Version classiqueVersion mobile
OpenEdition Books

Yeats’s Legacies

 | 
Warwick Gould

Essays

How Yeats Learned to Scan1

Hannah Sullivan

Texte intégral

  • 1 Note—Further information may have been gathered since this article was prepared for publication. If (...)
  • 2 Charles O. Hartman’s Free Verse: An Essay on Prosody (Princeton: Princeton University Press, 1980) (...)
  • 3 Joseph Phelan, The Music of Verse: Metrical Experiment in Nineteenth-Century Poetry (Basingstoke: P (...)
  • 4 See, for example, the adjectives employed by d. h. Lawrence in ‘Poetry of the Present’, in Michael (...)

1According to one familiar, much reiterated version of literary history, free verse was invented in pre-war London by Ezra Pound. It then quickly and decisively triumphed over poetry in traditional metres to become a new stylistic orthodoxy.2 In fact, English poets had been writing in versions of free verse for many centuries; by the end of the nineteenth century metrical discomfort and some degree of resistance to inherited iambic forms had become practically the norm. As Joseph Phelan puts it, in one of several recent revisionist accounts of late nineteenth-century metrical practice, the movement towards free verse was ‘halting and interrupted’ rather than teleologically straightforward.3 But the old-fashioned story of stylistic ‘rupture’ is not entirely unfounded, because it was only in the twentieth century that the ‘free verse’ picked up its full range of polemical and ideological associations. Now ‘free’ became prescriptive rather than descriptive. Instead of referring to a phonologically arbitrary principle of lineation, ‘free’ began to denote a particular kind of poetry—flexible, organic, democratic, and hospitable to the rhythms of ordinary speech.4

  • 5 t. s. Eliot, ‘American Literature and the American Language’, 1953, republished in To Criticize the (...)
  • 6 From a search across the Google Books corpus. By 1918, ‘free verse’ is used about seven times more (...)

2It is hard to believe that Google’s N-Gram viewer can be accurate in its determination that no book in English published in 1900 or 1901 contained the term ‘free verse’. But the statistical story that it tells of the term’s rapid upswing in popularity, and eventual cannibalization of the French ‘vers libre’, is remarkable. By 1910, the year nominated by t. s. Eliot as the ‘point de repère’ of modern poetry, the term was in modest use.5 Over the next decade, however, it exploded in popularity, occurring forty times more frequently in books published in 1920 than 1910.6 The French term ‘vers libre’ enjoyed a similarly meteoric explosion in popularity from 1910 until 1918, at which point it seems to decline in favour of the anglicised version. In fact, by 1920, ‘free verse’ was a more commonly used phrase than ‘iambic pentameter’.

  • 7 Ezra Pound, letter to Harriet Monroe, Jan. 1915, in d. d. Paige ed., Selected Letters of Ezra Pound (...)

3Given that Yeats is the only major poet in whose career 1910 or 1914 could be seen as a midway point, we might expect the free-verse explosion to have affected him more than anyone else. By 1912, Tennyson, Swinburne and Hopkins were dead. Thomas Hardy was seventy-two. On the other hand, Eliot and Stevens were still some years away from publishing a book, and Pound only in the infancy of his career. So what if anything happened to Yeats’s metrical practice in the twentieth century? Did he learn to scan differently, or even to stop scanning at all? Yeats certainly wasn’t unaware of the practice or polemic of the free versifiers; during the teens he was a close friend, almost a collaborator, of Pound’s. He was repeatedly exposed to Pound’s tripartite goals for modern poetry: precisely observed images, musical free verse, and natural syntax, ‘nothing that you couldn’t, in some circumstance … actually say’.7 By 1937, when he crafted ‘A General Introduction to My Work’, he even proposed that naturalness of language had been his own contribution to modern English poetry:

Then, and in this English poetry has followed my lead, I tried to make the language of poetry coincide with that of passionate, normal speech. I wanted to write in whatever language comes most naturally when we soliloquise, as I do all day long, upon the events of our own lives or of any life where we can see ourselves for the moment. I sometimes compare myself with the mad old slum women I hear denouncing and remembering; ‘how dare you’, I heard one say of some imaginary suitor, ‘and you without health or a home’. If I spoke my thoughts aloud they might be as angry and as wild. It was a long time before I had made a language to my liking; I began to make it when I discovered some twenty years ago that I must seek, not as Wordsworth thought words in common use, but a powerful and passionate syntax, and a complete coincidence between period and stanza. Because I need a passionate syntax for passionate subject-matter I compel myself to accept those traditional metres that have developed with the language. Ezra Pound, Turner, Lawrence, wrote admirable free verse, I could not. I would lose myself, become joyless like those mad old women. The translators of the Bible, Sir Thomas Browne, certain translators from the Greek when translators still bothered about rhythm, created a form midway between prose and verse that seems natural to impersonal meditation; but all that is personal soon rots; it must be packed in ice or salt. (E&I 521; cf., CW5 212–13)

  • 8 He also cites Yeats’s troubled, but self-critical, remark to Lady Gregory, ‘Ezra… helps me to get b (...)
  • 9 These include the 1937 ‘General Introduction’, the 1936 Oxford Book of Modern Verse, and the poems (...)
  • 10 Warwick Gould, ‘Writing the Life of the Text: The Case of w. b. Yeats’, Miscelánea: A Journal of En (...)

4In fact, the early Yeats used a great deal of blurry and imprecise language. A. Walton Litz has suggested that Pound’s 1913 precepts for Imagism, which advise ‘Don’t use such an abstraction as “dim lands of peace”. It dulls the image’ were written in direct chastisement of Yeats, who had used the word ‘dim’ twenty-six times in The Wanderings of Oisin.8 The actual story of Yeats’s engagement with free verse needs to be untangled from his clear-eyed and self-promoting retrospectives in the 1930s.9 As an endemic sufferer of what Warwick Gould has called ‘textual restlessness’, we should also ask whether he metrically updated his work in post-publication revision.10 One hypothesis might be that Yeats’s poetry became metrically freer—if not entirely free—as his career progressed, and that he used revision to loosen up or make more natural phrasing that strict form had necessitated in his books from the 1890s and 1900s.

  • 11 Thomas Rees, ‘Ezra Pound and the Modernization of w. b. Yeats’, Journal of Modern Literature 4.3 (F (...)
  • 12 Martin J. Duffell, A New History of English Metre (London: Legenda, 2008), 189.
  • 13 Laura O’Connor, Haunted English: The Celtic Fringe, the British Empire, and De-Anglicanization (Bal (...)
  • 14 Peter Howarth, British Poetry in the Age of Modernism (Cambridge: Cambridge University Press, 2005) (...)
  • 15 Thomas Parkinson, w. b. Yeats: The Later Poetry (Berkeley and Los Angeles: University of California (...)
  • 16 Helen Vendler, Our Secret Discipline: Yeats and Lyric Form (Cambridge, MA: Harvard, 2007). The plur (...)
  • 17 Marjorie Perloff, Rhyme and Meaning in the Poetry of Yeats (The Hague: Mouton, 1970), 13.

5At stake in the question ‘did Yeats use or disavow free verse?’ is the more general question of his relationship to modernism. If it is true that ‘Yeats never could absorb the rhythms of Pound’s poetry, and his later poetry shows only a slight growth away from his customary iambic patterns’, then he can’t be easily folded into a modernism whose main technique was vers libre.11 But here we become involved in a vexing hermeneutic circle; it is hard to imagine defining modernism in poetry without talking about Yeats, and it is hard to talk about Yeats in any historical or intellectual context without referring to modernism. The process of clear metrical description ought to be easier, requiring only a weighing of the poems that Yeats wrote (perhaps in their various forms, as evidenced by the Variorum) against the criteria for free verse. In that case, how do we explain the relative lack of critical consensus on Yeats’s use of metre? Many critics write as if he occupied some nebulous middle ground between traditional form and free verse, and remained in the same metrical limbo for his entire career. Martin Duffell’s generally excellent history of English metre says not much more than that Yeats wrote in ‘loose iambics’;12 Laura O’Connor finds that Yeats ‘does not wish to “break the pentameter”, but to shake it up’;13 but Peter Howarth tells us that Yeats ‘disliked’ Pound’s free verse, and Donald Davie that he ‘abjured free verse’ in general.14 Thomas Parkinson suggests that Yeats counted the number of stresses in a line, but not the number of feet, and so never wrote traditional accentual-syllabic verse.15 Other critics seem to avoid making any general pronouncement on Yeats’s use of metre. Helen Vendler doesn’t use the word at all in Our Secret Discipline, although she does talk of a poem’s ‘outer form’ as ‘metrical and stanzaic shape’.16 In 1970, Marjorie Perloff began her book on Yeats and rhyme by regretting that the sound features of the lyric poems remain largely unexplored’.17 Because Yeats’s development runs counter to a meliorist history in which freedom always triumphs over form, it has sometimes been difficult for critics to see that there is a development.

6Curiously, this development ran counter to the general trajectory of English versifying; Yeats’s poems became more metrically constrained and traditional in the period when free verse flourished. This is true both when we compare revised to original versions, and when we compare the new poems from The Tower, say, or Last Poems to the shiftier, more metrically uneasy early work. His very last poems inscribe themselves clearly within an English tradition of accentual-syllabic poetry, where both stresses and syllable numbers are counted. In ‘Under Ben Bulben’, promoting the ‘well made’ over the shapeless, the artisanally crafted over organic form, he employs the trochaic tetrameter catalectic of Renaissance song.

  • 18 w. b. Yeats, ‘Under Ben Bulben’, in Peter Allt and Russell K. Alspach eds., The Variorum Edition of (...)

Irish poets, learn your trade,
Sing whatever is well made,
Scorn the sort now growing up
All out of shape from toe to top … (
VP 639)18

  • 19 Yeats uses the term to describe Adah Menken’s writing in a letter to Olivia Shakespear, 24 July, 19 (...)

7This essay contends that this gradual process of self-chastisement into form was a conscious decision, motivated first by Yeats’s sense of the interrelation of prosody and syntax and, second, by a growing ideological distaste for free verse, ‘the American vice’ (CL InteLex 6072).19

  • 20 w. b. Yeats, Poems (London: T. Fisher Unwin, 1895), vii.

8Yeats’s early poems are formally heterogeneous, taken as a group, but there is also a degree of formal unease or shiftiness within individual poems. Rather than being free in Pound’s sense, as an act of willful choice, they can seem merely or casually free by the strictest standards of English versification. At other times, the metrical grid is followed to the point of woodenness. The dedicatory verse to Poems (1895), To Some I Have Talked with by the Fire, begins rather metronomically, because of the almost complete coincidence of word boundary and metrical foot:20

While I wrought out these fitful Danaan rhymes
My heart would brim with dreams about the times
When we bent down above the fading coals;
And talked of the dark folk, who live in souls
Of passionate men … (
VP 136)

9But within a few lines the poem has swung to the other end of the pendulum, from woodenly over-metrical to unmetrical. ‘Talked’ is still the governing verb:

And of the embattled flaming multitude
Who rise, wing above wing, flame above flame,
And, like a storm, cry the Ineffable Name …

10It is almost impossible to assimilate ‘wing above wing’ and ‘flame above flame’ to the stress pattern of the metre; by including both a third foot and a fifth foot trochee, Yeats has effectively broken his pentameter. The next line then contains an extrametrical syllable and another reversed foot. The same pattern, when a fixed metre is set up only to dissolve without obvious reason, happens in many of these early poems. It is even more noticeable when Yeats writes in the less capacious tetrameter. He is prone to filling out the lines with monosyllables, giving a harsh, jerky effect, and to inserting a caesura exactly halfway. But, a few lines later, he will break the pattern entirely.

I must be gone: there is a grave
Where daffodil and lily wave
And I would please the hapless faun,
Buried under the sleepy ground. (
VP 67)

  • 21 See, for example, George Saintsbury’s ‘Rules of the Foot System’, numbers 19 and 22, prohibiting su (...)

11According to the traditional rules of English prosody, which look for artful variation within a fixed pattern, particularly a varied use of pauses, this is not a very elegant use of iambic tetrameter.21

12Pound’s manifesto for Imagism was a manifesto for freedom in poetry, but it also plays great attention to the ‘craft’ that a poet must acquire before freedom is possible:

  • 22 Ezra Pound, ‘A Few Don’ ts by an Imagiste’, Poetry 1.6 (March 1913), 200–6 (203).

Let the neophyte know assonance and alliteration, rhyme immediate and delayed, simple and polyphonic, as a musician would expect to know harmony and counterpoint and all the minutiae of his craft. No time is too great to give to these matters or to any one of them, even if the artist seldom have need of them.22

  • 23 Susan Grace Galassi, senior curator at the Frick, cited by Catesby Leigh, ‘Could Picasso Draw Bette (...)
  • 24 In a radio interview, 19 November 1933, reprinted by Walter Frisch, Schoenberg and His World (Princ (...)
  • 25 Letter to Edith Shackleton Heald, 10 August 1937. For ‘prosedy’, see CL InteLex 2995, 2998.

13In fact, across the arts, this narrative of craft before innovation, mastery before freedom, is sometimes used to counter the argument that abstract painting, modern music, or free verse are art-less, child’s play. An exhibition of Picasso’s drawings explains that he was, first, an ‘old master manqué’, a master of ‘the conventions of classical representation’.23 Arnold Schoenberg was a famously strict teacher who insisted that musical education ‘must be based on an acquaintance with the works of the masters’, and that the classical tradition should be mastered before beginning to write atonal verse.24 Pound, Eliot, and Joyce all wrote early poems that show much more rigid adherence to traditional versification than their mature works. Yeats is not really in this category. He did not learn how to scan as an unformed, even juvenile poet, and then slowly relax constraint to produce his greatest works: his development runs the opposite way. Even at the end of his life, he called prosody (which he had earlier been prone to misspell as ‘prosedy’) the ‘subject of which I am most ignorant’ if the ‘most certain of my instincts’ (CL InteLex 7037).25 Earlier in his career he had been less confident in his own capacities to manipulate metre without a scholarly understanding of its exigencies, writing in on 10 January 1897 to Robert Bridges:

I too would much like to discuss with you questions of rhythm, for though I work very hard at my rhythm I have but little science on the matter and as a result probably offend often. Without a consistent science it is difficult to distinguish between license and freedom. (CL InteLex 23)

  • 26 In a letter to Edward Dowden, cited by Ronald Schuchard, The Last Minstrels: Yeats and the Revival (...)
  • 27 r. f. Foster, The Apprentice Mage, 1865–1939 (Oxford: Oxford University Press, 1997), 336.

14This anxiety was perhaps fuelled partly by j. b. Yeats’s criticism of his son for writing in ‘bad metres’, a habit which he attributes in an 1884 letter to faulty and self-indulgent declamation: ‘His bad metres arise very much from his composing in a loud voice manipulating of course the quantities to his taste’.26 Roy Foster claims that it was only in 1906 that Yeats felt his lyric capacities ‘were now accomplished enough for his recent poetry to need no alteration’.27 And it is certainly true that he tended to revise out some of the gratuitously unmetrical lines in poems otherwise in fixed metres. For the 1895 Poems, for example, he got rid of the seven-syllable ‘They were of no wordy mood’, and the trochaic ‘In the verse of Attic story’ in ‘The Song of the Happy Shepherd’.

  • 28 John Todhunter, review of The Wanderings of Oisin and Other Poems, The Academy, January–June 1889.
  • 29 Matthew Campbell, ‘Poetry in the Four Nations’, in Richard Cronin, Alison Chapman, and Antony W. Ha (...)

15The majority of Yeats’s early work is not, however, in an entirely familiar traditional form; poems like ‘The Lake Isle of Innisfree’, ‘The White Birds’, or ‘The Meditation of the Old Fisherman’ employ long lines and ternary rhythms, playing with the bounds of the line in English. The Wanderings of Oisin is even more obviously experimental. Between the poem’s first appearance in 1889 and its second appearance in 1895, Yeats also made it more metrically free, sometimes turning lines of ‘well-formed’ iambic tetrameter into halting and bizarrely unmetrical new versions. How can we account for this? The three books are in distinct, and not entirely traditional, metrical styles. One early and complementary review described these as being fitted to three different kinds of subject matter, and as respectively ‘free octosyllabics’, ‘Keatsian decasyllabic couplets’, and ‘quatrains of long-lined dactylic and anapaestic verse’.28 But this is perhaps being over-precise—Matthew Campbell is more accurate when he describes the third part as ‘an accentual verse which challenges the bounds of metre in English’.29 I would go further and say that this sprawling poem, stuffed full of discordant things, is also a good, if accidental, showcase of some of the metrical tendencies or limit point of Victorian versification; iambic accentual-syllabic verse, both tetrameter and pentameter, is on the verge of being washed away into the dolnik, the ballad, and accentual long lines, amplified by some of the alliteration typical of the Old English stress line. Taken as a whole, it indicates that Yeats began as a more experimental prosodist than he ended up.

16The Wanderings of Oisin begins, in the 1895 text, with a thumping, monosyllabic tetrameter, ‘You who are bent, and bald, and blind’, before loosening itself up:

With a heavy heart and a wandering mind,
Have known three centuries, poets sing,
Of dalliance with a demon thing. (
VP 2)

  • 30 Thomas Parkinson, w. b. Yeats: The Later Poetry, 188–89.

17Here, as throughout the poem, we see total line lengths of more than eight syllables, and extra skipping beats inserted between stresses. (It wouldn’t have been impossible for Yeats to write ‘With heavy heart and wandering mind’.) Yeats’s free use of these extra beats is what leads Parkinson to conclude that he never employed accentual-syllabic verse: ‘it seems to me unlikely, however, that he used a foot prosody in view of the fact that his manuscripts give no example of scanning by feet’.30 But it is not the case that all English poetry in traditional forms has a perfect binary oscillation between unstressed and stressed syllables. Nor, I think, is Parkinson’s genetic claim robust: accentual-syllabic poetry is perfectly possible, as oral traditions show, without a lexical practice of marking out and scanning the feet. What is unusual about Yeats’s practice in the poem is not the pattern of any individual line taken by itself, but the lack of formal consistency, the running together of line types that, historically, belonged to different genres.

  • 31 Marina Tarlinskaja, ‘Beyond “Loose Iamb”: The Forms and Themes of the English “Dolnik”’, Poetics To (...)

18Marina Tarlinskaja argued some decades ago that ‘transitional forms’ (‘between syllabotonic and syllabic, between syllabic and accentual, or between accentual and free verse’) exist in every literature, and that, in certain periods, even established forms may ‘lean towards typologically adjacent meters’.31 In English, she finds that all verse forms contain at least some ‘extra’ syllables (traditionally we might called these ‘anapaestic substitutions’), but the frequency with which these occur depends very strongly on the verse genre. Her statistical analysis accordingly describes verse forms as bands within a continuous space, rather than the discrete labels of prosody textbooks. In strict, classical English verse, she finds that fewer than 3% of intervals between stresses are occupied by two syllables. She then identifies two other, major kinds of English verse writing: ‘loose’ or transitional iambics, and a form that is more balanced between ternary and binary beats. Borrowing from Russian terminology she calls this last form the ‘dolnik’. The dolnik is a poem that traditional metrical analysis described as anapaestic or in a ‘ternary’ metre; Yeats’s escapist, flyaway fantasy ‘The White Birds’ would be an example. (Arguably it is a 3-ictic dolnik written out with two lines compressed into one.)

I would that we were, my beloved, white birds on the foam of the sea!
We tire of the flame of the meteor, before it can fade and flee… (
VP 121–22)

  • 32 Derek Attridge, ‘The Case for the English Dolnik; or, How Not to Introduce Prosody’, Poetics Today (...)

19This may seem an unnecessarily cumbersome additional term, but Derek Attridge has argued robustly that: ‘a name that distinguishes this verse form more clearly from accentual-syllabic verse is desirable, in order to signal clearly its identity as a recognizable metrical genre in its own right and one that does not produce uncertainties about whether to divide up the line into iambic or trochaic feet’.32 Tarlinskaja’s own analysis suggests that in ‘loose’ or transitional iambics, of the kind that Yeats is often alleged to have written, the percentage of disyllabic intervals is about 3–12% (her examples are primarily from nineteenth-century poets—Wordsworth, Coleridge, Shelley, Tennyson and Browning), but in the dolnik the percentage is much higher again, from, roughly, 20% to 80%. Between these bounds, however, she finds a ‘frequency gap’: besides a few ‘older folk ballads’ very few poems in English have more than 10% but fewer than 20% of their interictic intervals filled with two syllables. In other words, the apparent freedom enjoyed by poets writing loose iambics and poets writing in dolniks is something of an illusion; either they use about one disyllabic interval every two lines (loose iambics), as a relatively pointed variation from the norm, or they use multiple disyllabic intervals in every line, so that the ear never settles comfortably into an anticipation of two or three.

  • 33 On the frequency gap between ‘still iambic’ and ‘something else’, see Marina Tarlinskaja, ‘Beyond “ (...)

20Curiously, the first part of The Wanderings of Oisin falls into this frequency gap and, in his process of revising the text in 1895, Yeats made its metrical aberrance stronger. The poem is too anapaestic (in the old terminology) to be ‘still iambic’, but not anapaestic enough to be definitely anything else.33 In the 1889 text, by my analysis, some 10% of the stress (interictic) intervals in its 506 lines are occupied by two syllables: this is about the same proportion as in Coleridge’s ‘Christabel’. But, in the slightly shorter 1895 first part (almost identical, apart from some name differences, to the next revise), over 14% of the intervals in 414 lines are disyllabic. So, by Tarlinskaja’s taxonomy, this new version is neither strictly iambic, nor even on the outer reaches of ‘loosely iambic’, as the first version was, but falls into a ‘gray zone instinctively avoided by literary authors’ (505–6).

  • 34 Russell K. Alspach, ‘Some Sources of Yeats’s The Wanderings of Oisin’, PMLA 58.3 (Sept. 1943), 849– (...)
  • 35 John O’Daly, ‘Introduction’, Transactions of the Ossianic Society for the Year 1856, Vol. 4 (Dublin (...)
  • 36 And which he acknowledges only obliquely in the 1895 glossary to the poem. Russell K. Alspach, ‘Som (...)
  • 37 Warwick Gould, ‘Lips and Ships, Peers and Tears: Lacrimae Rerum and Tragic Joy’, YA18 (2013), 15–56 (...)
  • 38 Muiris O Rochain claims that the form of the poem is ‘in a form described as rannaíocht mhór’, 64, (...)

21The fact that Yeats changed his versification in revision suggests that the turn away from an iambic pattern without embracing the deliberate dolnik of ‘The White Birds’ is a considered and deliberate aesthetic choice. But what kind of choice is it? One possibility is that the poem’s unusual metrics have some thematic or generic resonance, and are, if unconsciously, a deliberate way of declaring kinship with the poem’s diverse source materials. In a letter to the Spectator of 29 July 1889, Yeats said that the first few pages were ‘developed’ —an interesting word, like a film script from a novel—from ‘a most beautiful old poem written by one of the numerous half-forgotten Gaelic poets who lived in Ireland in the last century’. He added that ‘in the quarrels between the saint and the blind warrior, I have used suggestions from various ballad dialogues between Oisin and Patrick, published by the Ossianic society’. (CL1 176–77; unnumbered in CL Intelex). In other words, given that he knew no Gaelic, he was reliant on a series of verse translations published in the 1850s. And these translations, precisely because they aimed at fidelity to the original Gaelic text (which was published on the facing page), tended to be, prosodically at least, quite unlike late-nineteenth-century English language poetry. Russell Alspach gives some examples of the rather literal, prosaic verse that the Ossianic Transactions contained.34 Returning to the original volume shows, in addition, that readers were presented with little introductory or paratextual material to explain the method of translation, or the relationship between the facing-page English and Gaelic texts. O’Daly’s introduction is devoted mostly to explaining the complex genealogy of various Irish kings and militias, and some summary of the scribal manuscripts from which the text derives (xxxi); one paragraph at the end asks the English reader ‘to excuse the style, consequent upon our being obliged to adhere as closely as the idioms of the English language would admit to our originals’.35 But, according to Alspach, an even more important source than these antique, original materials was Michael Comyn’s much more recent poem, The Land of Youth, which was also published in the Ossianic Society’s Transactions, in a translation by Bryan O’Looney.36 As Warwick Gould explains, this was also an extremely faithful translation, with hanging indents allowing the English quatrains to mirror the original Irish.37 It is perhaps the rigidity of this typesetting that makes the prosodic irregularity of the quatrains more startling, if we read O’Looney’s text as an English poem. Individual lines tend to come as semantically complete units but they have no rhythmic integrity or principle of repetition. Might Yeats’s own, strangely halting prosody be declaring some affinity with the haiku-like, proto-Imagist simplicity of its source? By source, of course, we mean the literal English translation (which sacrifices prosodic regularity for semantic accuracy) rather than the original, syllabic Irish verse form.38

  • 39 ‘The Land of Youth’, ed. by Bryan O’Looney, Transactions of the Ossianic Society for the Year 1856 (...)

A royal crown was on her head;
And a brown mantle of precious silk
spangled with stars of red gold
Covering her shoes down to the grass.
39

  • 40 ‘The Lay of Oisin on the Land of Youth’ is set as prose in Alfred Graves, The Irish Fairy Book (Lon (...)

22In fact, in later editions like Alfred Graves’s 1909 The Irish Fairy Book, O’Looney’s translation was reset as prose, as if it has never been a genuine verse translation in the first place.40 In Yeats’s own Oisin poem, the lingering presence of these prosaic materials may be entirely accidental, a mild stylistic overlay that resulted from his having to work with translations. But the consequence was the production of a poem that is, by the standards of the 1890s, metrically innovative and which, in a longer literary history, we might see as looking forward to some of the free verse experiments of the 1910s (also heavily influenced by translation).

23The first book of The Wanderings of Oisin occupies a novel ‘gray zone’ in its use of disyllabic intervals, while retaining lines of conventional length, but in the second, and especially the third, books, Yeats’s lines expand and, in doing so, become freer still. In the second part, a four-beat line turns into a five-beat line, which is sometimes a well-formed iambic pentameter but equally often a ‘near miss’, a line that comes close enough to pentameter that its unmetricality is painfully marked.

Now, man of croziers, shadows called our names
And then away, away, like whirling flames;
And now fled by, mist-covered, without sound,
The youth and lady and the deer and hound;
‘Gaze no more on the phantoms’, Niamh said,
And kissed my eyes …. (
VP 29)

24Here we have three lines of relatively regular pentameter washing up in line four. ‘The youth and lady and the deer and hound’ only evokes the metre if the first and third ‘and’ are unstressed, but the second is stressed, a flouting of a self-instituted pattern that is bound to be awkward on the ear. Niamh’s first words, ‘Gaze no more on the phantoms’, also present aural problems. In traditional English poetry, ‘no more’ is invariably stressed on the second of the two syllables, as shown by such diverse examples as the falling trochaics of Cymbeline, ‘Fear no more the heat o’ the sun’; Shelley’s breathless lament in ‘Adonais’, ‘He will awake no more, oh, never more!’, and Tennyson’s ‘So sad, so strange, the days that are no more’. Here, we can only ‘right’ the verse to its iambic pattern by saying ‘Gaze NO more on the phantoms’ a strained and unnatural emphasis that the reading voice does not expect to make.

25If we were unaware that these lines were the consequence of deliberate post-publication revision, we would be tempted to assume that Yeats had difficulty meeting the requirement of iambic pentameter. But knowledge of the extensive post-publication revision contributes to a sense that the metre is deliberately being forced to breaking point. Pound’s precepts for free verse never suggest writing in a way that is deliberately unmetrical, but Yeats’s disregard here for the normal constraints of iambic pentameter is so flagrant as to seem aggressively rebellious. What is gained in doing so, in changing a line like the regular ‘In triumph with her arms around me’ to the unmetrical ‘With her triumphing arms around me’? The answer, I think, is that breaking the pentameter was the only solution Yeats found for making the syntax more natural. This is the first version.

  • 41 This text published by Yeats as The Wanderings of Oisin and Other Poems (London: Kegan Paul, 1889), (...)

Now, man of crosiers, phantoms drew around
Once more—the youth and lady, deer and hound;
Half lost in vapour, shadows called our names,
And then away, away like spiral flames.
‘These forms?’ ‘Vex not with speech the phantoms dread’.
And now sang Niam, swaying her bright head
And her bright body… (
VP 29)41

  • 42 Michael J. Sidnell, ‘j. b. Yeats’s Marginalia in The Wanderings of Oisin and Other Poems’, YA13 265 (...)
  • 43 See George Bornstein’s transcription (page not reproduced in facsimile) in Bornstein ed., The Early (...)

26The earlier version is metrically more traditional by virtue of being, by the standards of the late nineteenth century, syntactically rather archaic; to fit the metre, in other words, Yeats is having to make use of inversions like ‘vex not’ (for ‘do not vex’) and ‘phantoms dread’ (for ‘dread phantoms’). In the later version, the archaic ‘sang Niam’ is resolved into the more normal ‘Niamh said’, while appearing to retain the incorrect pronunciation of ‘Niamh’ that John Butler Yeats had queried. In his marginal comments on the first edition, he asked ‘but is not Niam (Niambh) pronounced as Neev’? as one syllable? Not “Nee-am?”’42Strangely, in the manuscript revise preceding the 1895 publication, Yeats changed ‘Niam’ to ‘Neeve’ and he printed ‘Neave’ in at least one 1895 edition.43

27The second book is freer than the first, and the third is more metrically inventive again. Here are its first and third stanzas:

Fled foam underneath us, and round us, a wandering and milky smoke, High as the saddle-girth, covering away from our glances the tide; And those that fled, and that followed, from the foam-pale distance broke; The immortal desire of immortals we saw in their faces and sighed ….

Were we days long or hours long in riding, when rolled in a grisly peace, An isle lay level before us, with dripping [1889: dripping with] hazel and oak? And we stood on a sea’s edge we saw not; for whiter than new-washed fleece Fled foam underneath us, and round us, a wandering and milky smoke. (VP 47)

  • 44 William Wordsworth, letter to John Thelwall, 1804, quoted by Peter McDonald in Sounds Intentions: T (...)
  • 45 In John Henry Newbolt, My World as in My Time: Memoirs of Sir Henry Newbolt (London: Faber & Faber, (...)
  • 46 Letter, June 8, 1901, in Donald E. Stafford ed., The Selected Letters of Robert Bridges, Vol. 1 (Ne (...)

28One danger of writing long, end-stopped, lines in English is that stresses pile up at the right edge of the page, with the nuclear stress of each sentence or clause coinciding with the special metrical stress or pause that the ‘passion of metre’, in Wordsworth’s phrase, inevitably demands.44 Yeats avoids this by heavy use of noun-compounds (‘saddle-girth’, ‘foam-pale’, ‘new-washed’) where stress is carried on the leftmost item, as well as by forms of syntactic inversion and displacement. The word-order rearrangement also produces clusters of alliterated consonants, both within and across lines. For example, the normal, unmarked order in English would be ‘[foam fled underneath us] [whiter than new-washed fleece]’, with subject preceding verb, and the adjunct clause following the argument. But we have instead ‘whiter than new-washed fleece/ Fled foam underneath us’, with the most strongly stressed words in each phrase (‘fleece’ and ‘fled’, by the nuclear stress rule) placed abutting each other across the line break. The metrical effect is both very old, in evoking of some features of the Old English verse line, and very new for 1889 or, more precisely, it is novel by virtue of its instinct for renovation. Pound’s ‘The Seafarer’ (1911) and Auden’s ‘The Wanderer’ are later, more self-conscious experiments with some of the same techniques. For Robert Bridges, who was something of a metre-nerd, the long lines in this third part of The Wanderings of Oisin certainly seemed highly original. When he read the poem in the 1895 Collected, he picked out the line ‘Fled foam underneath us, and round us, a wandering and milky smoke’, as a beautiful and also ingenious rhythm. ‘That’, he said, ‘is a new thing in English poetry—Yeats made it—it was not there before’.45 And we should take this judgment seriously: Bridges was a trained, analytical metrician, who had already published a book on Milton’s metre, and whose letters to Gerard Manley Hopkins have a kind of trainspotter’s fascination with prosodic irregularities and gems. Bridges in fact liked Yeats’s line so much that he was still quoting it four years later, now as a kind of in-joke. Inviting Yeats to visit him, he explains the trains: ‘There is a very first class train from Paddington to Pangbourne of an evening. It leaves Paddington at 6.10 and its wandering and milky smoke does not stop at Reading’.46

29The very fact that Bridges was able to quote Yeats’s phrase in a letter shows that it works in a natural prose sentence; there is nothing baroque or strained about it. Might this help to explain why Bridges was so favourable about Yeats’s line while so condemnatory about his friend Hopkins’s rhythmically very similar poem, ‘The Wreck of the Deutschland’? In a poem written ten years before Yeats began The Wanderings of Oisin, but published almost thirty years after it, Hopkins had described a shipwreck like this:

  • 47 g. m. Hopkins, ‘The Wreck of the Deutschland’, in Robert Bridges ed., Poems of Gerard Manley Hopkin (...)

One stirred from the rigging to save
The wild woman-kind below,
With a rope’s end round the man, handy and brave—
He was pitched to his death at a blow,
For all his dreadnought breast and braids of thew:
They could tell him for hours, dandled the to and fro
Through the cobbled foam-fleece, what could he do
With the burl of the fountains of air, buck and the flood of the wave?
47

30Both Yeats’s and Hopkins’ descriptions contain a metaphoric alignment between white foam on the sea and a sheep’s fleece, and both are attempting to capture some of the erratic movement and noise of the sea. In doing so, they recruit a pattern of metrical arrangement that is decidedly different from the accentual-syllabic ‘norm’ of English poetry. Both poets are writing in long lines that unlike the lines in, say, ‘The White Birds’ are not resolvable into two shorter lines. Moreover, these lines contain an unusual mixture of interictic intervals: sometimes there is one syllable between stresses, sometimes two, sometimes three, and sometimes none. ‘For all his dreadnought breast and braids of thew’ is actually an iambic pentameter; so too, there is only one syllable between each stress in Yeats’s ‘for those that fled’, ‘new-washed fleece’ and ‘milky smoke’. In Yeats’s lines there is a particularly high proportion of disyllabic stresses, ‘were we days long or hours long in riding’. But we find this in Hopkins too: ‘stirred from the rigging to save’. In some places three light syllables intervene between stresses: this, in fact, is what gives the unusual effect to ‘wandering and milky smoke’, and in Hopkins ‘with a rope end round the man’. In some places stressed syllables jam up against stressed syllable with no relief: this produces one of Hopkins’ signature tricks, a kind of eerie, even panicked, upswell of emotion, where one syllable cracks into the next. In the terrible sonnets, it tunes us in to despair’s magnitude, its inevitability: ‘pitched past pitch of grief’. Here it is effective mimetically to suggest the ferocity of the ship-wrecking storm: ‘through the cobbled foam-fleece’ and then instead of a light syllable, an exclamatory ‘WHAT could he do’?

  • 48 Letter 38, August 21, 1877, in c. c. Abbott ed., The Letters of Gerard Manley Hopkins to Robert Bri (...)
  • 49 Joseph J. Feeney, The Playfulness of Gerard Manley Hopkins (Aldershot: Ashgate, 2008), 74.
  • 50 The rule is usually stated the other way round, so Guglielmo Cinque argues ‘stress prominence in a (...)

31Robert Bridges was probably the only person before 1918 who read both of these poems. He approved of Yeats’s work as something entirely new but, as is well known, pronounced himself shocked by Hopkins’ ‘presumptuous jugglery’, and added that he wouldn’t for any money read the poem again.48 How do we make sense of this? Some of his objections must, of course, have been to the peculiarly maudlin and erotic subject matter of Hopkins’ poem, handled with a degree of fussiness in strophe structures. But why did he find Hopkins presumptuous, while approving Yeats’s similar-sounding lines so fulsomely? One answer is that Yeats has a better sense of syntactic decorum. Yeats’s lines contain some metrically reinforcing but uncolloquial inversions of word order (verb-subject instead of subject-verb, for example) but are otherwise fairly simple; Hopkins is pushing the limits of language in all places at the same time: the drowning sailor’s chest, by a queer and slightly grim punning metaphor, is like a ‘dreadnought’; and instead of saying that his knotted muscles look like braids, the epithet is transferred, ‘braids of thew’. Where Yeats spells out the metaphor, and uses only easily legible noun compounds, ‘whiter than new-washed fleece | Fled foam underneath us’ |‘whiter than new-washed fleece | Fled foam underneath us’, Hopkins compresses everything, so that the adjective ‘cobbled’ modifies an already compact ‘foam-fleece’. Joseph Feeney, in his 2013 study, says this is: ‘An incongruous, brilliant, terse bisociation of cobblestones (hard, cold, and grey), foam (liquid and white), and lamb’s wool (soft and warming)’.49 But metaphors work by mapping one semantic field on to another, not by linking three separate things. Here the problem is driven primarily by the ambiguity of ‘foam-fleece’ which, in the context, has to be read as ‘fleece-like foam’, with ‘foam’ working as the compound’s head. But this is abnormal, because in English noun compounds the second word is usually the head (so a ‘foamflower’ is a kind of flower, but a ‘flowerpot’ is a kind of pot). According to some linguists, it is in fact because the second word is grammatically more important that it is less stressed.50 Here the logical conclusion is that ‘cobbled’ is a property of the foam (also seen as fleece-like), but the structure and stress-patterning of the phrase points to the possibility that ‘cobbled’ in fact modifies ‘fleece’ (after the pattern of ‘cobbled riding-boots’). In the first case, the double metaphor leads to ambiguity; in the second, there is a degree of redundancy.

  • 51 Hopkins’ seven-page selection is sandwiched between Robert Bridges and William Ernest Henley (obmv (...)

32When Yeats edited the Oxford Book of Modern Verse in 1936, he strategically relegated Hopkins to the start, to his moment of composition and not publication: he described this as ‘putting him among the Victorians’ (CL InteLex 6289).51 And the introduction manages an anxiety of contagion, if not quite of influence, by insisting on an enormous gulf between Hopkins’ generation and Yeats’s own.

I read Gerard Hopkins with great difficulty, I cannot keep my attention fixed for more than a few minutes; I suspect a bias born when I began to think. He is typical of his generation where most opposed to mine. His meaning is like some faint sound that strains the ear, comes out of words, passes to and fro between them, goes back into words, his manner a last development of poetical diction. My generation began that search for hard positive subject matter, still a predominant purpose. (OBMV xxxix)

  • 52 Walt Whitman, November Boughs (Philadelphia: David McKay, 1888), 65.
  • 53 t. s. Eliot, ‘Swinburne as Poet’, The Sacred Wood (London: Methuen, 1920), 131–36.

33This, of course, is a bravura rewriting of literary history. When Yeats began to think and write, he sounded quite a lot like Hopkins, and his early poetry shares much of the slackness of both subject-matter and syllable count that he decries. It was Pound’s generation, not Yeats’s, who fetishized ‘hard positive subject matter’. In fact, even Yeats’s rhetorical method here seems somewhat derivative of earlier modernist attacks on ‘messy’ or blurry Victorian writing. He attacks Hopkins for being a poet of blurry surfaces, inadequately referential, untethered in things. This is the complaint that Whitman brings against Tennyson for his ‘finest verbalism’.52 And, at more length, it is the substance of Eliot’s argument with Swinburne’s ‘diffuseness’, a poetry that is ‘merely the hallucination of meaning’, where sound, image, and idea are blurred together into one thing.53 We see the same kind of strategic distancing in a letter from three years earlier.

My generation revolted against ‘poetical diction’ & Hopkins was of the generation that elaborated it. He had it in one form Swinburne in another. His whole life was a form of ‘poetic diction’. He brought his faint theatrical Catholicism to Ireland where it was mocked by the sons of peasants & perhaps died of the shock. (CL InteLex 5623)

34What Yeats does here is to align the Victorian period with failures of both style and observation—an unnatural, ‘poetic’ diction slipping and sliding over its weakly observed visual content, in an ugly metre that suggests ‘hurried conversation’ —and his own poetry with the modernist revolution that overturned them. It is an ingenious and persuasive bit of rhetorical rewriting, despite its heavy reliance on supposition (‘Hopkins would have disliked increase of realism’).

  • 54 Ezra Pound, review of Responsibilities, Poetry 2.4 (May 1914), 64–69 (65).
  • 55 Warwick Gould, ‘The Unknown Masterpiece’, in Andrew Gibson ed., Pound in Multiple Perspectives (Bas (...)

35The Oxford Book of Modern Verse gives us a modernizing Yeats, but it also gives us a modernism without free verse. It is an intelligent piece of literary positioning, less arrière-garde than cannily retrospective. Imagism’s semantic clarity and visual objectivity are separated from its prosodic solution, which Yeats had never liked. In 1914 Pound had claimed Yeats as a slightly prickly associate for the Imagists: ‘Mr. Yeats is a symbolist, but he has written des Images as have a good many poets before him; so that is nothing against him, and he has nothing against them, at least so far as I know—except what he calls their “devil’s metres”’.54 The year before, Yeats had been even more direct in a letter to Harriet Monroe, praising Pound’s vigorous creativity ‘although I do not really like with my whole soul the metrical experiments he had made for you’ (CL InteLex 2284). The mild modifier ‘really’ does little to qualify Yeats’s confident and instinctive preference for metrical over free verse; as Warwick Gould puts it, in these encounters with Pound, we see ‘the underlying robustness of Yeats’s own artistic self-assurance’.55 But it was to be another twenty years before he found the discursive rhetoric to propose his own brand of modern poetry, retaining the bits of Poundian modernism that he liked, while getting rid of the parts he didn’t. Once again, he does this by alignment and association—and some rhetorical subterfuge. If the problem with Hopkins is that his poetry in the end is evanescent and meaningless, the problem with free verse, the anthology argues, is that it is prose.

  • 56 Charles Williams to w. b. Yeats, 11 October 1935, quoted in John Stallworthy, ‘Yeats as Anthologist (...)
  • 57 Robert Alden Rubin, ‘Some Heroic Discipline: William Butler Yeats and the Oxford Book of Modern Ver (...)
  • 58 Daniel Albright’s word in Quantum Poetics: Yeats, Pound, Eliot, and the Science of Modernism (Cambr (...)
  • 59 Roy Foster, w. b. Yeats: A Life II, The Arch Poet, 1935–1939 (Oxford: Oxford University Press, 2003 (...)
  • 60 Jerome McGann, Black Riders: The Visible Language of Modernism (Princeton: Princeton University Pre (...)

36When the editor at Oxford University Press saw Yeats’s provisional list of selections, he was ‘perplexed’ by the free verse poem intended to sit at the anthology’s front, like a kind of advertisement: ‘One entry perplexed me, and no-one here can help. “Pater: The Mona Lisa”— was there such a poem?’56Critics have tended to take Yeats’s resetting of Pater into free verse at face value: Robert Rubin, for example, argues that ‘in presenting the selection as rhythmic vers libre, Yeats defamiliarizes Pater’s words and challenges readers to re-examine them’.57 And almost everyone who has written about Yeats’s free verse poem agrees that it is, in fact, in free verse: a ‘rearrangement’ 58or ‘recasting’59or ‘reformat[ting] of Pater’s lush prose as a free verse poem’.60 In fact, Yeats’s poem is not very like most free verse poems. It begins:

She is older than the rocks among which she sits;
Like the Vampire,
She has been dead many times,
And learned the secrets of the grave;
And has been a diver in deep seas,
And keeps their fallen day about her … (
OBMV 1)

  • 61 t. s. Eliot, ‘American Literature and the American Language’, 58.
  • 62 In the 1800 version of a heavily reworked passage, Wordsworth argues that ‘not only the language of (...)

37Some of its lines, such as ‘And learned the secrets of the grave’, and ‘And tinged the eyelids and the hands’, are perfectly well-formed lines of iambic tetrameter, and the relentless anaphora and grammatical parallelism are highly rhetorical. If Poundian free verse shrugs off all forms of decorative and ornamental ‘poeticism’, to become more economical and sparing even than prose, Pater’s prose is aspiring to the condition of musical verse. Most tellingly of all, Yeats’s line endings do little more than mark out the boundaries between intonational phrases or syntactic units. Rather than being used to create additional and unexpected breaks in sense or sound—this form of counterpoint being free verse’s most important semantic tool—the line breaks are more or less redundant, flaccid echoes of an underlying structure. But this, undoubtedly, is Yeats’s polemical purpose. If free verse is something that can be effected after the fact, by a mechanical process of relineation, then it really is no more than the ‘chopped-up prose’ its common-sense detractors claim. And by rendering a famous bit of Victorian purple prose into Poundian free verse, Yeats counters two important tenets of modernism in poetry. The first is the historical claim that modernism begin in (and not before) 1910, after Pound had arrived in London; the idea, in Eliot’s words, that ‘the point de repère of modern poetry… is the group denominated “imagists” in London about 1910’.61 The second is the connection forged by Pound between plain style and ‘direct focus on the thing’ and a break with traditional metres. (That there is no necessary connection between these two principles is illustrated by Wordsworth’s Preface to Lyrical Ballads, which calls for ‘the language of prose’ to be allied to metrical regularity.)62 Instead, Yeats takes a sardonic tone towards both Paterian aestheticism and free verse. His introduction describes aestheticism flippantly as the moment when poets ‘said to one another over their black coffee—a recently imported fashion—“We must purify poetry of all that is not poetry”’(OBMV ix). Inevitably, it was self-limiting and ‘in 1900 everybody got down off his stilts; henceforth nobody drank absinthe with his black coffee; nobody went mad; nobody committed suicide; nobody joined the Catholic church …’ (xi). Rather than being an attempt to rescue Pater for modernism, then, I would suggest that this relineation was a clever attempt to make ‘vers libre’ seem an embarrassing French affectation, as dated and adolescent as absinthe and black coffee.

38Yeats’s 1936 introduction to the Oxford anthology, presenting a decidedly slanted and polemical vision of modern poetry as a whole, needs to be read alongside the 1937 ‘General Introduction to my Work’, which performs the same summary job on the poet’s own work. The two prose essays are companion pieces. But where the anthology is playful or elliptical in its dismissal of free verse, the ‘General Introduction’ comes out straightforwardly in favour of traditional metres. In this famous passage, Yeats shows an acute awareness of what modern linguistics calls the syntax-prosody interface.

It was a long time before I had made a language to my liking; I began to make it when I discovered some twenty years ago that I must seek, not as Wordsworth thought, words in common use, but a powerful and passionate syntax, and a complete coincidence between period and stanza. Because I need a passionate syntax for passionate subject-matter I compel myself to accept those traditional metres that have developed with the language. Ezra Pound, Turner, Lawrence wrote admirable free verse, I could not. (E&I 521–22)

  • 63 Quoted by Thomas Parkinson, w. b. Yeats: The Later Poetry, 185.
  • 64 j. s. Mill continues ‘What we have said to ourselves we may tell to others afterwards; what we have (...)
  • 65 t. s. Eliot, ‘Tradition and the Individual Talent’, The Sacred Wood, 42–53 (52–53).

39Twenty years before 1937 is 1917: the year after the Easter uprising, the year before the end of the Great War, and the year in which Prufrock and Other Observations was published. This is a very epigonal ‘beginning’ for a poet who published his first poem in the 1880s, but perhaps Yeats is correct to date his mature style to this point in time And Yeats makes it clear that his own style is founded not on any specialty of lexis, or even metrical ingeniousness, but on its syntax. It is to serve the goal of a ‘passionate syntax’ that he is forced to ‘accept’ traditional metres; the decision, he implies, is made for him. What exactly ‘passionate syntax’ consists in is harder to say. From comments made elsewhere, it seems clear that Yeats does not mean a mimetic groping after spoken language with its hesitancies and repairs; in an unpublished 1927 notebook, he is negative about Browning’s attempt to give ‘an impression of reality from ejaculations and suppressions’ for these are ‘all an avoidance of the expression of passion’.63 But if a passionate syntax is not Browning’s patina of spoken language, nor is it to be found in the meditative mourning at-one-remove of In Memoriam, pronounced ‘detestable because of its syntax’. It is associated with ‘common personal speech’ and ‘profound feeling’ and, as the ‘Introduction to My Plays’ adds, with the ear rather than the eye—that is, not with the phanopoeic poetry of Imagism. ‘I have spent my life in clearing out of poetry every phrase written for the eye, and bringing all back to syntax that is for ear alone’ (E&I 529). In particular, this kind of speech seems to be associated with self-talk, ‘whatever language comes most naturally when we soliloquise, as I do all day long’. Yeats’s focus on poetry as soliloquy rather than communication, as speech free from any particular speech act, is reminiscent of j. s. Mill’s belief that ‘all poetry is of the nature of soliloquy’: a kind of language which may afterwards be repeated in front of an audience, but which owes its essential nature to its hermeticism.64 By soliloquy, however, he doesn’t seem to mean a rational, even philosophical form of cogitation and meditation; instead he associates himself with the ‘angry’ and ‘wild’ thoughts of ‘mad old slum women… denouncing and remembering’. In fact, it is because this language is impassioned, personal, and wild that it needs the corrective strictness of traditional form: to write a passionate syntax in free verse would risk being sloppy, egotistical, and dull: ‘If I wrote of personal love or sorrow in free verse, or in any rhythm that left it unchanged, amid all its accidence, I would be full of self-contempt because of my egotism and indiscretion, and foresee the boredom of my reader’ (E&I 522). This word ‘accidence’ take us back to the very beginning of the essay, where Yeats contrasts the work of the poet with that of the novelist. ‘A poet writes always of his personal life, in his finest work out of its tragedies, whatever it be, remorse, lost love or mere loneliness; he never speaks directly as to someone at the breakfast table, there is always a phantasmagoria’. As a result, Yeats goes on to say, the poet is ‘most himself’ when he is also least himself: ‘he is never the bundle of accident and incoherence that sits down to breakfast; he has been re-born as an idea, something intended, complete. A novelist might describe his accidence, his incoherence, he must not, he is more type than man, more passion than type’. This idea—of a poetry that is impersonal because of, rather than despite, its intense origin in the personal life of the poet—is obviously reminiscent of Eliot’s formulation in ‘Tradition and the Individual Talent’ that ‘poetry is not the expression of personality, but an escape from personality’.65

  • 66 t. s. Eliot, ‘Ulysses, Order, and Myth’, The Dial 75.5 (November 1923), 480–83; ‘Tradition and the (...)
  • 67 t. s. Eliot, ‘Reflections on Vers Libre’, The New Statesman, 3 March 1917, https://www.newstatesman (...)
  • 68 Stravinsky even judged that Wagner’s opera ‘must have been one of the most passionate experiences i (...)

40This argument via antithesis, where two opposites are reconciled to produce a tensile whole, is reminiscent of t. s. Eliot; so many of his essays work through an act of Hegelian synthesis. For Eliot, the mythical method is not a way of paying attention to the past but ‘of controlling, of ordering, of giving a shape and a significance to the immense panorama of futility and anarchy which is contemporary history’, and the ‘most individual’ parts of a writer’s work those ‘in which the dead poets, his ancestors, assert their immortality most vigorously’.66 Eliot also applied this line of thought to form, most notably in his 1917 essay on ‘Vers Libre’, which finds that so-called free verse is anything but: ‘freedom is only truly freedom when it appears against the background of an artificial limitation’.67 Most of all, he pursued it in his own work, writing a poetry of undeniably modern content and language in hard, classical forms. The kinship between Eliot and Yeats on this point seems obvious, and yet Yeats failed (or chose not) to recognize it. His introduction describes Eliot as technically competent in a completely classical way: ‘He is an Alexander Pope, working without apparent imagination, producing his effects by a rejection of all rhythms and metaphors used by the more popular romantics rather than by discovery of his own, this rejection giving his work an unexaggerated plainness that has the effect of novelty’ (OBMV xxi). In some ways, he presents Eliot as Hopkins’ opposite: the one baroque, eccentric, ornamental; the other harshly and tediously plain, writing a monotonous poetry without any special subject matter where ‘Tristram and Isoult were not a more suitable theme than Paddington Railway Station’ (E&I 499).68 This is not literally quite true, of course: the story of Tristram and Isoult appears in several places in Eliot’s poetry, via its Wagnerian mediation as Tristan und Isolde, while Paddington station appears not once. Just as with Hopkins, the brittle criticism papers over an uneasy and significant anxiety of relation. For is this ‘unexaggerated plainness’, or the use of traditional form for its own sake, not exactly what Yeats claims in the 1937 General Introduction to be doing himself? In describing the productive counterpoint between strict form (the five notional stresses of iambic pentameter) and the ‘natural’ speech pronunciation of the same words, he even uses the same ghostly metaphor as Eliot.

If I repeat the first line of Paradise Lost so as to emphasise its five feet I am among the folk singers—‘Of man’s fírst dísobédience and the frúit’, but speak it as I should I cross it with another emphasis, that of passionate prose—‘Of mán’s first disobédience and the frúit’, or ‘Of mán’s first dísobedience and the frúit’; the folk song is still there, but a ghostly voice, an unvariable possibility, an unconscious norm. What moves me and my hearer is a vivid speech that has no laws except that it must not exorcise the ghostly voice. I am awake and asleep, at my moment of revelation, self-possessed in self-surrender …

41Eliot’s essay on ‘Vers Libre’, published in 1917 (perhaps, not so coincidentally, the date Yeats chooses for his new beginning) also uses the metaphor of a ghost throughout. His source is Hamlet. ‘The ghost of some simple metre should lurk behind the arras in even the “freest” verse; to advance menacingly as we doze, and withdraw as we rouse’.

  • 69 ‘Vacillation’ IV (VP 501); ‘Under Ben Bulben’ (VP 640); ‘The Tower’ (VP 411).

42In broadest outline, then, Yeats began as a late Victorian prosodic innovator, wrestling with long lines, heavy alliteration, and an accentual-syllabic poetry tipping into accentual lines alone, as a kind of Hopkins. He ended up sloughing off these habits: by the time we get to, The Tower (1928), not only is Yeats’s versification more ‘regular’ but it is also, in a stronger sense, like Eliot’s ‘more classical’. Yeats’s depiction of Eliot as ‘Popian’ is curious, I think, and leads to a fresh point of triangulation with Pound. For without Pound’s aggressive editing of the 1920–1921 draft, The Waste Land would have been much more Popian: Yeats is describing Eliot’s initial instincts for that poem, which he could not have known, more accurately than its final form. In his own case, modernization meant accepting many of Pound’s dicta about visual observation, while repudiating his metrical experiments. His poetry from the mid-teens onwards gives brilliantly detailed flashes of other places and times, like ‘that dolphin-torn, that gong-tormented sea’ in ‘Byzantium’, as well as admitting realistic detail into a poetic landscape at once more everyday and more specified that the ‘dim’ worlds of The Wanderings of Oisin. Beginning, perhaps, with the description in ‘Adam’s Curse’ of going down on marrow-bones to ‘scrub a kitchen pavement’ (VP 204) this new poetry finds space for henwives, a ‘crowded London shop’, ‘porter drinkers’ randy laughter’, and good fellows shuffling cards ‘in an old bawn’.69 As he admitted more mundane content, Yeats also became more precise about a line’s syllable count and more artful in producing varied effects within a fixed grid. To ‘The Song of the Happy Shepherd’ or the first part of ‘The Wanderings of Oisin’, we might compare the more metrically uniform, while internally varied, effect of lines like these: now there is no regular or repeated coincidence between metrical positions and word boundaries.

In pity for man’s darkening thought
He walked that room and issued thence
In Galilean turbulence … (
VP 437)

  • 70 Eliot writes: ‘A good poet will usually borrow from authors remote in time, or alien in language, o (...)

43Some of these techniques he learned, like the younger generation of modernists, by study of Elizabethan and Jacobean writers. In 1912, he read Grierson’s edition of Donne, like everybody else, and he wrote to Grierson to thank him. Here, as in the 1937 General Introduction, we see the equation of passion with precision: ‘Poems that I could not understand, or could but vaguely understand are now clear & I notice that the more precise & learned the thought the greater the beauty, the passion. The intricasy & subtlety of his imagination are the length & depth of the furrow made by his passion’ (CL InteLex 2015). But, in fact, he had been ahead of the fashion. In 1908, he was already advising, in language curiously reminiscent of Eliot’s essay on Philip Massinger: ‘I am always telling young writers here that even Shelley & Wordsworth are too near us in mood & manner to be safe models. We should all go back to Spenser & Donne’ (CL InteLex 894).70 Of course, when he himself was a young writer, back in the 1880s and 1890s, Donne had not been one of his models.

  • 71 Pound describes his sense that the ‘dilutation of vers libre’ had gone ‘too far’ in ‘Harold Monro’, (...)

44The focus of this essay has been on Yeats’s metres. But it is not as a prosodist that Yeats was an innovator. I have said that his re-embrace of traditional metres was an act of modernism, a conscious regression, and yet it was also a shared action: Eliot and Pound were already imitating Gautier’s ‘hard’ metres in the mid-teens, to correct an Imagism that had gone too far.71 What is innovative, and distinctive, is his focus on the relationship between prosody and syntax, both of which he puts under serious constraint.

  • 72 See Chris Golston and Tomas Riad, ‘The Phonology of Classical Greek Meter’, Linguistics 38.1 (2000) (...)
  • 73 Ezra Pound, ABC of Reading (New York: New Directions, 1934), 35.
  • 74 j. c. c. Mays gives the history in ‘Coleridge and Yeats: The Romantic Voice’, Variants 6 (2007), 65 (...)

45Contemporary linguistics has suggested that traditional poetry (across languages) is a kind of speech where prosody outranks syntax.72 In other words, a sentence like ‘Him who disobeys me disobeys’, which, as Ezra Pound noted, is not a felicitous English sentence in itself, is acceptable in Paradise Lost because it fits the prosodic constraints of Miltonic blank verse.73 (In fact, contingent on using the word ‘disobey’, Milton’s may be the only sentence that fits the metre.) Imagism’s insistence that poetry be as well-written as prose can then be understood, in linguistic terms, as constraint reranking. Now, the free verse poets claimed, prosody should never be more important than syntax. The uniqueness of Yeats’s achievement—and its difficulty (as, I think, his ‘restless’ bouts of revision illustrate)—was in agreeing always to put natural syntax first while also working within the constraints of traditional metres. This makes the kind of ‘containment’ that metre offers rather different: instead of pouring a flexible material (sentences where word order is flexible, where contractions are possible, where lexical archaism is allowed, etc.) into a metrical shape, Yeats insists on fitting normal, contemporary speech into the same pattern, tessellating words and position. The idea of rewriting Lyrical Ballads so that the poems include no scansion-enabling syntactic archaisms will give a sense of how difficult this is, in practice. And the possibility that the procedure could go too far also seems a real danger; where Henry James managed to drive his novels, in their final New York Edition, to magisterial befuddlement, Yeats was, at times, in danger of revising towards the banal. In a BBC recording of his poems, for example, we are told that he condemned the line ‘That is no country for old men. The young …’ as ‘the worst piece of syntax I ever wrote’, proposing instead the syntactically easier, but workaday alternative ‘Old men should quit a country where the young’.74 The second version is not only more cohesive (it gets rid of the deixis to an unspecified referent), but, curiously, it is more metrically regular, in fitting the syntactic clause to the line, rather than breaking unevenly, after eight syllables.

  • 75 From, respectively: the first lines of ‘A Prayer on Going into my House’ (VP 371) and ‘Easter, 1916 (...)

46No one can regret that Yeats did not live to see ‘Old men should quit’ into print. But the revisionary instinct, which is to replace a copula with a finite verb, reflects Yeats’s belief that passionate syntax involved the use of finite verbs to predicate something or inquire about predication. It is remarkable how confidently Yeats’s mature poems propose, declare, and inquire, often confining their sentences within the small orbit of the line. ‘God grant a blessing on this tower and cottage’, ‘I have met them at the close of day’, ‘How can we know the dancer from the dance?’ ‘What can I but enumerate old themes’, ‘We were the last romantics’, ‘What’s water but the generated soul?’75 Stanza structures allow more complex propositions to be developed, the subordinate clauses nested within two bars of white space. Think, for example, of how Yeats manages a complex question in the fifth stanza of ‘Among School Children’, placing his subject ‘what youthful mother’ in line one, the main verb ‘would think her son’ in line five, and its complement in line seven ‘a compensation for …’ His most dramatic and successful revisions of the early poems are often, primarily, syntactic reworkings. Take, for example, the heavily rewritten ‘The Sorrow of Love’. Yeats had originally (1891–1892) built the second and third stanzas up agglutinatively, starting with ‘and then you came with those red mournful lips’, adding three end-stopped noun phrases in qualification, before turning back, chiastically, to the sparrows:

And then you came with those red mournful lips,
And with you came the whole of the world’s tears.
And all the sorrows of her labouring ships,
And all the burden of her myriad years.
And now the sparrows warring in the eaves,
The crumbling moon, the white stars in the sky,
And the loud chanting of the unquiet leaves,
Are shaken with earth’s old and weary cry. (
VP 120)

  • 76 Roman Jakobson’s slightly longwinded analysis of the two versions agrees that the 1925 version ‘con (...)
  • 77 For Louis MacNeice, this is not an advantage: ‘But perhaps this poem ought to be languid’. The Poet (...)
  • 78 Warwick Gould discusses why Yeats might have made the Trojan frame of reference clearer, ‘Writing t (...)

47Just as ‘came’ is a weak verb to support the whole verbal action of the second stanza, so ‘are shaken’ is delayed and indefinite in the third.76 The impression given by the ‘then’/’now’ partition, and the anaphoric ‘ands’ is flat and languid; the casual relationship between the woman with red lips and the disenchantment of the last stanza is not pursued. Something happens to sour the contemplative plenitude of the first stanza, but what is not exactly stated. The 1925 revised version is more precise in every respect.77 The vague Homeric flavour of ‘labouring ships’ and ‘myriad years’ is explained; finite verbs are used with more purpose; and the boxed-in stanzas of the first version are broken down.78 In the final version the source of sorrow is made clear by the repetition of ‘arose’:

A girl arose that had red mournful lips
And seemed the greatness of the world in tears,
Doomed like Odysseus and the labouring ships
And proud as Priam murdered with his peers;

Arose, and on the instant clamouring eaves,
A climbing moon upon an empty sky,
And all that lamentation of the leaves,
Could but compose man’s image and his cry.

48This poem is so substantially rewritten that the metrical changes are not simple replacements. Once again, however, it is noticeable that slacker versifying in the original version, which contains several hypermetrical lines, and others that require metrical stress on syllables unstressed in normal speech (‘And the loud chanting of the unquiet eaves’ is not virtuosic), is replaced in the final form by clean, more classical iambic pentameter.

  • 79 In Poetry 2.1 (April 1913), the poem is printed with a colon at the end of the first line; a semi-c (...)
  • 80 Ezra Pound, Canto 1, The Cantos of Ezra Pound, 7.
  • 81 On superposition, see Ezra Pound, Gaudier-Brzeska (London: John Lane, 1916), 103.
  • 82 This aversion from the indicative is seen even in Eliot’s epigraphs: ‘o quam te memorem virgo’ in ‘ (...)

49Yeats’s use of verbs could not be more different from Pound’s. Imagist poems are, as syntax, often little more than rubble. ‘In a Station of a Metro’ is an elliptical conjoining of two apparently unrelated things, a kind of copula without an ‘are’ or ‘seem like’ to specify the relationship. Instead of a verb, bibliographic coding—a colon or semi-colon, in different printings—is required to provide the pivot of meaning between two lines.79 And this is true not only of Pound’s early experiments in Imagism; he is a poet who seems, throughout his career, to have had a neglectful or aberrant relationship with finite verbs, preferring instead to heap up noun phrases. Canto 1 begins ‘And then went down to the ship’: who is not specified until line three, when ‘we’ arrives.80 This was, of course, a somewhat theorized and conscious activity. Pound’s belief in the value of ‘superposition’ is a commitment to not making sentences, in merely allowing various charged bits of textual matter to lie suggestively beside each other.81 Eliot does not shy away so completely from the finite verb, but his poetry repeatedly hedges and qualifies, modifying the apparently confident proposition (‘April is the cruelest month’) with hanging participle phrases (‘breeding… mixing… stirring…’), turning from the indicative to the conditional mood (‘Let us go’, ‘And how should I presume’), or professorially standing back in self-critique (‘That was a way of putting it—not very satisfactory’).82

  • 83 w. h. Auden, ‘The Public v. the Late Mr. William Butler Yeats’ [1939], in Edward Mendelson ed., w. (...)
  • 84 In Yeats’s drafts, we very often find a list of complete phrases running down the page, connected, (...)

50It is because of the extraordinary propositional density of his poetry that Yeats is so quotable, so memorable. In an era of writers in both prose and poetry who avoided direct propositions, embracing instead the deferral of anaphora (as in the passage from Pater), the ambiguity of the omitted verb, or the sinuous hesitancy of subordination, Yeats’s active lines—active in the strong sense of being propelled by active, finite verbs—are unique. This was something that w. h. Auden recognized, in a backhanded way, when he arraigned the recently dead poet for being excessively rhetorical, seducing us with ‘the fallacious belief that art ever makes anything happen’.83 But this drive to clarity was also something that Yeats had to learn. It is achieved by highly active syntax and traditional metres, and also by a pretty close fit between the syntactic unit and the prosodic unit. A relatively high proportion of punctuation marks and phrase boundaries occur at the end of the line, but ‘complete coincidence’ is avoided: Yeats did not want the obviousness of Pater’s free-verse ‘Mona Lisa’, where lines do nothing other than mark boundaries between phrases.84 By keeping the syntactic and the verse unit close, but just slightly off-kilter, Yeats forged a ‘natural’ style capable, at times, of pointed metrical artifice. By refusing to write free verse he also, ironically, enjoyed a liberty with finite verbs and complete statements that is unique in twentieth-century English poetry.

Notes

1 Note—Further information may have been gathered since this article was prepared for publication. If you would like to find out if any further information has been discovered that may help your own research, why not write to the author at hannah.sullivan@new.ox.ac.uk? Quite apart from anything else, feedback is always welcomed.

2 Charles O. Hartman’s Free Verse: An Essay on Prosody (Princeton: Princeton University Press, 1980) gives an extended and thoughtful version of this story, which grows out of modernist polemic. Robert Bernard Hass describes free verse as the prosodic change accompanying modernism’s other disjunctions of content and structure in ‘(Re)Reading Bergson: Frost, Pound, and the Legacy of Modern Poetry’, Journal of Modern Literature 29.1 (2005), 55–75 (67).

3 Joseph Phelan, The Music of Verse: Metrical Experiment in Nineteenth-Century Poetry (Basingstoke: Palgrave Macmillan, 2012), 11. And see Meredith Martin’s ‘Introduction’ to The Rise and Fall of Meter: Poetry and English National Culture 1860–1930 (Princeton and Oxford: Princeton University Press, 2012) for careful critique of ‘the narrative we have been taught’ (3).

4 See, for example, the adjectives employed by d. h. Lawrence in ‘Poetry of the Present’, in Michael Herbert ed., d. h. Lawrence: Selected Critical Writing (Oxford: Oxford University Press, 1998), 75–79 (78).

5 t. s. Eliot, ‘American Literature and the American Language’, 1953, republished in To Criticize the Critic (London: Faber & Faber, 1978), 43–60 (58).

6 From a search across the Google Books corpus. By 1918, ‘free verse’ is used about seven times more often, and ‘vers libre’ five times more often, than ‘iambic pentameter’. But ‘vers libre’ had a shorter useful history than ‘free verse’, and remained in greater use than the fairly stable ‘iambic pentameter’ only until 1937. The rapid upswing in the use of both terms is really remarkable (compared, say, to the slower rise in the use of ‘imagist’), and significant even with the usual concerns about Google Books as a corpus (including but not limited to: OCR errors, duplication in corpus, bias towards academic/scientific writing). Google Ngram Viewer: ‘[free verse]’, ‘[Free verse]’, ‘[Free Verse]’, ‘[FREE VERSE]’, ‘[vers libre]’, ‘[Vers Libre]’, ‘[Vers libre]’, ‘[VERS LIBRE]’, ‘[iambic pentameter]’, ‘[Iambic pentameter]’, ‘[Iambic Pentameter]’, 1890–1940 in English.

7 Ezra Pound, letter to Harriet Monroe, Jan. 1915, in d. d. Paige ed., Selected Letters of Ezra Pound: 1907–1941 (New York: New Directions, 1950), 49.

8 He also cites Yeats’s troubled, but self-critical, remark to Lady Gregory, ‘Ezra… helps me to get back to the definite and concrete and away from modern abstractions. To talk over a poem with him is like getting you to put a sentence into dialect. All becomes clear and natural’. A. Walton Litz, ‘Pound and Yeats: The Road to Stone Cottage’, in George Bornstein ed., Ezra Pound Among the Poets (Chicago: University of Chicago Press, 1985), 128–48 (138–39).

9 These include the 1937 ‘General Introduction’, the 1936 Oxford Book of Modern Verse, and the poems themselves in their final form. Having published a Collected Poems in 1933, Yeats was, in 1937, anticipating an expansion of the edition ‘in about two years time’. Richard J. Finneran, ‘Preface’, The Collected Works of w. b. Yeats, Vol. 1: The Poems (New York: Scribner, 1996), 2nd ed., xxiii.

10 Warwick Gould, ‘Writing the Life of the Text: The Case of w. b. Yeats’, Miscelánea: A Journal of English and American Studies 30 (2004), 9–34 (9).

11 Thomas Rees, ‘Ezra Pound and the Modernization of w. b. Yeats’, Journal of Modern Literature 4.3 (February 1975), 574–92 (580).

12 Martin J. Duffell, A New History of English Metre (London: Legenda, 2008), 189.

13 Laura O’Connor, Haunted English: The Celtic Fringe, the British Empire, and De-Anglicanization (Baltimore: Johns Hopkins University Press, 2006), 84.

14 Peter Howarth, British Poetry in the Age of Modernism (Cambridge: Cambridge University Press, 2005), 9. Donald Davie, Articulate Energy: An Inquiry into the Syntax of English Poetry (London: Routledge and Kegan Paul, 1976), 125.

15 Thomas Parkinson, w. b. Yeats: The Later Poetry (Berkeley and Los Angeles: University of California Press, 1966), 187–89. Despite the diachronic coverage of his work, Parkinson also tends to write as if Yeats worked in the same way for his whole career: see the description of his general practice of composition, 182–83.

16 Helen Vendler, Our Secret Discipline: Yeats and Lyric Form (Cambridge, MA: Harvard, 2007). The plural ‘metres’ is used, but both when quoting Yeats, not in analysis: 16, 145; ‘free verse’ once, 347.

17 Marjorie Perloff, Rhyme and Meaning in the Poetry of Yeats (The Hague: Mouton, 1970), 13.

18 w. b. Yeats, ‘Under Ben Bulben’, in Peter Allt and Russell K. Alspach eds., The Variorum Edition of the Poems of w. b. Yeats (New York: Macmillan, 1957), 639. Unless otherwise stated, future references to poems are to this edition.

19 Yeats uses the term to describe Adah Menken’s writing in a letter to Olivia Shakespear, 24 July, 1934, in John Kelly, Eric Domville, Warwick Gould, and Ronald Schuchard eds., The Collected Letters of w. b. Yeats: Electronic Edition, Unpublished Letters (Charlottesville: InteLex, 2002).

20 w. b. Yeats, Poems (London: T. Fisher Unwin, 1895), vii.

21 See, for example, George Saintsbury’s ‘Rules of the Foot System’, numbers 19 and 22, prohibiting substitutions that confuse the basis of the metre, and claiming that ‘the most important and valuable engine in the constitution of English verses is the variation of the middle or internal pause’, Historical Manual of English Prosody (London: Macmillan, 1910), 32–33.

22 Ezra Pound, ‘A Few Don’ ts by an Imagiste’, Poetry 1.6 (March 1913), 200–6 (203).

23 Susan Grace Galassi, senior curator at the Frick, cited by Catesby Leigh, ‘Could Picasso Draw Better than Raphael?’ Standpoint Magazine, June 2012, http://standpointmag.co.uk/node/4499/full

24 In a radio interview, 19 November 1933, reprinted by Walter Frisch, Schoenberg and His World (Princeton: Princeton University Press, 1999), 300.

25 Letter to Edith Shackleton Heald, 10 August 1937. For ‘prosedy’, see CL InteLex 2995, 2998.

26 In a letter to Edward Dowden, cited by Ronald Schuchard, The Last Minstrels: Yeats and the Revival of the Bardic Arts (Oxford: Oxford University Press, 2008), 1.

27 r. f. Foster, The Apprentice Mage, 1865–1939 (Oxford: Oxford University Press, 1997), 336.

28 John Todhunter, review of The Wanderings of Oisin and Other Poems, The Academy, January–June 1889.

29 Matthew Campbell, ‘Poetry in the Four Nations’, in Richard Cronin, Alison Chapman, and Antony W. Harrison eds., A Companion to Victorian Poetry (Oxford: Blackwell, 2002), 438–56 (447).

30 Thomas Parkinson, w. b. Yeats: The Later Poetry, 188–89.

31 Marina Tarlinskaja, ‘Beyond “Loose Iamb”: The Forms and Themes of the English “Dolnik”’, Poetics Today 16.3 (Autumn 1995), 493–522 (496).

32 Derek Attridge, ‘The Case for the English Dolnik; or, How Not to Introduce Prosody’, Poetics Today 33.1 (2012), 1–26 (8).

33 On the frequency gap between ‘still iambic’ and ‘something else’, see Marina Tarlinskaja, ‘Beyond “Loose Iamb”’, 504–6.

34 Russell K. Alspach, ‘Some Sources of Yeats’s The Wanderings of Oisin’, PMLA 58.3 (Sept. 1943), 849–66.

35 John O’Daly, ‘Introduction’, Transactions of the Ossianic Society for the Year 1856, Vol. 4 (Dublin, 1859), xxi–xxxii (xxxii).

36 And which he acknowledges only obliquely in the 1895 glossary to the poem. Russell K. Alspach, ‘Some Sources of Yeats’s The Wanderings of Oisin’, 849–50.

37 Warwick Gould, ‘Lips and Ships, Peers and Tears: Lacrimae Rerum and Tragic Joy’, YA18 (2013), 15–56 (32).

38 Muiris O Rochain claims that the form of the poem is ‘in a form described as rannaíocht mhór’, 64, which is a syllabic metre.

39 ‘The Land of Youth’, ed. by Bryan O’Looney, Transactions of the Ossianic Society for the Year 1856 (Dublin, 1859), 227–80 (237).

40 ‘The Lay of Oisin on the Land of Youth’ is set as prose in Alfred Graves, The Irish Fairy Book (London: t. f. Unwin, 1909), 71–84.

41 This text published by Yeats as The Wanderings of Oisin and Other Poems (London: Kegan Paul, 1889), 21.

42 Michael J. Sidnell, ‘j. b. Yeats’s Marginalia in The Wanderings of Oisin and Other Poems’, YA13 265–91 (269).

43 See George Bornstein’s transcription (page not reproduced in facsimile) in Bornstein ed., The Early Poetry, Vol. 2, ‘The Wanderings of Oisinand Other Early Poems to 1895: Manuscript Materials (Ithaca: Cornell University Press, 1994), 141; w. b. Yeats, Poems (London: T. Fisher Unwin, 1895), 27. Why this text is different from that listed in the Variorum as ‘9’ (1895 British and American printing) is unclear to me.

44 William Wordsworth, letter to John Thelwall, 1804, quoted by Peter McDonald in Sounds Intentions: The Workings of Rhyme in Nineteenth-Century Poetry (Oxford: Oxford University Press, 2012), 78.

45 In John Henry Newbolt, My World as in My Time: Memoirs of Sir Henry Newbolt (London: Faber & Faber, 1932), 194.

46 Letter, June 8, 1901, in Donald E. Stafford ed., The Selected Letters of Robert Bridges, Vol. 1 (Newark: University of Delaware Press, 1983), 382.

47 g. m. Hopkins, ‘The Wreck of the Deutschland’, in Robert Bridges ed., Poems of Gerard Manley Hopkins, 2nd ed. (London: Humphrey Milford, 1930), 11–23 (16).

48 Letter 38, August 21, 1877, in c. c. Abbott ed., The Letters of Gerard Manley Hopkins to Robert Bridges (London: Oxford University Press, 1935), 46.

49 Joseph J. Feeney, The Playfulness of Gerard Manley Hopkins (Aldershot: Ashgate, 2008), 74.

50 The rule is usually stated the other way round, so Guglielmo Cinque argues ‘stress prominence in a phrase is a mere reflection of the depth of embedding’, ‘A Null Theory of Phrase and Compound Stress’, Linguistic Inquiry 24.2 (Spring 1993), 239–97 (245).

51 Hopkins’ seven-page selection is sandwiched between Robert Bridges and William Ernest Henley (obmv 17–23).

52 Walt Whitman, November Boughs (Philadelphia: David McKay, 1888), 65.

53 t. s. Eliot, ‘Swinburne as Poet’, The Sacred Wood (London: Methuen, 1920), 131–36.

54 Ezra Pound, review of Responsibilities, Poetry 2.4 (May 1914), 64–69 (65).

55 Warwick Gould, ‘The Unknown Masterpiece’, in Andrew Gibson ed., Pound in Multiple Perspectives (Basingstoke: Macmillan, 1993), 40–92 (45).

56 Charles Williams to w. b. Yeats, 11 October 1935, quoted in John Stallworthy, ‘Yeats as Anthologist’, in A. Norman Jeffares and k. g. w. Cross eds., In Excited Reverie: A Centenary Tribute to William Butler Yeats, 1865–1939 (London: Macmillan, 1965), 171–92 (180–81). On Charles Williams work for Oxford University Press on this anthology, see below, 327-35.

57 Robert Alden Rubin, ‘Some Heroic Discipline: William Butler Yeats and the Oxford Book of Modern Verse’, unpublished Ph.D. dissertation, University of North Carolina at Chapel Hill, 2011, 65.

58 Daniel Albright’s word in Quantum Poetics: Yeats, Pound, Eliot, and the Science of Modernism (Cambridge: Cambridge University Press, 1997), 64.

59 Roy Foster, w. b. Yeats: A Life II, The Arch Poet, 1935–1939 (Oxford: Oxford University Press, 2003), 556.

60 Jerome McGann, Black Riders: The Visible Language of Modernism (Princeton: Princeton University Press, 1993), 81.

61 t. s. Eliot, ‘American Literature and the American Language’, 58.

62 In the 1800 version of a heavily reworked passage, Wordsworth argues that ‘not only the language of a large portion of every good poem, even of the most elevated character, must necessarily, except with reference to the metre, in no respect differ from that of good prose’, in Stephen Gill ed., William Wordsworth: 21st Century Oxford Authors (Oxford: Oxford University Press, 2010), 63.

63 Quoted by Thomas Parkinson, w. b. Yeats: The Later Poetry, 185.

64 j. s. Mill continues ‘What we have said to ourselves we may tell to others afterwards; what we have said or done in solitude we may voluntarily reproduce when we know that other eyes are upon us. But no trace of consciousness that, any eyes are upon us, must be visible in the work itself’. See ‘Thoughts on Poetry and its Varieties’, Dissertations and Discussions, 2nd ed. (1867) and reprints, I, 51–76 at 58. The passage occurs in the extended discussion of the difference between poetry and eloquence: ‘eloquence is heard, poetry in overheard’(57). The title of the essay is footnoted, ‘Monthly Repository, January and October 1833’. The essay had been republished from ‘What Is Poetry?’ Monthly Repository, 7 (Jan., 1833), 60–70; and ‘The Two Kinds of Poetry’, ibid., 7 (Oct., 1833), 714–24. Both had been signed: ‘Antiquus’. See John M. Robson and Jack Stillinger (eds.), The Collected Works of John Stuart Mill, Volume I—Autobiography and Literary Essays [1824] (Toronto: University of Toronto Press; London: Routledge & Kegan Paul, 1981), 349. From John Butler Yeats’s unpublished ‘Memoirs’, it becomes clear that he thought himself a ‘disciple’ of Mill: it is therefore probable that Dissertations and Discussions would have been among the family’s books. See William M. Murphy, Prodigal Father: The Life of John Butler Yeats, 1839–1922 (Ithaca and London: Cornell University Press, 1978), passim, esp. 65.

65 t. s. Eliot, ‘Tradition and the Individual Talent’, The Sacred Wood, 42–53 (52–53).

66 t. s. Eliot, ‘Ulysses, Order, and Myth’, The Dial 75.5 (November 1923), 480–83; ‘Tradition and the Individual Talent’, The Sacred Wood, 42–53 (43).

67 t. s. Eliot, ‘Reflections on Vers Libre’, The New Statesman, 3 March 1917, https://www.newstatesman.com/culture/culture/2013/05/t-s-eliot-reflections-vers-libre

68 Stravinsky even judged that Wagner’s opera ‘must have been one of the most passionate experiences in his life’, in Themes and Episodes (New York: Knopf, 1966), 125.

69 ‘Vacillation’ IV (VP 501); ‘Under Ben Bulben’ (VP 640); ‘The Tower’ (VP 411).

70 Eliot writes: ‘A good poet will usually borrow from authors remote in time, or alien in language, or diverse in interest’, ‘Philip Massinger’, The Sacred Wood, 112–30 (114).

71 Pound describes his sense that the ‘dilutation of vers libre’ had gone ‘too far’ in ‘Harold Monro’, published in Eliot’s Criterion 11.45 (July 1932), 581–92.

72 See Chris Golston and Tomas Riad, ‘The Phonology of Classical Greek Meter’, Linguistics 38.1 (2000), 99–167 (103).

73 Ezra Pound, ABC of Reading (New York: New Directions, 1934), 35.

74 j. c. c. Mays gives the history in ‘Coleridge and Yeats: The Romantic Voice’, Variants 6 (2007), 65–84 (76).

75 From, respectively: the first lines of ‘A Prayer on Going into my House’ (VP 371) and ‘Easter, 1916’ (VP 391); the last line of ‘Among School Children’ (VP 446); line 9 of ‘The Circus Animals’ Desertion’ (VP 629); line 41 and line 8 of ‘Coole Park and Ballylee’ (VP 490–91).

76 Roman Jakobson’s slightly longwinded analysis of the two versions agrees that the 1925 version ‘contains a higher number of finites and, at the same time, exhibits a greater grammatical uniformity in their use’, Verbal Art, Verbal Sign, Verbal Time, ed. by Krystyna Pomorska and Stephen Rudy (Oxford: Blackwell, 1985), 79–107 (91).

77 For Louis MacNeice, this is not an advantage: ‘But perhaps this poem ought to be languid’. The Poetry of w. b. Yeats (London and New York: Oxford University Press, 1941), 70–71.

78 Warwick Gould discusses why Yeats might have made the Trojan frame of reference clearer, ‘Writing the Life of the Text’, 25.

79 In Poetry 2.1 (April 1913), the poem is printed with a colon at the end of the first line; a semi-colon was introduced in Lustra (London: Elkin Matthews, 1916).

80 Ezra Pound, Canto 1, The Cantos of Ezra Pound, 7.

81 On superposition, see Ezra Pound, Gaudier-Brzeska (London: John Lane, 1916), 103.

82 This aversion from the indicative is seen even in Eliot’s epigraphs: ‘o quam te memorem virgo’ in ‘La Figlia Che Piange’ (the subjunctive), published, like ‘The Love Song of J. Alfred Prufrock’, in Prufrock and Other Observations (London: The Egoist Press, 1917).

83 w. h. Auden, ‘The Public v. the Late Mr. William Butler Yeats’ [1939], in Edward Mendelson ed., w. h. Auden: Prose, Vol. 2, 1939–1948 (London: Faber & Faber, 2002), 3–7 (7). And in the elegy, ‘For poetry makes nothing happen’, ‘In Memory of w. b. Yeats’, in Edward Mendelson ed., w. h. Auden: Selected Poems (London: Faber & Faber, 1979), 80–83 (82).

84 In Yeats’s drafts, we very often find a list of complete phrases running down the page, connected, like the free-verse ‘Mona Lisa’, through anaphora: see, e.g. the draft for the ‘honey of generation’ stanza in ‘Among School Children’, in Richard J. Finneran et. al. eds., The Tower (1928): Manuscript Materials (Ithaca and London: Cornell University Press, 2007), 369.

Auteur

Associate Professor of English at New College, Oxford. Her first book, The Work of Revision (2013), was a comparative study of modernist writers’ practices of writing and redrafting, with a particular focus on the use of typescript. She was awarded a Leverhulme Prize in 2013 to write a book on the theory, polemic, and practice of free verse from Wordsworth to the present. In fact, as this article on Yeats (written at the beginning of research for the book) begins to suggest, the equation between freedom and prosodic irregularity is not always as simple as it may seem. Her debut poetry collection, Three Poems, will be published by Faber & Faber in 2018.

Acheter