13This strain of primitivism is distinctive from, although no doubt related to, the primitivism of root languages, which we may be permitted to call radical linguistic primitivism. What we designate as vernacular linguistic primitivism (‘put a sentence into dialect’ in Yeats’s neat formula) is in its Poundian version hardly distinguishable from a much older and presumably more powerful branch of linguistic primitivism that puts the concrete image before the abstractions of everyday speech, as Hegel noticed in the lecture Wer denkt abstrakt? (Who Thinks Abstractly?).

14This imagist linguistic primitivism is of course, qua primitivism, dependent on a theory of language that seeks the origin of language—in some Cratylus-like fashion—in concrete relations between words and things. There are a number of such theories whose internal relations seem far from obvious.


15This preliminary mapping of variants of linguistic primitivism has rapidly made the whole picture much more complicated, and it seems imperative that we go back much further in time in order to trace the genealogies of these ideas in modernist poetics. Let us first, however, go back only one step in time, to consider briefly a poet who has been regarded as a harbinger of modernism, but whose poetic ideas clearly had a great deal of independence from mainstream modernism, and who has more recently suffered a noticeable decline in his reputation. I mean Gerard Manley Hopkins.

16A decisive moment in the decline of Hopkins’s reputation comes with Donald Davie’s early essay from 1953’s The Purity of Diction in English Poetry, ‘Hopkins as a Decadent Critic.’ Although ostensibly concerned with Hopkins’s criticism, this powerful essay is ultimately addressed to the decadence of his poetry. But even if we consult an earlier Kenyon Critics’ symposium on Hopkins, which strikes a much more laudatory note (it contains, for instance, the eulogy by Robert Lowell), we shall still find some of the same unease about Hopkins’s linguistic primitivism, notably in essays by Austin Warren and Arthur Mizener.

17It is easy to make a case for Hopkins’s linguistic primitivism, finding it in his preoccupations with techniques from the distant past, with kenningar (the old Norse tradition of circumlocution) and alliteration, with Welsh cynghanned (harmonic sound arrangement), and even with Duns Scotus’s haecceitas (quiddity or particularity). All are all tangible archaising devices and can be seen as very much in the tradition of primitivism, according to Ransom’s description of ‘an antique or outmoded cast of thought.’ It is also easy to see that such archaising elements of Hopkins’s poetic method are in no way foreign to his contemporaries, although the particular uses of these devices in his poetry were anomalous enough to make the poetry unacceptable to them.

18The matter of primitivism constitutes a problem touching not only on Hopkins criticism but also on the reception theory of modernism as a whole. The question to be asked is evidently one of priority: did Hopkins come to his position in philosophical poetics, his ‘inscape’ theory, through his metrical and linguistic experiments, or did he invent or re-invent his elaborate system of diction and sprung rhythms in order to create an instrument capable of expressing his indomitable desire for the concreteness of the sensual and particular? Harold Whitehall, the metrical expert among the Kenyon Critics, has no doubts on the issue:

Sprung rhythm, the overstressing devices and a distinctive, if obscure, vocabulary are the interlocking segments of the Hopkins problem. To write sprung rhythm, he was obliged to use alliteration, internal rhyme, and assonance and word repetition. To use these devices, he needed new compounds and syntactic shortcuts. In nothing more metaphysical than this does his breaking down of the barriers of language consist. […] His verbal innovations exist merely to assure the precise ordering of the musical elements of the line. (354)

19Anyone who has read the notebooks of Hopkins, where the obsessive descriptions of particulars are present from the very beginning in simple unaffected prose, may very well be sceptical about such a view. After all, it seems a fitting example of linguistic reductionism—a most peculiar result of the New Critical ideology, in view of its holism in philosophy. The problem, however, remains. I am not proposing a solution to it, as it seems to be a variant of the familiar chicken and egg conundrum, and thus insoluble. But let me remind you of the famous quip attributed to Samuel Butler about evolutionary mechanics, that a chicken is just the means available for an egg to make another egg. This seems as good a description as any of the evolutionary mechanisms in some theories of literary genealogy, where the active creative will or intention of the artist is similarly reduced.

20In order to avoid reductionism one is forced to consider the whole gamut of canon formation. It seems inevitable that the discussion of Hopkins’s innovation in language and diction must be related to his views on the poetical practices of his contemporaries and immediate forerunners, as much as to the technical devices he found worthy of imitation in ancient poetry.

21This is, as could be expected, very much the starting point of Donald Davie’s essay. He quotes a letter from the twenty year-old Hopkins to Baillie (10–11 September 1864), where the author outlines a fairly elaborate typology of poetic styles, of which ‘the first and highest is poetry proper, the language of inspiration’ (154). This is the voice of the greatest poetry, and Hopkins has very little to say about it. He has more to say about the second type, which he calls Parnassian language, and which he regards as of dubious worth. He exemplifies it with Tennyson, and it seems that doubts about Tennyson had been the first incitement for Hopkins to construct his taxonomical scheme. There is also a sub-species of the Parnassian that he calls Castalian, and identifies with Wordsworth in his less inspirational moments. The third category is the most interesting from our present point of view of linguistic primitivism:

The third kind is merely the language of verse as distinct from that of prose, Delphic, the tongue of the Sacred Plain, I may call it, used in common by poet and poetaster. Poetry when spoken is spoken in it, but to speak it is not necessarily to speak poetry. (Ibid. 158) [Second emphasis mine]

22Delphic poetry is not exemplified. It seems to me that Hopkins, in his description of the Delphic style, anticipates his much later assertion that poetry ‘should be current language heightened, to any degree heightened and unlike itself, but not […] an obsolete one’ (letter to Bridges, 14 August 1876) (xxxiii). This indeed has a striking resemblance to his last sub-species of the Delphic, the Olympian, which consists of ‘the language of strange masculine genius which suddenly, as it were, forces its way into the domain of poetry, without naturally having a right there’ (ibid. xxxiii).

23The idea of the ‘tongue of the Sacred Plain,’ which is not a common language and even less a secular one, but both sacred and plain, should be compared with Ezra Pound’s formula in the letter to Harriet Monroe referred to earlier: ‘a fine language departing in no way from speech save by a heightened intensity’ (49).

24It is necessary to notice that both Hopkins’s and Pound’s positions are prompted by their opposition to Tennyson, and to a generalised view of poetic diction associated with Victorian poetry: Yeats’s ‘abstractions.’ While Hopkins and Pound are united regarding Tennyson, in another respect they are opposed, as Davie duly points out, and that is in their views of Milton. For Hopkins, Milton is the supreme example of propriety of language, and Hopkins himself wants to achieve a more ‘balanced and Miltonic style’ (235). For Pound, as for Eliot (at least to begin with), Milton is the epitome of idiosyncratic style and the great stumbling block in the development of English poetic diction. How can that be? Given the similarities of their positions on linguistic primitivism it seems hardly possible to believe Hopkins and Pound are speaking of the same poet.


25Pound never wrote a full account of his views on Milton’s poetry, although his comments in ABC of Reading on the Latinate syntax of Milton are very pertinent. Milton’s ‘misdeeds as a poet have been called attention to, as by Mr Ezra Pound, but usually in passing,’ T.S. Eliot writes in his first essay on Milton (from 1936), a highly critical but more circumspect and less vitriolic statement than Pound’s obiter dicta (258). Eliot concentrates his criticism of Milton on his legacy more than on his practice, but in substance his strictures are the same as Pound’s: Milton has created a rhetoric which is ‘not necessarily bad in its influence,’ but ‘bad in relation to the historical life of a language as a whole’ (262). Here Eliot brings in a favourite parallel with Dryden:

Of the two, I still think Dryden’s development the healthier, because it was Dryden who preserved, so far as it was preserved at all, the tradition of conversational language in poetry: and I might add that it seems to me easier to get back to healthy language from Dryden than it is to get back to it from Milton. (262)

26In his second essay on Milton (from 1947), an essay often regarded as a fairly weak recantation of his earlier criticisms, Eliot in fact reiterates and clarifies his charges against Milton, although tempering them with a much wider and more pragmatic perspective. He now maintains ‘the remoteness of Milton’s verse from ordinary speech, his invention of his own poetic language, seems to me one of the marks of his greatness’ (ibid., 268). But just prior to that, the strictures seem to remain:

there are the great poets from whom we can learn negative rules: no poet can teach another to write well, but some great poets can teach others some of the things to avoid. They teach what to avoid, by showing us what great poetry can do without—how bare it can be. (268)

27This is a concept Davie was to develop in a somewhat different direction, in his idea of the purity of English poetical diction. But the norm for Eliot is not a poetic diction, however pure, but a vague notion of something called variously ‘prose,’ ‘ordinary speech,’ ‘common language,’ or the like. Towards the end of his essay, Eliot gives one of his clearest accounts of this version of linguistic primitivism:

I have on several occasions suggested that the important changes in the idiom of English verse which are represented by the names of Dryden and Wordsworth, may be characterized as successful attempts to escape from a poetic idiom, which had ceased to have a relation to contemporary speech. This is the sense of Wordsworth’s Prefaces. (272)

28Indeed it is, and I shall presently follow that trail. But at this point we can safely assign a mode to this kind of primitivism, well known from antiquity in the great inventory of Boas and Lovejoy. But if cyclical primitivism is going to work, it has to be dialectical as well: the historical process has to have enough tension to develop at an even pace. Eliot, philosophically resourceful as always, has some ingenious things to say about this process:

If every generation of poets made it their task to bring diction up to date with the spoken language, poetry would fail in one of its most important obligations. For poetry should help, not only to refine the language of the time, but to prevent it from changing too rapidly: a development of language at too great a speed would be a development in the sense of a progressive deterioration, and that is our danger today. (273)

29These principles, so well-known and so much part of the modernist heritage that we have almost, through familiarity, lost sight of them, have not elicited the same degree of commentary as have other critical concepts central to high modernism in the Pound/Eliot mode, such as dissociation of sensibility, or individual talent, or the objective correlative, or melopoeia. This may be because of the vagueness of the model invoked, and the reasonable nature of its implied historicism. Clearly, the conversational model for poetic language cannot be used in any mechanical or even methodical way as a critical tool for censuring poetic diction: it is historically determined as part of the practitioner’s art and limited to specific situations when poetic diction is deemed to have deviated too much from a norm which is intuitively or subjectively (in fact poetically) determined. In other historical situations, the opposite principle can be invoked and a formal poetic diction may have to be cultivated and codified.

30Pound, although subscribing to a similar kind of primitivism, is much less systematic and relativist in his remarks, and more assertive in relating the deviation from a natural conversational norm to specific linguistic causes. In ABC of Reading he blames Milton’s unnatural syntax on the inflectional model of Latin: ‘the great break in European literary history is the change over from inflected to uninflected language. And a great deal of critical nonsense has been written by people who did not realize the difference’ (50).

31There is a passage of Ernest Fenollosa’s The Chinese Written Character as a Medium for Poetry where the author gives his radically primitivist view of Aryan etymology, concluding that ‘Nature has no grammar.’ In his footnote to the passage, Ezra Pound tells us:

Even Latin, living Latin, had not the network of rules they foist upon unfortunate schoolchildren. These are borrowed sometimes from Greek grammarians, even as I have seen English grammars borrowing oblique cases from Latin grammars. Sometimes they sprang from the grammatising or categorising passion of pedants. Living Latin had only the feel of the cases: the ablative or dative emotion. (50, n.1)

32There are obvious grounds for confusion here. It is true that the formal syntax of the classical languages has played an enormous role in forming the poetic diction (and other areas of formalised language) of later times: we can quote the monumental works of Eduard Norden and Einar Löfstedt to this effect. But there is no firm evidence that all cases of syntactical license in poetry are to be regarded as calqué on classical models. Moreover, although we can perhaps assume that colloquial Latin was less rigorously free in its word order than the speeches in Livy or Tacitus would make us believe, this does not imply that we can reduce the rules of grammar for that language to ‘feel’ or ‘emotion’ (whatever that might mean). Pound here has confused the borrowing of grammatical terms with the actual internalised rules of accidence or syntax employed by the competent speaker.

33This situation is today much clarified by recent investigations of case grammar and syntactical typology. Had he had more than a hearsay account of even the contemporary views on these matters, for instance in Sapir or Jespersen, Pound might have been able to avoid this confusion—or perhaps not. It is now becoming abundantly clear that there is a built-in confusion or contradiction in the idea of linguistic primitivism as a doctrine of high modernism. In radical linguistic primitivism (of which the Fenollosa/Pound position is a particularly salient example), one posits a primordial form of language made up of root words signifying things or acts in a direct way, and elevates this as a model for poetic language. In vernacular linguistic primitivism, one posits a model for poetry in a language that is colloquial, contemporary, and non-archaising. While the first model can lay some claim to absolute validity, the second—if for no other reason because of the fact that colloquial speech changes with the times—has to be relativised and historicised in a cyclical way, in particular at a stage when modernism is in need of gaining some respectability. Eliot and Pound tried to embrace both models in their reckoning with traditional poetic language, which they saw embodied in Miltonic diction in particular. But Pound, being much more of a radical (in this specialised sense if in no other) had to be more violent in his rejection of Milton, as he so easily could have embraced the principle of Latinate syntax, or free word order within metrical constraints, as a form of radical linguistic primitivism. This was of course what Hopkins did, and quite correctly from his premises.


34When trying to assess the importance of linguistic primitivism for the practice of poetry in high modernism and subsequent literary phases (whether we want to call them postmodern or not) it is imperative to keep in mind the neutral and non-committal nature of the vernacular variety of primitivism. As compared to classical variants of primitivism it is not imbued with any great moral purpose. It is not because of any great vision for the future of mankind that the poet should make his language colloquial: it is a mechanical operation very much like increasing or reducing speed when driving a car. When the poet notices that a formal poetic diction is clogging up the traffic, he steps on the accelerator and makes his language more colloquial; in the next curve he may have to reduce speed and let the language congeal to standard formulae.

35Nonetheless, both Pound and Eliot remained on the whole faithful to their principles in their poetic practice. When considering the deletions and emendations prompted by Pound for the most extensive first version of The Waste Land, one may be reminded of Yeats’s comment on Pound’s good principles and deficient taste. Pound evidently had more taste than he got credit for in making Eliot exclude his most abandoned colloquial exercises: ‘He do the police in different voices’ would have been a far from suitable epitaph on Eliot’s poetical talent had he decided to follow a more prolix and polyphiloprogenetive poetic career. As for Pound’s conversational cluster technique in The Cantos, it is no secret that it has had a most insistent influence on American poetry, noticeably even today, via the legacies of objectivism and projective verse.

36The provenance of vernacular linguistic primitivism is obvious, even without Eliot’s helpful hint. Nearly all the ideas and also the confusions can be found in the various versions of Wordsworth’s prefaces. Since the literature on Wordsworth is so enormous, and the ideas so well known, I shall mention just a few important points. When Ezra Pound writes to Harriet Monroe in favour of a poetic language containing ‘nothing that you couldn’t in some circumstance, in the stress of some emotion, actually say (49), he echoes not only Wordsworth’s repeated contention that he had been aiming at ‘a selection of the language really used [or spoken] by men’ (13), but also his insistence on linking their language with ‘the language of extraordinary occasions’ (42) and ‘language exquisitely fitted for the passion’ (24). And when Pound, earlier in the same letter, argues that ‘poetry must be as well written as prose,’ (49) he is likewise indebted to the observation of Wordsworth ‘that some of the most interesting parts of the best poems will be found to be strictly the language of prose when it is well written’ (19). The relativist and historicist bias to be found in Eliot is also often prefigured in Wordsworth’s prefaces, in the rejection of the category of taste, particularly in the appendix of 1802, with its historical account of the origin of poetic diction in a perversion of the daring and figurative language or the earliest poets.

37Scholars and critics have often emphasised this spirited trans-valuation of 18th century values. In a tightly argued short article, Professor Hans Aarsleff of Princeton has given good evidence for considering an influence in general terms from French thinkers, in particular Condillac, but perhaps also Madame de Staël in her 1800 relativist and historicist account of literature in relation to its social setting, De la littérature (On Literature). One tends to forget that the notion of the primacy of figurative language and its derivation from the passions is an 18th century commonplace, not only to be found in its most eloquent champion, Giambattista Vico, but also in Rousseau, Herder, Hamann, and Monboddo.

38Still, the Wordsworthian version has a distinctive flavour of its own in the history of linguistic primitivisms. In one important aspect it differs completely from the otherwise strikingly similar views of Eliot or Pound: it is firmly committed to a social perspective most assuredly stated in its earliest form in the Advertisement of 1798, where Wordsworth says ‘the majority of the following poems are to be considered as experiments. They were written chiefly with a view to ascertain how far the language of conversation in the middle and lower classes of society is adapted to the purpose of poetic pleasure’ (1).

39In choosing a social class not his own, observable and close at hand certainly, but not relegated to a distant time or land, the poet has to assume the role of spokesman or translator, as Wordsworth puts it. We would perhaps today say documentary journalist or anthropologist. The reportorial elements in Wordsworth’s poetry may not have been widely noticed, but are in fact both considerable and theoretically essential. More important still, it is from his respect for the social permanence of rustic life that his poetry gathers its strength—a strength not matched in Eliot or Pound, for all their reliance on craftsmanship and for all their considerable (if sometimes confused) social speculation:

Low and rustic life was generally chosen, because in that condition the essential passions of the heart find a better soil in which they can attain their maturity, are less under restraint, and speak a plainer and more emphatic language; because in that condition of life our elementary feelings co-exist in a state, and consequently, may be more accurately contemplated, and more forcibly communicated; because the manners of rural life germinate from those elementary feelings; and from the necessary character of rural occupations, are more easily comprehended; and are more durable; and lastly, because in that condition the passions of men are incorporated with the beautiful and permanent forms of nature. (14)

40This is a powerful statement of mainstream primitivism in social terms with a considerable explanatory force regarding the origin of poetic thought. For beyond Wordsworth’s primitivism, and imbuing it with moral purpose, is the whole tradition of the sermo humilis: the humble and earth-like language derived ultimately from the Synoptic Gospels and impregnating so much of medieval literature. Erich Auerbach observed this pristine form of linguistic primitivism in his magisterial Mimesis, and also wrote the classical essay on sermo humilis, which takes as its point of departure the levels of style recognised by St Augustine in his examination of pagan rhetoric. But even beyond that we can listen to the ironic voice of Socrates in Plato’s Phaedrus, castigating the art of writing and rebuking the speech-writers for their rhetoric: ‘In fact, the people in those days, lacking the wisdom of you young people, were content in their simplicity to listen to trees or rocks, provided these told the truth’ (157).

41This tradition of linguistic primitivism is one step ahead of other more limited versions, as it does not, in its humbleness or simplicity, invest much trust in writing as an activity, or regard it as a necessity for poetic creation. Thus it is not very highly regarded today. I identify it with Hopkins’s despised Delphic of the Sacred Plain: ‘Poetry when spoken is spoken in it, but to speak it is not necessarily to speak poetry’ (158).


42Linguistic primitivism has put its distinctive mark on modernist poetry in the whole Anglo-American tradition. When we look at modernism in a wider European setting, we find very little that corresponds to it. The voices of modernist poetry in French or German poetry, from Hölderlin to Celan, Baudelaire to Bonnefoy, speak defiantly in another dialect, lofty, vatic, solemn, sermo sublimis, rather than sermo humilis. Paul de Man has, in a masterful and pregnant early essay, traced the primary vocabulary of that dialect to its sources in figurative language of great simplicity, ‘The Intentional Image in Romantic Poetry.’ In this essay he reminds us that whatever the demands of the language, of the social world or the sensual world, or even the ontological primacy of the natural object, words are used in poetry not as signs or names, but in order to name, in an Adamic act. He quotes the words of Mallarmé (from his epitaph on Poe) ‘donner un sens plus pur aux mots de la tribu’ (70). Is this to be taken, with the English poet, as an act of social acquiescence in the face of the intractability of language, or with the continental poet, as a defiance of social demands in order to reach the silences behind the words? Those are the questions that will be present for us in further investigations of these problems in this series of essays. Let me conclude here with a quotation from Wordsworth’s third ‘Essay Upon Epitaphs,’ which marvellously comprises both views in two sentences: ‘Words are too awful an instrument for good and evil to be trifled with’ (129) and ‘language, if it do not uphold, and feed, and leave in quiet, like the power of gravitation or the air we breathe, is a counter-spirit, unremittingly, and noiselessly at work to derange, to subvert, to lay waste, to vitiate and to dissolve’ (ibid.).


Works Cited

Eliot, T.S. Selected Prose of T.S. Eliot. Ed. by Frank Kermode. New York: Harvest, 1975.

Fenellosa, Ernest, and Ezra Pound. The Chinese Written Character as a Medium for Poetry, ed. by Huan Saussy, Jonathan Stalling, and Lucas Klein. New York: Fordham University Press, 2008.

Hopkins, Gerard Manley Poems and Prose. Ed. by W.H. Gardner. Harmondsworth: Penguin, 1953.

Mallarmé, Stéphane. Collected Poems and Other Verse. Ed. and trans. by E.H. Blackmore and A.M. Blackmore. Oxford: Oxford University Press, 2006.

Plato. Plato’s Phaedrus. Trans. by R. Hackworth. Cambridge: Cambridge University Press, 1972.

Pound, Ezra. ABC of Reading. New York: New Directions, 1960.

The Letters of Ezra Pound, 1907–1941. London: Faber and Faber, 1951.

Ransom, John Crowe. ‘Poetry as Primitive Language,’ in Speaking of Writing: Selected Hopgood Lectures. Ed. by Nicholas Delbanco. Ann Arbor: University of Michigan Press, 1990: 73–83.

Whitehall, Harold. ‘Sprung Rhythm,’ in The Kenyon Review. 6.3 (1944): 333–54.

Wordsworth, William. Wordsworth’s Literary Criticism. Ed. by Nowell C. Smith. London: Frowde, 1905.

Yeats, W.B. ‘Letter to Lady Gregory,’ in W.B. Yeats, Man and Poet. Ed. By A.N. Jeffares. New Haven: Yale University Press, 1949: 167.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search