A Poem for all Seasons: Yeats, Meaning, and the Publishing History of 'The Lake Isle of Innisfree' in the 1890s.

MCDONALD, PETER D.

Books are not innocent. They have designs on the texts they contain and on their readers. For Yeats's more discerning commentators this much has been evident since the beginning of his publishing career. With no doubt Poems (1895) on his mind, T. W. Rolleston remarked as early as 1900:

Few poets have revised and retouched their work more than Mr. Yeats, and this may perhaps be one cause of the singular unity of the impression which it leaves upon the mind. In the final edition of his poems, where much is altered and much early work struck out altogether, one sees naturally but little sign of the immature and experimental stages which every poet must go through. He appears to have struck the rock, and the water flowed; we do not see it led with pain and toil from distant sources, through miry channels, and by feeble streamlets into its true bed.[1]

While many might now feel less confident about the 'singular unity' of Yeats's early work, few would dispute Rolleston's claims about the artful effects of his books. Indeed, taking up what has now become a dominant theme in Yeats studies, George Bornstein develops Rolleston's early impressions in his essay 'Remaking Himself: Yeats's Revisions of His Early Canon'. Like Rolleston, he argues that Yeats's impulse to revision involved much more than the stylistic alteration of individual poems. Tracing the process back to The Wanderings of Oisin (1889), he shows it also entailed 'a complex and deliberate reworking of the larger divisions into which he organized his work' which amounted to 'a careful construction both of a canon and of an appropriate poetic self '.2 A book such as Poems (1895) is, he suggests, not simply a calculated selection of revised texts arranged in a particular order, but a portrait of the artist as a young man. Of course, like Rolleston, he argues that the result was not a faithful record of a young poet's tortuous development, but a construct projecting an idealized public self-image. As he puts it, the principles of selection at work in the first three volumes, Wanderings (1889), The Countess Kathleen and Various Legends and Lyrics (1892) and Poems (1895), reveal 'a strategy more aesthetic than autobiographic' (p. 340). As Yeats's contemporaries and modern scholars have both recognized, then, books for him were not mere modes of transmission. They were opportunities for poetic self-creation.

Taking his cue from Hugh Kenner's remark of the late 1950s that Yeats 'didn't accumulate poems, he wrote books', Bornstein makes a compelling case for a bibliographically informed style of critical interpretation which resists the allure of the text's ideality.[3] He reminds us that we do not read, say, 'The Lake Isle of Innisfree' in Poems (1895). We read it as the 'The Lake Isle' of Poems (1895). Its specific placement in the book as a whole, and among the contending ensemble of poems for the first time entitled 'The Rose'; its relations to the neo-primitive transcendentalism of the volume's 'Prefatory Poem: To Some I have Talked with by the Fire', and to the more equivocal attitudes adopted in the poems that open and close 'The Rose', respectively, 'To the Rose Upon the Rood of Time' and 'To Ireland in the Coming Times'; and the implications for it of Yeats's explanation in the prose preface that the poems in 'The Rose' represent 'the only pathway whereon he can hope to see with his own eyes the Eternal Rose of Beauty and Peace' -- all this specifically bibliographical and paratextual scaffolding, which stands at what Genette would call the threshold of interpretation, influences, more or less effectively, the reader's reception of the text.[4] More generally, demonstrating that books and not simply texts have implied authors, Bornstein suggests the reader is also implicitly guided by the image of the 'fictive Yeats constructed by the 1895 Poems', a figure who emerges 'as an Irish poet devoted to tradition and rural Ireland, to legendary heroes and to the peasantry, and to esoteric pursuit of ideal Beauty', and who shuns 'contemporary causes, urban life, and the emerging middle classes'. This implied author, he adds, bears little resemblance to the 'historical young Yeats', who was, among other things, thoroughly metropolitan, worldly, and politically engaged.[5] In his conclusion, Bornstein pauses to reflect on the wider scholarly and theoretical implications of his analysis:

In remaking his poems Yeats remade himself, as his quatrain on the subject reminds us. In so doing, he offers us today a middle ground between the old fixed, stable author and fixed, stable text on the one hand, and the elimination of the author and substitution of endless textual free-play on the other. For what Yeats finally created was a process rather than a product, in which a successive but finite remaking of texts and selves substitutes for the fixing of them.[6]

Here Bornstein offers an inviting vision of a world beyond the cold war between history and theory, bibliography and literary interpretation, a world in which authors are neither worryingly absent nor oppressively present. Far from being a stable guarantor of the text's fixed meaning, Yeats appears now as the protean agent of its rich but finite multivocality.

But can the conflicts and divisions within literary studies be so easily and neatly resolved? And does Bornstein's argument really lead forward into a promised land beyond tired polemics about the 'death of the author'? Attractive as it is, his vision loses some of its conciliatory force, I would suggest, when we set it against his own methodological practice. Most importantly, by focusing on the ways in which what might for now be called Yeats's 'own books' affect meaning, he introduces the radical idea of the decentred author while remaining firmly committed to the traditional preoccupation with authorial control. True, a more complex image of the author emerges from his investigation which uncovers the strategies the 1890s Yeats adopted 'first to create a self-image' and then to devise 'a foil to the allegedly more politically and socially-involved Yeats of the early twentieth century' (p. 339). But, in the end, all we have is what Yeats, the less integral but still privileged author, created. The modes of publication, and hence the possible meanings and selves, over which he had less direct influence are inevitably obscured. In short, for Bornstein, the meanings of 'The Lake Isle of Innisfree', to stay with my example, are relatively finite because Yeats is still firmly in charge. What I would like to ask in this paper is what happens to the meanings of the poem, and indeed to one's understanding of Yeats's position in the literary field of the 1890s, if one takes a more comprehensive look at the first decade of 'The Lake Isle's' publishing history in Britain. Bearing in mind that it appeared not only in The Countess Kathleen (1892) and in Poems (1895 and 1899), but in W. E. Henley's National Observer (13 December 1890), The Book of the Rhymers' Club (1892), the Religious Tract Society's Leisure Hour (August 1896), William and Elizabeth Sharp's anthology Lyra Celtica (1896), and Stopford Brooke and Rolleston's A Treasury of Irish Poetry in the English Tongue (1900), can one remain as confident in the relative stability of its meanings as Bornstein believes? Indeed, is it not possible that the image Yeats wished to project in his 'own books' was not just that he was a particular kind of author, but that he was one who had some control over his text's reception, a privilege not as easily accorded him in other publishing formats?

In early November 1890, he sent 'The Lake Isle', along with 'The Old Pensioner', directly to Henley, saying 'I enclose two short poems for Scots Observer if suitable.'[7] Though this marked the beginning of the poem's complex public life, he had been sitting on the manuscript for almost two years. As he told Katherine Tynan in a letter of 21 December 1888, in which he included an incomplete and rather laboured early draft, the poem was a by-product of his second work of prose fiction, John Sherman (1891), which he had begun earlier that year.

In my story I make one of the charecters [sic] when ever he is in trouble long to go away and live alone on that Island -- an old day dream of my own. Thinking over his feelings I made these verses about them -- (CL, i, 120-21)

The early two-stanza draft is, in effect, a lyric version of Sherman's narrative recollection of his childhood fantasy (Part iv, Section iv). Not surprisingly, in relocating these 'feelings' from one genre to another, Yeats changed not only the linguistic register and the literary form in which they were expressed, but the nature of their object. In keeping with the practice of the novel as a whole (all the Irish place names are fictionalized) the actual Innisfree becomes the invented Inniscrewin, and, given this was Yeats's first attempt at a bildungsroman in a realist mode, the narrator represents Sherman's longings in a studiedly plain, even colloquial style.

Often when life and its difficulties had seemed to him like the lessons of some elder boy given to a younger by mistake, it had seemed good to dream of going away to that islet and building a wooden hut there and burning a few years out, rowing to and fro, fishing, or lying on the island slopes by day, and listening at night to the ripple of the water and the quivering of the bushes -- full always of unknown creatures -- and going out at morning to see the island's edge marked by the feet of birds.[8]

Though a solitary place apart, and despite the hint of 'unknown creatures', the Inniscrewin for which the young Sherman longs is firmly of this world. By contrast, Innisfree, the object of the lyric speaker's desire, is, even in the early draft, a far more magical place. Though more concretely realized (the birds have become linnets, the creatures bees and crickets) it is also now a more mysterious refuge where 'peace will come down dropping slow | Dropping from the veils of the morning'; and where Sherman's prosaically contrasting day and night have evolved into an altogether richer chronology encompassing 'dawn', 'morning', 'noontide', 'evening', and 'midnight', each of which offers its own enigmatic consolations. Moreover, while Sherman simply dreams 'of going away to that islet', the lyric speaker's statement of intent already has the oratorical conviction of the Biblical 'I will arise and go', a piece of stagy intertextuality Yeats would later come to regret. As Yeats insisted in his letter to Tynan, the Innisfree he now had in mind was 'a little rocky Island with a legended past' (CL, i, 120).

The version Henley received in late 1890 was radically improved and expanded. The musically and practically unappealing 'dwelling of wattles -- of woven wattles and wood work made' has become a 'small cabin [. . .] of clay and wattles made'; more plausibly, noon now has a 'purple glow' and 'midnight's all a glimmer', not the other way round; and, most importantly, a third stanza, sketching in the speaker's immediate urban context and the profoundly Romantic source of his longing, has been added. Assuming Henley did not tamper with the manuscript (though Yeats claimed he 're-wrote my poems as he re-wrote the early verse of Kipling', there is no evidence to suggest he did in this case), the version he saw, and which subsequently appeared in the National Observer (by then its name had changed) for 13 December 1890, was more or less final.[9] In fact, in the course of its long public life, the verbal text of 'The Lake Isle', the 121 words that constitute the literary critic's traditional object of study, remained for Yeats unusually stable. The text we encounter in the Observer is not substantially different from the one we find in Poems (1895) or, indeed, in Allt and Alspach's Variorum Edition (p. 117).[10] Yet, if the text had already achieved some measure of closure, its meanings and uses, for Yeats and others, had become an open question.

By mid-December 1890, Yeats had already consolidated his important early relationship with Henley and, with three poems, three reviews, and four articles, established himself as one of the Observer's major literary contributors. Only two months after the appearance of his first article in the issue for 2 March 1889, he was on the weekly's promotional list of 'signing contributors', among such leading literary figures as William Archer, Edmund Gosse, Andrew Lang, Richard Garnett, and Robert Louis Stevenson.[11] Yet from the start his association with Henley was incongruous. As he admitted in the first draft of his autobiography, written more than a decade after Henley's death in 1903, the political and even literary gulf that divided them was enormous.

He despised all of Rossetti but 'The Blessed Damozel', never spoke of Pater and probably disliked him, praised Impressionist painting that still meant nothing to me, was a romantic but not of my school, and founded [the] declamatory school of imperialist journalism.[12]

In the final analysis, however, none of this really seemed to matter. 'I always felt that I suffered no loss of dignity from his opposition to all I hoped for Ireland' and that it 'made no difference that I dissented from his judgement of other men.' The advantages of being one of 'Henley's young men' made the gulf between them manageable, if not irrelevant; besides, as he always signed his contributions to the Observer he could, as he put it, 'go my own road in some measure'.[13] For one thing, at a time of real financial need, Henley was, as he later recalled, 'my chief employer', paying probably as much as [pound]1 1s.0d. for a short poem like 'A Cradle Song' and [pound]1 a column for reviews and articles.[14] Moreover, as an established man of letters, Henley used his connections and influence to try, not always successfully, to secure him other paid work, and he introduced him to a large, though for Yeats not always congenial, network of writers, critics, and journalists.[15] Yet these financial and social benefits, significant as they were in themselves, were also only a practical expression of Henley's real importance for the young Yeats. What he offered him, above all, was the kind of assurance about his literary quality and status he sought at that stage in his career. As Yeats put it in a letter to Tynan, 'one of his good points is his sympathy with young writers' (CL, i, 252). Indeed, given his later thoughts about the meaning of their association, it could be argued that Henley played a major part in the formation of his early literary identity. Speculating about Henley's charismatic 'hold' in the draft autobiography, he felt it 'was perhaps that he was never deceived about his taste, that he wished one well, and could not flatter'; or, again, 'I was drawn to him also, I doubt not, by his aristocratic attitudes, his hatred of the crowd and of that logical realism which is but popular oratory'.[16] On this view, Henley was a distinguished critic whose judgement he could trust, and a prominent man of letters who bolstered his own high-minded anti-populism. A few years later, just before his own father's death in February 1922, he added a further psychological dimension to this analysis. Recalling his early career, and after referring specifically to his own father's 'infirmity of will', he noted 'I had to escape this family drifting, innocent, & helpless, & the need for that drew me to dominating men like Henley & Morris' (CL, i, 520). Part alternative father figure, then, part purist literary role model, and part established impresario, Henley offered Yeats the cultural legitimization he sought as a young Irish writer on the make in the English 'Republic of Letters'. In what must remain one of the richer ironies of literary history, that is, Henley, the jingoistic author of 'A Song of England', helped fashion Yeats's 1890s Irish poetic self by reflecting back to him the literary self-image he desired.

The other person involved in this dialectical process of self-creation was, of course, Yeats himself. Indeed, any reading of 'The Lake Isle' in the context of the Observer must begin with the fact that the poem was, for Yeats, part of a conscious self-promotional strategy targeted at Henley. Though his general sense of Henley's position, gleaned no doubt from various sources (his own reading, hearsay, and so on) would have enabled him to pursue this odd 'elective affinity', he also had Henley's unsigned review of The Wanderings of Oisin and Other Poems to guide him. The image Yeats saw of himself in the review, which appeared in the Observer for 9 March 1889, was extremely flattering (he told Tynan it was a 'splendid article') but it was also tellingly incomplete (CL, i, 152).[17] Despite the varied nature of his first major book, Henley emphasized Yeats's Irishness by entitling the review 'A New Irish Poet' and by saying his 'verses have the wilding charm, the wayward grace touched with elfishness, characteristic of true Irish song'.[18] At the same time, he placed him firmly in an English literary tradition by comparing him favourably to Shelley, Tennyson, and Morris, and, more importantly, he made sure that his Irishness would be understood in a particular way. Given his own commitment to the late-Victorian cult of manliness, he was, for instance, eager to report that, for all his admirable interest in the 'rainbow-coloured lands of fantasy', this new Irish poet was not an effeminate Arnoldian Celt: 'There are lines in which the tones of war and hunting and heroic comradeship ring out bravely; the writer can stir the blood as well as beguile and lull with sensuous dreams' (pp. 446-47). But, in case this gave the impression that the Observer and its editor were backing a manly but also possibly insurrectionary Irish poet (Arnold's feminine Celts were, of course, famously 'ineffectual in politics')19 he hastened to add that none of the poems had any contemporary political relevance. Much to Yeats's delight, he praised, among others, Wanderings, 'King Goll', 'The Meditation of the Old Fisherman', and 'The Song of the Last Arcadian', but he kept silent about 'How Ferencz Renyi Kept Silent' and insisted that the Fenians mentioned in Wanderings were simply part of the mythic past. In his conclusion, he invited his readers to welcome a poet who 'can speak out with the right heroic accent, and kindle the blood with tales of the (strictly historical) deeds that were done in the brave old days "When the Fenians made foray at the morning with Bran, Sgeolan, Lomair" ' (p. 447).

With this clear opinion at hand, Yeats understandably ensured that Henley got the first option on only a very select range of poems. In the five years from 1890 to 1894, during which his sixteen Observer poems appeared, he also contributed eleven others to a wide range of periodicals, including the Parnellite United Ireland (the Observer's Irish bete noire), the Nationalist Irish Weekly Independent, the new populist London literary review The Bookman, and the Religious Tract Society's illustrated family monthly Leisure Hour.[20] Not all of these would have suited Henley's taste. Some, such as 'In the Firelight' (Leisure Hour, March 1891), were immature, not to say mawkish, but others, such as 'Mourn -- and Then Onward!' (United Ireland, 10 October 1891), the elegy for Parnell, and the Nationalist 'Ballad of Earl Paul' (Irish Weekly Independent, 8 April 1893), were too overtly political. Just how conscious Yeats's careful selection of poems for Henley was can be seen from one of his remarks to Tynan. 'You should not have sent the poem [most probably 'The Wild Geese: A Lament for the Irish Jacobites'] to the Scots Observer,' he told her in a letter of May 1890, 'it was too political -- Irish exiles are out of their range I think' (CL, i, 217). He then immediately went on to say he had just sent Henley 'Father Gilligan', the wry folk ballad which, in the Scots Observer version, came with the authenticating paratext: 'A Legend told by the People of Castleisland, Kerry'.[21] Clearly, in Yeats's view, disillusioned priests miraculously saved from disgrace and memorialized by the 'people', not revolutionary exiles, represented an acceptable version of Ireland and Irishness for Henley's Observer, a version he was quite willing to help promote. As importantly, however, the intersection between the set of Observer poems and the 1895 canon suggests Yeats treated Henley's weekly as a nursery for the poetic self he created in his first collected volume. Of the thirteen poems from the key section 'The Rose' that first appeared in periodicals (eight never did), twelve came from the Observer. Indeed, while only less than half (five out of eleven) of the poems not published by Henley in the early 1890s made it into any of Yeats's books ('Mourn -- and Then Onward!' and 'The Ballad of Earl Paul', for instance, were excluded) versions of all his Observer poems, as well as the two he contributed to Henley's next venture The New Review, became and remained part of the canon. When one thinks about Yeats's protean existence in the early 1890s, then, one should not focus simply on the contrast between his 'poetic' and his 'historical' selves, as Bornstein puts it, but on the more complex interrelationship between the various poetic selves embodied in the periodicals and the more selective construct that emerges in the books. With the Observer Yeats clearly had designs not only on Henley, but on posterity.

The poetic self he revealed to Henley by selecting suitable poems was also closely related to the implied authorial self he projected in his more consciously inflected prose contributions. In the early articles and reviews, which Henley actively encouraged, he presented himself as a distinctly Irish writer who celebrated the peasantry, folklore, and rural life of the West of Ireland, Sligo in particular, and who seemingly had more interest in the 'bloodless, dim nation' than in questions of Irish nationhood.[22] As he remarked in Autobiographies, 'that I might avoid unacceptable opinions, I wrote nothing but ghost or fairy stories' (p. 160). While his first articles show him to have been especially responsive to Henley's initial Scottish and largely Protestant readership, he also assumed he was writing for an educated, urban, sceptical, generally non-Irish audience, those who 'listen to eloquent speaking, go to discussions, read books and write them, settle all the affairs of the universe', a fair estimation of Henley's intended readership. [23] To them he explained:

It is far easier to be sensible in cities than in many country places I could tell you off [sic]. When one walks on those grey roads at evening by the scented elder-bushes of the white cottages, watching the faint mountains gathering the clouds upon their heads, one all too readily seems to discover, beyond the thin cobweb veil of the senses, those creatures, the goblins, hurrying from the white square stone door to the north, or from the Heart Lake in the south.[24]

'To the wise peasant', he added later, 'the green hills and woods round him are full of never-fading mystery'.[25] For Henley and the readers of the Observer, then, he defined himself, as much in these early articles as in a poem such as 'The Lake Isle', as a champion of the West of Ireland figured as a lost paradise in which a happily primitive, Irish culture, transcendental in outlook though still rooted in the natural world, held out against the disenchantments of modernity. Contemporary political realities formed no part of this mythic vision, though, once, he did allow himself, or he was allowed, to reflect that for the Irish peasant Heaven was much like earth, except that 'now and then a landlord or an agent or a gauger will go by begging his bread, to show how God divides the righteous from the unrighteous'.[26] This was in a signed review of Lady Wilde's Ancient Cures, Charms, and Usages of Ireland (1890) in which he none the less remained silent about her Nationalist sympathies. Keeping up this intricate balance between his own diverse allegiances and Henley's was, of course, in his interests, but it was not always easy, and on some occasions it defeated him. He found it impossible, for instance, to review Ellen O'Leary's collection Lays of Country, Home and Friends (1891), for 'so ultra-Tory a paper', and about an article on the 'Revival of Irish Literature', he told Henley the 'subject was thorney [sic] & I fear I may not in making it suitable for my own purposes have made it suitable for you' (CL, i, 250, 521). Henley evidently agreed as it never appeared in the Observer.

The extent to which Yeats marketed himself for Henley's Observer might suggest that the differences in the degree of influence he had over his reception in periodicals as opposed to his 'own books' was negligible. Yet the realities of periodical publication also inevitably complicated matters. This was not only because a strong editor, as was Henley, influenced his output by encouraging, commissioning and/or revising particular texts. Such interventions were, after all, not the prerogative of periodical editors alone. Rather, it had to do with the effect the specific conditions of periodical publication had on his own stratagems, in particular, the scattered and relatively unadorned appearance of the poems, and the juxtaposition of miscellaneous textual matter. If books are, on Genette's definition, the natural home of the paratext, periodicals are a mode in which, for those who need a single term, the circumtext predominates. This is not to say Yeats could not add certain paratextual elements to the periodical versions of his poems. Like 'Father Gilligan', 'The White Birds', for instance, appeared in the National Observer, but not in any of the books, with a brief explanatory note inserted below the title:

(The birds of fairyland are said to be white as snow. The Danaan Islands are the islands of the fairies.)[27]

If the paratext is, as Genette claims, 'always the conveyor of a commentary that is authorial or more or less legitimated by the author [. . .] at the service of a better reception for the text', then this supplementary note, designed to guide readers not familiar with Irish legends and folklore, and set apart typographically, is, strictly speaking, paratextual.[28] Yet in the periodical context these important occasional notes, like the poems themselves, were set within a vast body of miscellaneous matter, the effects of which were neither as clear nor as manageable. To analyse these, it would be useful to distinguish between what could be called the authorial and the non-authorial circumtext. As I have shown, the meaning of 'The Lake Isle' in the context of the Observer is neither fully determined by, nor separable from an understanding of its function as a component of the mutually acceptable version of Ireland and Irishness negotiated between Yeats and Henley in the early 1890s. Here all the texts, poems, stories, reviews, and articles, authored by Yeats and authorized by Henley are relevant to the reading. Together they constitute the authorial circumtext: 'authorial' because Yeats still had some measure of control over it, and hence over his self-presentation in the Observer, 'circumtextual' because it is, unlike the paratext, only indirectly and implicitly 'at the service of a better reception' for the poem. The non-authorial circumtext, in contrast, is by definition an editorial matter, but it does not follow from this that it has no bearing on the reception of 'The Lake Isle'. In theory, the persons responsible for the contents and/or layout of the periodical (proprietor, editor, designer, printer, illustrator, and so on) may use the opportunity to create a commentary which may affect, but not necessarily enhance, a particular text's reception. The most obvious example here would be the addition of accompanying illustrations, a point which is, as Warwick Gould has shown, particularly well demonstrated by the one 1890s version of 'The Lake Isle' I shall not be discussing in detail.[29] The version of the poem which appeared in The Leisure Hour for August 1896 was accompanied by a full-page illustration by the poet's father, John Butler Yeats, who told his daughter in a letter of April that year:

I drew an Irish longhaired chief in chain armour stalking slowly along using his spear as a staff and leaning heavily on it, an Irish wolf hound following him, its tail between its legs, the place a wood of oak trees black with forest darkness.[30]

This was, on the face of it, a perplexing illustration for a poem which has nothing to say about primitive chieftains, wolf hounds, or dark oak forests. Yet, as Gould rightly points out, the image does offer the reader an illuminating, if lateral-minded, commentary on the text. Most generally, it emphasizes, as the young Yeats would have appreciated, the legendary associations Innisfree had for both him and his father; and, more specifically, by placing the figure of the chieftain at the centre of things, it reveals both Yeats's indebtedness to William Wood-Martin's History of Sligo, County and Town (1882-92): there Wood-Martin described the island's connection to the legend of Free, an ill-fated young warrior, a story Yeats would use in 'The Danaan Quicken Tree' in 1893. Ineffect, the illustration as non-authorial circumtext says more about the poem's intertextual origins and implicit mythopoeic significance than the young Yeats himself had allowed. Then again, the non-authorial circumtext may have no planned effect, but this does not preclude the possibility of its still having some unintended influence, fortunate or unfortunate. This is especially true of certain materials surrounding 'The Lake Isle' in the National Observer.

For various reasons, some more obvious than others, Yeats, as we have seen, excluded or played down any explicit political references in his contributions to the stridently Unionist Observer. Yet, for the same reasons, it is difficult to read 'The Lake Isle' in its periodical context without being aware of contemporary political issues. This is the most telling effect of the non-authorial circumtext that begins with Frederick Greenwood's article entitled 'Sedition-Sops for Ireland' that appears alongside 'The Lake Isle', extends to the issue for 13 December 1890 which, given the recent rupture in the Irish Parliamentary Party, devoted twelve-and-a-half columns, almost half the issue, to Irish matters, and finally encompasses the entire run in which the West of Ireland, in particular, figures regularly as a desperate rural backwater populated by an inferior race. In fact, given its political outlook and Henley's sympathy towards Yeats's carefully managed project, the Observer not unexpectedly offered its readers two competing images of Ireland, even of the West of Ireland: Yeats's celebratory mythopoeic version, and its own politically expedient 'ultra-Tory' version. In the contemporary Conservative debates about Balfour's new pacificatory Land Purchase Bill, for instance, which the Observer supported but which Greenwood, the fervid but always independently-minded Unionist, criticized in his article, the West of Ireland figured prominently. As the issue for 31 August 1889 noted, it was deemed to be a special region beyond the reach of any land reforms:

There is a large tract of country in Ireland, comprising West Donegal, the greater part of Mayo and Sligo, Kerry, and a portion of Galway, which no Land Act of the kind hitherto passed can ever really benefit, or even seriously affect. It is no exaggeration to say that, political expediency apart, it would distinctively pay the Government to buy up every tenant within the whole of this area, and provide him with a pension for life conditional upon his consenting to remove from his present habitation.[31]

The problem did not simply have to do with the poverty and over-crowding in these areas officially designated 'congested districts', or with the fact that 'the population was not able to draw from the soil a safe and sufficient living', but with the supposed character of the 'Irish peasant' in the region.[32] While the Observer supported Balfour's plan to provide special State relief to the area, though it predictably preferred to encourage 'emigration or migration' and 'the regeneration of the spirit of self-help', it also endorsed his view that 'the poverty of the West Coast of Ireland is mainly due to the idleness, the ineptitude, the lazy indifference of the people to the wretchedness in which they live'.[33] True, this did not, in its view, really mean they were all that different from the Irish 'national type'. Characterized as it was by its 'subjection to current needs', its 'utter thriftlessness and waywardness of mind', and its political ineptitude ('not the ruling but the ruled') the 'peasants' of the West simply reinforced the Observer's anti-Irish stereotype.[34] But it was none the less the case that the problems of the West were exacerbated by the fact that region was one of the key refuges of the 'unadulterated Irish Celt'.

For, after all, the history of the settlement of this country is the history of the extrusion of the Celt, westwards and northwards, by the Norman, Saxon, and Dane, until he found security in the impregnable highlands of Scotland, Wales, and West and South Ireland. If such facts do not cover some real difference in the fibre and worth of the contending races, they constitute a class of phenomena wholly anomalous and inexplicable.[35]

Still it was unfortunate that the 'mere Irish' did not 'disappear before the colonists, as the Maoris have done in recent times in New Zealand'. 'It is not wonderful', another of Henley's observers added, 'if the wearied Saxon have sometimes heaved a sigh over the strange partiality of Nature which dooms the noble savage of other continents but protects that other kind of wild man he grows at home.'[36]

Reading 'The Lake Isle' of the Observer is, then, in part, a matter of understanding its relations to the authorial circumtext. While its elevated language and sonorous metre create an atmosphere all of its own, its representation of Innisfree as a magical primitivist idyll which answers to the modern, urban speaker's most profound, unconditioned longings is incidentally informed by the mythic image of the West of Ireland Yeats projected in his other contributions. Reading it in this context involves, I have suggested, not only a textual analysis of these contributions and their interrelationships, but a sociological analysis of Yeats's association with Henley and of his career strategies in the early 1890s. What 'The Lake Isle' meant in the Observer and what Yeats used it for, the literary interests it was meant to serve, at that point in his career cannot easily be distinguished. At the same time our perception of the poem is coloured by the highly politicized accompanying textual materials over which he, as a contributor, had no control. Just as his mythopoeic celebration of the mysteries of the Sligo landscape is pitted against the Observer's politically expedient insistence on its destitution, so his idealization of the primitive is set against a racist ideology of the uncivilized 'wild man'. The extent to which one sees these discourses as complementary, contradictory, or simply contiguous depends mainly on how one reads Henley. One could, for instance, take John Frayne's sceptical view. He felt Henley's support for Yeats was not 'innocent', since he knew his contributions 'would have confirmed in an English reader the prejudice that the Irish were hopeless dreamers and their affairs had best be kept in capable English hands'.[37] Though many of Henley's clubland readers might well have responded in this way, and seen Yeats and 'The Lake Isle' as two good reasons for maintaining the Union, the little available evidence, notably Henley's review of Wanderings, suggests he was, for all the violence of his often inconsistent opinions, seldom capable of such cynical calculation. It is more likely that Yeats's contributions were, for him, acceptable because they could be seen to have no contemporary political relevance. 'The Lake Isle's' appearance in the Observer was, in other words, less a consequence of Henley's political sophistication, and more a result of his no less ideological assumptions about the nature of the 'literary'. Yet the case for considering the promiscuous effects of the non-authorial circumtext need not rest on speculations about Henley's motives. Whether taken to be complementary, contradictory, or merely contiguous, the simultaneous presence of Yeats's consciously inflected 'literary' version of the West of Ireland and the Observer's tendentiously 'political' version reveals just how contested and unstable the boundaries between these two discursive categories were in the 1890s.

The fact that 'The Lake Isle' was first published in the Observer does not only tell us about the company Yeats chose to keep in the early 1890s. It reveals something about his interests and priorities at that point in his career. Though he claimed to have previously declined an offer of a sub-editorship for the Unionist Manchester Courier on political grounds, when it came to the Observer the enormous professional advantages of his freelance association allayed his political and even literary qualms.[38] For him, Henley's outspoken, though never consistent, Republicanism in the English 'Republic of Letters' was always more important than his Unionist politics. Given the position Henley occupied, as a distinguished editor and established figure in the 1880s avant-garde, his backing was a mark of distinction confirming Yeats's ascendancy, as a young Irish poet, in the English literary field. The subsequent publishing history of 'The Lake Isle' in the 1890s reveals a similar interplay between the diversity of Yeats's alliances and the growing complexity of his evolving interests. If 'The Lake Isle's appearance in the Observer showed something of his eagerness to earn the recognition of the established English avant-garde, its re-emergence in The Book of the Rhymers' Club (1892) reinforced his ties to what he called 'the very newest literary generation' in London.[39] Though, as he admitted, not all the club's members were young (John Todhunter, for one, was in his fifties) most of the major players were, like him, in their twenties. Their first anthology was, then, intended to be a provocative literary statement (Yeats called it a 'manifesto') designed to seize the cultural high ground by naming and publicizing the new self-appointed 1890s avant-garde.[40] Though published by the enterprising young London publisher Elkin Mathews, the anthology was initiated allegedly by Yeats, edited by a committee of Rhymers: Richard Le Gallienne (Mathews's reader), George Greene, Lionel Johnson, Todhunter, and Ernest Dowson, and financed by the twelve participating members themselves.[41] Prefaced by a list of the contributors, many of whom were not English, it was, in effect, the book of and by the arrivistes on the English literary scene. The advantages for Yeats of this dual alliance with London's established and emergent avant-garde were clearly evident in the Observer's review of the anthology. Though its title, 'Minors', ensured the newcomers knew their place, the anonymous reviewer, who gave special praise to 'The Lake Isle', singled out Yeats who was 'ever an artist and a poet', and added 'for the other contributors, let it in fairness be recorded of them that they do their best'.[42]

The next two anthologies in which 'The Lake Isle' appeared had less to do with Yeats's advances in the English literary field, and more to do with his increasingly intricate Irish and Pan-Celtic connections. Lyra Celtica (1896) was a Scottish initiative published by the new firm Patrick Geddes and Colleagues of Edinburgh, edited by Elizabeth Sharp, and introduced and annotated by her husband William (alias 'Fiona Macleod') who was Geddes's literary adviser. Though Geddes himself had a wide range of interests (he was a progressive biologist, sociologist, town-planner, educationalist, and a patriotic Scottish Nationalist) he shared the Sharps' commitment to the Pan-Celtic cause. As Elizabeth put it, Geddes and her husband were both 'ardent Celts who believed in the necessity of preserving the finer subtle qualities and the spiritual heritage of their race against the encroaching predominance of the materialistic ideas and aims of the day'.[43] While the firm's bi-annual five-shilling magazine The Evergreen: A Northern Seasonal (1895-96), edited by William Sharp, reflected its full range of interests, the anthology was the flagship of its 'Celtic Library'. Never as comprehensive as its subtitle announced (An Anthology of Representative Celtic Poetry: Ancient Irish, Alban, Gaelic, Breton, Cymric, and Modern Scottish and Irish Celtic Poetry) it was, as Sharp admitted in his introduction, 'intentionally given over mainly to modern poetry' and to the myriad subdivisions within recent Celtic culture, ranging in twelve sections from the 'Scoto-Celtic (Middle Period)' to the 'Irish (Modern and Contemporary)', and from the 'Contemporary Anglo-Celtic Poets' (Welsh, Cornish, and Manx) to the largely North American 'Celtic Fringe'.[44] It was, in short, a propaganda vehicle for the Celtic literary revival which, as Yeats remarked in a letter to Sharp, 'is certain to be very influential & to help forward a matter I have myself much at heart -- the mutual understanding & sympathy of the Scotch Welsh & Irish Celts'.[45]

Though published four years before Brooke and Rolleston's Treasury of Irish Poetry in the English Tongue (1900), Sharp's collection was, as he made clear in his introduction, in direct competition with it from the start. Recalling the inaugural address entitled 'The Need and Use of Getting Irish Literature into the English Tongue' which Brooke gave to the new London-based Irish Literary Society in March 1893 (it was here that Brooke first called for a 'Golden Treasury' of Irish verse) Sharp gently rebuked him for his 'over-emphasis on the word Irish, which he frequently uses instead of Celtic', and then went on to quote and re-write the conclusion to his lecture:

When we have got the old [Celtic] legendary tales rendered into fine prose and verse, I believe we shall open out English poetry to a new and exciting world, an immense range of subjects, entirely fresh and full of inspiration. Therefore, as I said, get them out into English, and then we may bring England and [Celtdom] into a union which never can suffer separation.[46]

In their revised form, these were, for Sharp, 'inspiring words'.[47] Yet the conflict of interests between the anthologists of the Irish and the Pan-Celtic literary revivals remained chiefly his bugbear. When Brooke's Treasury eventually appeared in 1900, it was preoccupied with other disputes, internal to the Irish literary tradition itself. Published by the eminent London firm Smith, Elder & Co., promulgated at the Irish Literary Society's first meeting, edited by its newly elected President and his son-in-law, the former Rhymer T. W. Rolleston, who was its first secretary, and dedicated to Sir Charles Gavan Duffy, its first President and, as the dedication pointed out, the editor of 'the first worthy collection of Irish National Poetry', the Treasury embodied the Society's position in book form.[48] As the carefully worded title declared, and as Brooke reiterated in his introduction, it was not only intended to be an Irish version of F. T. Palgrave's famous Golden Treasury (1861), it was also a rejoinder to those in the Gaelic League who took their inspiration from the President of the Irish National Literary Society, Dublin, Douglas Hyde, and more particularly from an especially uncompromising reading of his 1892 lecture, 'The Necessity for De-Anglicising Ireland'. Not that the anthology and its Nationalist editors were against De-Anglicization. 'Were it possible that Irish literature should be anglicised', Brooke insisted, 'there would soon be no literature worth the name in Ireland' (p. xix). Rather, while retaining some sympathy for the League's efforts in the spheres of education and scholarship, he, like Rolleston, believed Gaelicization was not the only, or the most effective, means of asserting Ireland's cultural independence. On pragmatic and political grounds, he argued that 'if Irish writers do not deviate into an imitation of English literature, but cling close to the spirit of their native land, they do well for their country when they use the English tongue' (p. viii). His anthology was, therefore, committed to those poets, like Yeats, who affirmed Ireland's national traditions by translating its 'ancient myths, legends, and stories' into a language that was 'rapidly becoming universal'.

Yeats was at the centre of the various social networks and cultural controversies out of which these three anthologies emerged. So too was 'The Lake Isle'. The only poem of his to appear in all three, it was repeatedly co-opted by the advocates of these various causes and recontextualized by the books themselves. By placing it in particular bibliographical, textual, and paratextual configurations, each anthology had designs on its reception, designs Yeats implicitly supported but which for the most part remained outside his direct control. The Book of the Rhymers' Club was the exception. He had a considerable say in both its production and reception. Though it was, in his view, 'the manifesto of the circle', the book itself contained only the vaguest indications of their collective rationale.[49] To obtain any sense of this, readers had to rely on the conjunction of the title and the two paratextual poems that opened and closed the volume, respectively 'At the Rhymers' Club: The Toast' by Ernest Rhys and 'The Song of the Songsmiths' by George Greene, both of which were italicized and so typographically set apart from the book's fifty-five other poems. Their central point was that 'Rhyme', their metonym for poetry, was not what it used to be: Rhys looked back to Ben Jonson and Herrick, Green simply to the 'ancient fire' in 'a far, forgotten clime', and that its fate lay in their 'prentice-hand'.[50] Predictably enough, they found, as Rhys put it, 'the Muse degraded, | And changed, I fear, and faded'. Limited as it was, their diagnosis of this decline was no less expected: poetry was impoverished because it had been sold out to various alien interests. Testifying to the club's new purism, Rhys proclaimed: 'We drink defiance | To-night to all but Rhyme, | And most of all to Science'. Adopting a characteristically regal figure to reinforce the idea of poetry's sovereign authority, Greene added that their 'task' was to restore 'the queenly rhyme' with 'reverent toil'. The only specific indication of the way they intended to undertake this was the selection itself which expressed their Paterian commitment to the short lyric as the highest form in the hierarchy of poetic genres. Though the anthology contained a wide variety of short forms, including sonnets, ballads, odes, and villanelles, it clearly declared itself for concise, highly charged modes of poetic expression and against the potentially impure discursiveness of longer, narrative modes. This new poetic austerity was underscored by the book itself. Derived from the innovative designs associated principally with Whistler and Walter Blaikie, its rather uninspired plain yellow cloth cover, simple, unornamented typographical layout, bold use of white space, and untrimmed pages, combined with its five-shilling price and well-advertised limited print run, embodied the Rhymers' modernity and purist exclusivity in bibliographical form.[51]

The person who developed these suggestive generalizations into something more substantial was Yeats himself. He had as much to do with the creation of the anthology as with the production of its meanings, since he spent his career writing and re-writing the club's manifesto, producing, in effect, a series of retrospective paratexts, ranging from the essay of April 1892 entitled 'The Rhymers' Club', which formed part of his 'Celt in London' series for the Boston Pilot, to his introduction to The Oxford Book of Modern Verse (1936). These statements varied according to the period and context in which he was writing, and most often said more about Yeats than about the Rhymers, but each also underscored and developed Rhys and Greene's general claims. As Yeats emphasized from the start, the Rhymers were too dissimilar to be a 'school of poets in the French sense'.[52] What united them, more than any particular aesthetic (he mentions, by way of contrast, the French Decadents, Symbolists, Parnassians, and Naturalists) was their reaction to the alleged bankruptcy of the English poetic establishment. In the 1892 article, he set the frivolous 'search for new forms merely' of 'the Gosse, Lang and Dobson school', the 1870s avant-garde, against the new seriousness with which the Rhymers' wished to look 'once more upon the world' (pp. 57-58). Referring specifically to the Music Hall preoccupations of John Davidson (though a regular Rhymer, he was not in the first anthology) and Arthur Symons, he added 'the typical young poet of our day is an aesthete with a surfeit, searching sadly for his lost Philistinism' (p. 59). Only six months later, however, he was emphasizing the Rhymers' continuities with the aestheticism of their immediate predecessors, notably their belief that 'poetry is an end in itself'.[53] True, this was in an article for United Ireland, entitled 'Hopes and Fears for Irish Literature', in which he typically played the English off against the Irish and vice versa: he set mature, but enervated, English aestheticism against the youthful, but too highly politicized energies of the artless Irish. Aware as always of his own sensitive position, for this readership he also distanced himself from the English by stressing the 'dependence, as I conceived it, of all great art and literature upon conviction and upon heroic life'. Yet, over the years, the Rhymers' austere commitment to 'all but Rhyme' established itself in Yeats's mind as the key to their collective rationale. In 1912, after mentioning his own reaction to the 'rhetorical poetry of the Irish politicians' (i.e. the Young Irelanders, in particular) he claimed to have found in London 'a group of young lyric writers who were also against rhetoric'.[54] He now saw them, as he put it in 1936, not reacting to the formalist aestheticism of Dobson, Lang, and Gosse but to the supposed moral, scientific, and political 'discursiveness' of the previous generation: Browning, Tennyson, and Swinburne.[55] For him, the Rhymers were, in short, purist allies whose generational struggles within the English poetic tradition resembled and reinforced his own quarrels with the Irish literary establishment.

In so far as his association with the Rhymers consolidated his position within the English avant-garde, it built on the achievements of his alliance with Henley's Observer. This much is reflected in the poems he contributed to their first anthology. Though 'Dedication of "Irish Tales" ', with its references to 'Exiles', Ireland's 'Sorrow' and its nostalgia for the time 'when her own people ruled in wave-worn Eri', is more explicitly political than anything he submitted to Henley, his five other contributions: 'The Lake Isle', 'A Man who dreamed of Fairyland', 'Father Gilligan', 'A Fairy Song', and 'An Epitaph', all came from the Observer.[56] Yet, in the anthology, they not only appeared in a new publishing context (a miscellaneous collection of fifty-seven short poems under one title) they were made to do new work. Supplemented by Rhys and Greene's paratextual poems and Yeats's own commentaries, the anthology invited its readers to attend to them as a manifestation of the Rhymers' collective bid to redefine literariness for the 1890s. In this context, for instance, the Irishness of 'The Lake Isle"s theme and setting, which had been crucial in the Observer, was now less important than its short first-person lyric form, its metre, diction, and syntax. In fact, compared to the many humdrum lyrics that made up the supposedly ground-breaking collection, it was one of the few poems that managed to bring the Rhymers' grand, self-promoting theory somewhere in line with their practice. In giving voice to the atavistic desires of the first-person speaker, it rejected modernity, and implicitly the authority and advances of 'Science', in a modern literary idiom the novelty of which was seldom evident elsewhere in the volume. While its determined but pliant rising triple rhythm and variable line lengths hold up well against the ploddingly consistent iambic tetrameter of Ernest Radford's wholly conventional 'A Sundial -- Flowers of Time' (p. 3), its precise, often inventive diction ('bee-loud', 'purple glow', and so on) acquires new life when read alongside the tired abstractions of Richard Le Gallienne's 'What of the Darkness?' which ends: O! is the Darkness too a lying glass, Or undistracted do ye find Truth there? -- What of the Darkness? Is it very fair? (p. 85)

Even its familiar languorous tone crackles with surprising energy when compared to the staggering banality of George Greene's lines: 'Full of world-weariness, and of the sense | Of unachievement, lies the toiler down'.[57] Indeed, in this motley company its own stylistic flaws seem relatively minor. As Yeats himself recalled in his Autobiographies, 'a couple of years later I would not have written that first line with its conventional archaism -- 'Arise and go' -- nor the inversion in the last stanza' (presumably, 'pavements grey'). 'I only understood vaguely and occasionally', he added, 'that I must for my special purpose use nothing but the common syntax'. Yet, despite these afterthoughts, he was still willing to concede that the poem was 'my first lyric with anything in its rhythm of my own music'. Attesting to the affinity between his own aims and the Rhymers' purist literary values, he explained: 'I had begun to loosen rhythm as an escape from rhetoric and from that emotion of the crowd that rhetoric brings' (pp. 153-54). Read in the context of his own oeuvre, then, it could be argued that the poem represents an early partial failure of his individual poetic project. In The Book of the Rhymers' Club, however, it is one of the few poems affirming the partial success of their collective attack on the English poetic establishment.

If the Rhymers' anthology advocated, with Yeats's explicit support, a formalist reading of 'The Lake Isle', Lyra Celtica did precisely the reverse with at most his tacit agreement. It privileged a biographical and thematic reading. This was not only because William Sharp, in his notes, used the last stanza of the poem to confirm Yeats's Celtic roots; or because the poems were arranged according to their authors' geographical and historical origins ('Irish (Contemporary)' in Yeats's case). Rather, it was because the anthology rested on the assumption that 'The Lake Isle' was, like all the other poems it contained, in the first instance an expression of its author's racial origins. As Sharp typically pointed out, with some inventive inaccuracy, in his short biographical note at the back of the volume, Yeats had the right Celtic pedigree and upbringing:

Born (of an Irish father, and of a Cornish [sic, Susan Pollexfen's family were, in fact, originally from Devon] mother come of a family settled in Ireland) at Sandymount, Dublin, in 1866 [sic]; but early life chiefly spent in Sligo, and on the Connaught seaboard. Of late years, Mr Yeats has passed much of his time in London, but is never absent from Ireland for any long period -- '. . . for always night and day I hear lake-water lapping with low sounds on the shore; While I stand on the roadway, or on the pavements grey, I hear it in the deep heart's core.' (p. 398)

For Sharp, these lines were clearly no longer an expression of an anonymous urban speaker's longing for a primitive, rural idyll; they were a testament to the depth of Yeats's racial feeling, genuine or invented. Yet, never one to fuss over precise details or restrictive definitions, and eager as he was to avoid the charge of being a 'zealous Celticist', Sharp also accepted that 'Celtic' was as much a cultural as a racial category (p. xxviii). In the absence of 'Celtic blood', one could always detect the 'Celtic strain'; and, more bizarrely, one also had to recognize one 'may at any moment encounter the Celtic brain in the Anglo-Saxon flesh' (pp. xxv, xxvii). On these grounds he was willing to make a case for the inclusion in the anthology of Milton, Keats, Byron, Scott, Burns, Shelley, Coleridge, Swinburne, Tennyson, and Stevenson. After all, he rhapsodized 'it is an unfrontiered land, this pleasant country in the geography of the soul which we call Bohemia; and here all parochial and national, and even racial distinctions fall away' (p. xxviii). But, he added, in a more pragmatic mood:

To avoid confusion, the Editor has refrained from representing poets whose 'Celtic strain' is more or less obviously disputable; hence the wise ignoring of the claims even of Scott and Burns. Byron was more Celtic in blood than in brain, and is represented really by virtue of this accidental kinship. (p. xliii)

With this makeshift vision of Pan-Celticism without borders, it is not surprising that Yeats occupied the position he did in Sharp's book. He was not simply one indisputably Celtic author among the eighty-seven named in the table of contents. Celtic in blood, brain, and style, he was 'pre-eminently representative of the Celtic genius of today' (p. xliv).

Sharp's critical opinions were as derivative as the intricately entwined harps, shamrocks, and lattice-work of the anthology's 'Celtic' cover design and ornamentations by Helen Hay. In his introduction, he simply rehashed Arnold whom he described as 'the most sympathetic and penetrating critic of the Celtic imagination' (pp. xliii, xlviii; like Arnold, he also cited Renan's 'La Poesie de la Race Celtique' (Essais, 1859) with approval). Admittedly, his Arnold was only a pro-Celtic cultural revisionist eager, like him, to reclaim a marginalized sensibility for English literature. He makes no mention of Arnold's view that the Celts were not capable of producing 'true art, the architectonice `which shapes great works'; nor does he consider the political interests at stake in Arnold's analysis.[58] For Sharp, the primary value of the 1867 essay lay in its characterization of the 'Celtic strain' which he proceeded to rehearse in the course of his introduction. He directed his readers' attention to the 'natural magic', 'strange melancholy', 'rare music', 'deep yearning emotion', and 'cosmic note' of all 'authentic documents of Celtic genius', qualities that justified the juxtaposition of poems as diverse as Byron's doleful love lyric 'When we Two Parted' and the Breton poet, Louis Tiercelin's death-wish lyric 'By Menec'hi Shore' which struck 'the keynote of the poetry that is common to all the Celtic races'.[59] He also re-affirmed Yeats's particular pre-eminence in explicitly Arnoldian terms. 'In almost every poem he has written', Sharp noted, 'there is that exquisite remoteness, that dream-like music, and that transporting charm which Matthew Arnold held to be one of the primary tests of poetry, and, in particular, of Celtic poetry' (p. xliv). This was especially true, of course, of those in the anthology. Though Elizabeth Sharp had wanted to include five of Yeats's poems, in the end, Unwin, the copyright holder, allowed her to publish only three: 'The Lake Isle', 'The White Birds', and 'They went forth to the Battle, but they always fell' (afterwards 'The Rose of Battle').[60] William, however, managed to get around these restrictions to some extent by including ample quotations from The Wanderings of Oisin, 'The Madness of King Goll', 'The Stolen Child', and 'The Rose of the World', in his introduction. All these confirmed Yeats's status as the exemplary Arnoldian Celt. While the title 'They went forth to Battle, etc.', from Ossian, recalled the epigraph to Arnold's essay, Sharp's reading of 'The Lake Isle' gave some weight to his observation that 'the Irish themselves have given us the most poignant, the most hauntingly sad lyric cries in all modern literature', through which one could hear 'the lament of exiles', 'the history of the Celtic race itself'.[61]

The extent to which Yeats endorsed this flattering but also flamboyant and derivative Arnoldian reading of his poems and position is a vexed question. Initially extremely hostile towards Sharp (when they first met in July 1887, he told Tynan he 'hated his red British face of flaccid contentment'), he soon accepted him as a useful pro-Celtic ally (CL, i, 24). Privately, he supported his Pan-Celtic initiatives, praised Lyra Celtica, and made inviting references to 'our movement' (CL, ii, 149). In public, he lauded 'Fiona Macleod' as 'le nom le plus intimement lie 'a` ce mouvement, a` cette rennaissance celtique, comme disent les journaux'.[62] Admittedly this statement appeared in L'Irlande Libre, the propaganda magazine for Maud Gonne's Paris-based Association Irlandaise, in April 1898 when he was most probably still in the dark about 'Fiona's' real identity. And, of course, he agreed to let the Sharps select and comment on some of his work. As he noted in a letter to Unwin about Elizabeth's request for poems, 'Celtic poetry is so much my business that I can hardly refuse her leave to include something' (CL, i, 468-69). Yet, other factors, notably the fracas over the chairmanship of his own lecture on 'The Celtic Movement' to the Irish Literary Society on 4 December 1897, suggest he was also eager to keep his distance in public from Sharp and his nayvely Arnoldian views. When faced with the prospect of Sharp chairing the lecture, he was, according to Lady Gregory, 'furious [. . .] & declared it wd bring ridicule on the whole movement'.[63] Tactful and accommodating as ever in private correspondence, he was none the less delighted when Gregory managed to strong arm the reluctant Sharp into backing down.[64] In the lecture, which subsequently appeared, with revisions, as the article entitled 'The Celtic Element in Literature' in the June 1898 issue of Unwin's 'International Review' Cosmopolis, he was, after all, far less accepting of Arnold's analysis than Sharp had been. In it he felt obliged to 're-state a little' Arnold's (and Renan's) arguments, to 'see where they are helpful and where they are hurtful'.[65] In fact, the lecture/essay performed an intricate rhetorical manoeuvre which retained the terms of Arnold's argument while subverting its central pro-Celtic point. When Arnold saw 'Celtic melancholy', Yeats saw 'primitive melancholy'; or when he

asks how much of the Celt must one imagine in the ideal man of genius. I prefer to say, how much of the ancient hunters and fishers and of the ecstatic dancers among hills and woods must one imagine in the ideal man of genius? (pp. 183-84)

In effect, he shifted the focus of the cultural debate by replacing Arnold's politically charged Celticism with his own more neutral brand of neo-primitivism. This ostensibly apolitical, even anti-political strategy had its political uses, however, in so far as it left him free to answer any Nationalist motivated scepticism about his own Pan-Celtic sympathies. Responding to D. P. Moran's hard-line public attack on his use of the terms 'Celtic note' and 'Celtic Renaissance', and his supposed Arnoldian enthusiasms, he replied, in a letter to Moran's The Leader: 'All I have said or written about Mathew Arnold since I was a boy is an essay in "Cosmopolis", in which I have argued that the characteristics he has called Celtic, mark all races just in so far as they preserve the qualities of the early races of the world'.[66] Far from being un-Irish, in other words, his affirmation of Ireland's ancient Celtic roots had everything to do with his reaction to the hollow materialism of modern, especially English, 'civilization'. On this analysis, the 'heart's deep core' in 'The Lake Isle' was the site of the modern speaker's longings for a primitive Ireland which happened to be Celtic, not of his buried racial feelings as a displaced Celt, a reading Yeats encouraged in Poems (1895) and even to some extent in the Observer, but which Sharp occluded in Lyra Celtica. It should be remembered, however, that the ever opportunistic Sharp did not take long to fall in line with Yeats's reassessment of Arnold. In an article for The Fortnightly Review of January 1899, conveniently disguised as 'Fiona Macleod', 'she' denounced his earlier phrases such as 'Celtic glamour' and 'Celtic Renaissance' as the 'shibboleths of the journalists', and went on, at some length, to praise Yeats's 'admirable article on the Celtic element in literature: an exposition with which I, for one, find myself wholly in accord, believing as I, too, do, that much of what is specifically called Celtic is simply ancient'.[67] Such were the benefits of a secret double identity.

Despite Yeats's often uneasy personal relations with its editors, especially Rolleston, A Treasury of Irish Poetry in the English Tongue was in many ways a more promising venue for 'The Lake Isle' than Sharp's Lyra Celtica. This time, at least, he supported both the cause the anthology was intended to advance and the editors' interpretation of it. Though Brooke had initially planned to make literary merit (what he called a 'relatively high standard of excellence') the sole principle of editorial selection, Rolleston convinced him to produce a 'history of Irish poetry in English', partly to promote their cause by documenting the diversity of that tradition, partly to differentiate their product from Yeats's own 'brief Anthology', A Book of Irish Verse (1895).[68] This did not mean they saw it as a work of disinterested historical scholarship, however. The 'history of Irish poetry in English' was, for them, not a comprehensive record but a triumphant bildungsroman. As Brooke puts it in his introduction, readers would find 'here a school of poetry in the making, a child growing into a man' (p. x). In another rhapsodic blend of figures, he sheds a light of sorts on what this developmental process actually entailed.

The river of Irish poetry in the English language [. . .] rose a hundred years ago in the far-off hills, and wrought its turbulent way down the channelled gorge it carved for its stream out of its own mountains. Other streams have joined it, bearing with them various waters; and it has only just now issued from the hills, and begun to flow in quieter and lovelier lands, glancing from ripple to pool and from pool to ripple, among woods and meadows, happy, and making its lovers happy. It is the youngest child of the Goddess Poesy. (p. xxxiv)

Despite the confusing return of the child figure, used now to situate Irish poetry in relation to world (especially English) literature, his general point is clear enough. The story of Irish poetry in English was, for Brooke, not simply one of growing cultural richness and diversity, but one of progressive autonomy. Without ceasing to be 'Irish', the tradition had gradually become more universally 'poetic': that is, less narrowly 'political', particularly after the decline of the Fenian movement. 'After '67', Brooke notes, 'patriotic rage seldom recurs as a separate motive for poetry' (p. xiii). Hence his lofty image of the 'youngest child of the Goddess Poesy': 'It is only quite lately that modern Irish poetry can claim to be a fine art' (p. iix). The aim of his value-laden Treasury was, then, not only to answer the Gaelic League hard-liners by promoting an alternative version of Irishness but to reconstruct the 'history of the development of a special national art' (p. x).

These aims governed the arrangement of the book as a whole, and, most importantly, the physical placement of Yeats's nine poems. Located at the end of Book v, they were placed for the reader not simply in the context of a century-old Irish poetic tradition, but as its telos. They marked the highest literary achievement to date of a history that began, according to the anthology, with 'The Wearin' o' the Green', dated 1798 by the editors and described as the 'finest of Irish street-ballads' (p. 2). The exact significance of this positioning is not immediately obvious. For one thing, we have to notice that the twenty-nine authors included in Book v (among others, Todhunter, Johnson, 'A. E.', Tynan-Hinkson, Hyde, Rolleston, and Brooke) are not arranged alphabetically or in order of seniority. For once, Yeats is at the end of the list not simply because his surname happened to begin with a 'Y'. For another, we need to see that Book v is ideologically, if not physically, the end of the anthology. The final section, Book vi, which includes such figures as Edward Dowden, George Savage-Armstrong (two of Yeats's key Irish enemies) and the Rhymer George Greene, is beyond the pale of the editors' official Irish canon, the site of the poetic West Britons. As Brooke commented charitably but firmly, and no doubt with an eye on his Gaelic League critics:

It is impossible not to admire the subtlety, tenderness, and love of nature of these poets, but their place is apart in an Anthology of Irish poetry. They have not kept, along with their devotion to their art, the spirit of their native land. (p. xiv)

The arrangement of the six books was, in other words, not simply chronological. They were intended loosely to represent 'on the whole distinct phases' in the 'general movement of Irish poetry during the nineteenth century', from the popular 'ballads and songs' of the late eighteenth century in Book i, to the patriotic and humorous early-nineteenth-century poems of Book ii, to the 'propaganda' 'Poets of The Nation' included in Book iii, to Mangan and Ferguson, who are credited with starting 'a new Celtic movement' and so given Book iv to themselves.[69] While the poets in Book vi represent an unfortunate coda to the triumphal narrative from 'popular' to 'high' Irish culture, from 'political' to 'fine' national poetry, those in Book v comprised its latest and most promising chapter, with Yeats as the final paragraph. Though they, like Yeats, had 'studied and honoured the great masters of song, and, as they write in English, the English masters', they had 'yet endeavoured to secure and retain in their poetry not only the national and spiritual elements of the character of the Irish people, but also that appealing emotion which lives like a soul in the natural scenery of Ireland' (p. xv).

This combination of values -- 'national', 'spiritual', and 'natural' -- explained not only Yeats's particular eminence as the last author in Book v, but the significance of 'The Lake Isle's inclusion in the anthology. For both editors, Yeats's status was partly a consequence of his prominent role among the new generation of Irish poets who rejected 'political poetry' and 'consecrated their verse towards the support of a vigorous and vital nationality: first, by the representation in a modern dress of the Irish myths and sagas; and secondly, by the representation of the spiritual elements of the modern world from an Irish standpoint, and in an Irish spirit' (p. xxix). Yet what made Yeats stand out among his 'fellow-poets' was the fact that his poetry had 'a wider range' (p. xxxii). While never losing his distinctive Irishness, he 'proved his universality', as Brooke puts it, by combining his interest in 'mysticism' and the 'spiritualised representation of the ancient Celtic stories' with 'some work, direct, simple, and humane, on actual life'. For a neo-Wordsworthian and Unitarian such as Brooke this last aspect of his oeuvre was crucial. For him, 'the better food and pleasanter delights of poetry' could best be found, not in 'a hundred passionate and mystic things', but 'in the daily life of men and women spiritualised by natural passion' (p. xviii). So, while he ensured the anthology reflected the generic and thematic diversity of Yeats's output, from the occultism of 'The Two Trees' to the legendary appeal of 'The Island of Sleep' (from Oisin), and from the confident transcendentalism of 'The Rose of the World' to the defiant realism of 'The Lamentation of the Old Pensioner', he had a special investment in 'The Lake Isle', or, rather, in a particular reading of it.[70] On this issue he had the full support of his son-in-law who shared his reservations about Yeats's mystical tendencies. In his headnote to Yeats's poems, Rolleston comments: 'The mystic in him is sometimes, especially in his later work, found adoring the mere stigmata of mysticism; and then one thinks with dismay that a finer and stronger genius than Blake's may some day lose itself in that dreary waste inhabited by Los and Orc and Enitharmion.' 'But', he goes on, 'these forebodings soon vanish when one hears again the "lake water lapping" on the shores of Innisfree' (p. 497). In the Treasury, then, 'The Lake Isle' was presented to readers not only with Yeats's other contributions as the clearest evidence of the 'development' of the tradition of Irish poetry in English. It was also offered as a justification of his particular eminence, based on his range, and as a special illustration of the fact that, despite his mystical otherworldliness, he had not forgotten the 'natural scenery of Ireland'. It was, in short, to be read as a gratifyingly 'fine', distinctly Irish, and reassuringly topographical lyric in English.

Yeats was broadly in sympathy with the Treasury's position on a number of issues. He supported its opposition to hard-line nativists, such as Moran who predictably fuelled a protracted controversy in The Leader on its publication in December 1900; he saw Brooke and Rolleston as allies in his disputes with West Britons such as Dowden; and he shared their outlook on the intricate relations among politics, poetry, and nationalism.[71] But he also had some misgivings about the way the anthology represented him. As he told George Russell, when Russell was enquiring about which poems of his to include in another anthology, 'Rolleston did not please me over well by giving long extracts from what I think immature verse' (CL, iii, 493). In his headnote to Yeats, Rolleston not only drew attention to his selective self-presentation in Poems (1895), he included twenty-two lines from The Island of the Statues (Act ii, Scene 3, ll. 131-34, 160-65, 201, 203-13), the suppressed Classical verse drama Yeats had published in The Dublin University Review (April-July 1885). This was not the dispassionate scholarly gesture it might seem. Rolleston no doubt took it a little personally that Yeats expunged poems he, as the former editor of the review, had initially deemed worthy of publication. Moreover, there were evidently some explicit tensions between Yeats's and the editors' views on the mystical tendencies in contemporary Irish poetry. In the four headnotes Yeats himself contributed to the Treasury on Lionel Johnson, 'A. E.', Nora Hopper, and Althea Gyles, he spoke of little else but their various interests in 'a spiritual life'; and, whereas Brooke chastised 'A. E.' for ignoring 'universal human life' and 'Nature as she seems to the senses', he called his poems 'the most delicate and subtle that any Irishman of our time has written'. He thought particularly highly of them 'because their writer has not come from any of our seats of literature and scholarship [vide Trinity College, Dublin men like Brooke and Rolleston], but from among sectaries and visionaries'.[72] These differences of opinion on aesthetic and meta-physical matters may well have made him uneasy about the way the Treasury presented 'The Lake Isle'. After all, as he had mentioned to Tynan, as his father's illustration in The Leisure Hour had reiterated, and as he also made clear in a note appended to the suppressed poem 'The Danaan Quicken Tree', which appeared in The Bookman for May 1893, he saw Innisfree not simply as a scenic spot on the West of Ireland, but as a mystical place, with a long history.

It is said that an enchanted tree grew once on the little lake-island of Innisfree, and that its berries were, according to one legend, poisonous to mortals, and according to another, able to endow them with more than mortal powers. Both legends say that the berries were the food of the Tuatha de Danaan, or faeries.[73]

Attesting, once again, to the inseparability of 'The Lake Isle"s meanings and its publishing history, and to the important but limited part Yeats, as author, had to play in the overall process, this periodical paratext to another poem becomes, via a footnote, a direct commentary on 'The Lake Isle' itself in Allt and Alspach's Variorum Edition.

One of the axioms of publishing history is that authors have little to do with book production. As Roger Stoddard succinctly remarks:

Whatever they may do, authors do not write books. Books are not written at all. They are manufactured by scribes and other artisans, by mechanics and other engineers, and by printing presses and other machines.[74]

Yet it is important to bear in mind the differences between the politics of publication and the material processes of book production. While most authors may indeed never participate in the physical production of their books, many may have a considerable say in (and hence responsibility for) the overall design process. This was certainly true for Yeats. Though Poems (1895), to take the most striking early example, was published by T. Fisher Unwin, London, printed by the leading Edinburgh firm T. & A. Constable, with decorative designs by H. Granville Fell, Yeats, as Bornstein has argued, had an extraordinary influence over its total design, and hence over the way it presented him and his poems to the reader.[75] In this context, for instance 'The Lake Isle', which now formed part of the complex, multi-voiced section entitled 'The Rose', represented an especially resolute assertion of his otherwise often ambivalent poetic quest for 'the Eternal Rose of Beauty and Peace'. Yet, as we have seen, the poem also appeared in many other publishing contexts in the 1890s where Yeats's powers over its reception as author and book designer were significantly, if not completely, restricted. Entering the gravitational field of the National Observer, The Book of the Rhymers' Club, The Leisure Hour, Lyra Celtica, or The Treasury, and on each occasion subject to the planned and unplanned effects of new paratextual or circumtextual commentaries, bibliographical conditions, and socio-cultural interests, it began a rich public life in which Yeats's own designs were variously embellished, subverted, or effaced. Depending on the form in which they happened to come across it, readers of the 1890s could have found a judiciously 'literary' celebration of the West of Ireland uneasily entangled in an otherwise hostile 'political' portrayal of its destitution, or a short lyric asserting the new purist imperatives of a young group opposed to the English poetic establishment, or a poetic accompaniment to an illustration of an ancient Irish legend, or a particularly clear expression of Arnoldian Celticism, or an especially 'fine', but also fortunately topographical, example of the latest achievements in the tradition of Irish poetry in English. Of course, whether or not these readers actually interpreted the poem in these ways is another question, and one which falls outside the scope of this paper. All that needs to be said now is that these various bibliographical forms created, with or without Yeats's full support, deliberately or incidentally, the material and textual conditions that legitimized such readings. After all, just as the history of meaning should not be confined to the history of the author's intended meanings, so it should not be restricted to the history of reception, since it is generated as much by the producers of bibliographical forms, who have palpable designs on the texts they transmit, as by their consumers, who have minds and interests of their own.

[1] A Treasury of Irish Poetry in the English Tongue, ed. by Stopford A. Brooke and T. W. Rolleston (London: Smith, Elder, 1900), p. 492.

[2] George Bornstein, 'Remaking Himself: Yeat's Revisions of His Early Canon', Text, 5 (1991), p. 340.

[3] Hugh Kenner, 'The Sacred Book of the Arts', Yeats: A Collection of Critical Essays, ed. by John Unterecker (Englewood Cliffs, NJ: Prentice-Hall, 1963), p. 13. The debate about the ordering of Yeats's oeuvre has, of course, moved on, not least since the Finneran/Gould controversies in 1984. For a summary of these, see Richard J. Finneran, Editing Yeats's Poems (London: Macmillan, 1983), Editing Yeats's Poems: A Reconsideration (London: Macmillan, 1990), and, for Gould's version, see Yeats's Poems, ed. by A. Norman Jeffares (London: Macmillan, 1989), esp. pp. 706-49. The most interesting recent exploration of the critical implications of the oeuvre's arrangement is Hazard Adams, The Book of Yeats's Poems (Tallahassee: Florida State University Press, 1990).

[4] W. B. Yeats, Poems (London: Fisher Unwin, 1895), p. vi. Yeats is WBY hereafter.

[5] Bornstein, p. 351.

[6] Bornstein, p. 356. The quatrain he refers to is the untitled 'The friends that have it I do wrong'. See The Variorum Edition of the Poems of W. B. Yeats, ed. by Peter Allt and Russell K. Alspach (New York: Macmillan, 1957; corr. 3 imp. 1966), p. 778.

[7] The Collected Letters of W. B. Yeats, ed. by John Kelly and others, 3 vols (Oxford: Clarendon Press, 1986-94), i, 232, 243. Hereafter CL, i. For the most detailed account of the genesis of 'The Lake Isle', see Russell K. Alspach, Yeats and Innisfree (Dublin: Dolmen Press, 1965).

[8] WBY, John Sherman & Dhoya (1891; repr. Dublin: Lilliput Press, 1990), p. 58.

[9] WBY, Memoirs, ed. by Denis Donoghue (London and Basingstoke: Macmillan, 1972), p. 38.

[10] The most substantive variation is the alteration of 'on the shore' (1.10) in the Observer version to 'by the shore' in all subsequent versions. Though not included in the Variorum Edition, the versions in Lyra Celtica and A Treasury also have no substantive variations, but The Leisure Hour version does introduce an unexpected 'the' in line 10: 'I hear the lake water lapping'.

[11] Advertisement, Scots Observer, 18 May 1889,p.731.

[12] WBY, Memoirs, p. 38. For Yeat's later, more theory-laden assessment of Henley, which focuses on his antithetical character, see Autobiographies (London: Macmillan, 1955), pp. 153-63.

[13] Memoirs, p. 38 and Autobiographies, p. 159.

[14] Memoirs, p. 37. For the details about Henley's payments, see CL, i, 220, and R. F. Foster, W. B. Yeats: A Life(Oxford: Oxford University Press, 1997), pp. 94-95.

[15] For Henley's efforts on Yeats's behalf and Yeats's responses to them, see CL, i, 100, 107, 116-17.

[16] WBY, Memoirs, pp. 38-39.

[17] He also claims here that the article was 'by Henley'.

[18] [W. E. Henley], 'A New Irish Poet', Scots Observer, 9 March 1889, pp. 446-47 (p. 446).

[19] Matthew Arnold, 'On the Study of Celtic Literature', in Lectures and Essays in Criticism, ed. by R. H. Super (Ann Arbor: University of Michigan Press, 1962), pp. 291-386 (p. 346).

[20] See Allan Wade, A Bibliography of the Writings of W. B. Yeats (London: Hart-Davis, 1951), pp. 291-303. Besides 'The Lake Isle', Yeats's poetic contributions to the Observer were 'A Cradle Song', 'Father Gilligan', 'The Old Pensioner', 'A Man who Dreamed of Fairyland', 'A Fairy Song', 'Kathleen' (later 'A Dream of a Blessed Spirit'), 'An Epitaph' (later 'A Dream of Death'), 'Rosa Mundi' (later 'The Rose of the World'), 'The Peace of the Rose' (later 'The Rose of Peace'), 'The White Birds', 'Fergus and the Druid', 'The Rose in my Heart' (later 'Aedh tells of the Rose in his Heart'), 'The Celtic Twilight' (later 'Into the Twilight'), 'The Faery Host' (later 'The Hosting of the Sidhe'), 'Cap and Bell' (later 'The Cap and Bells').

[21] WBY, 'Father Gilligan', Scots Observer, 5 July 1890, pp. 174-75.

[22] WBY, 'Kidnappers', Scots Observer, 15 June 1889,p.100.

[23] WBY, 'Village Ghosts', Scots Observer, 11 May 1889,p.692. For his early assumptions about Henley's readership, see also 'Scots and Irish Fairies', 2 March 1889, pp. 410-11. For a more detailed account of Henley's intended readership, see Peter D. McDonald, British Literary Culture and Publishing Practice, 1880-1914 (Cambridge: Cambridge University Press, 1997), pp. 47-53.

[24] WBY, 'Kidnappers', p. 101.

[25] WBY, 'Columkille and Rosses', Scots Observer, 5 October 1889,p.550.

[26] WBY, 'Tales from the Twilight', Scots Observer, 1 March 1890,p.409.

[27] National Observer, 7 May 1892,p.641.

[28] Gerard Genette, Paratexts: Thresholds of Interpretation (1987), trans. by Jane E. Lewin (Cambridge: Cambridge University Press, 1997), p. 2.

[29] I am very grateful to Warwick Gould for directing me to his thorough and extremely useful account of 'The Lake Isle' in The Leisure Hour, 'Yeats as Aborigine', in Four Decades of Poetry, 1890-1930, ii.2, (1978), pp. 65-76. As I note in the text, the poem appeared in the August issue of 1896, not in October as Gould claims.

[30] Letters from Bedford Park: A Selection from the Correspondence (1890-1901) of John Butler Yeats, ed. by William M. Murphy (Dublin: Cuala Press, 1972), p. 32.

[31] Unsigned, 'Irish Land Legislation', Scots Observer, 31 August 1889,p.397.

[32] 'Irish Land Legislation', and Unsigned, 'Notes', National Observer, 22 November 1890,p.1.

[33] Unsigned, 'A Word to Mr. Balfour', National Observer, 22 November 1890, pp. 4-5.

[34] Unsigned, 'National Types: In Ireland', Scots Observer, 27 September 1890, pp. 485-86.

[35] Unsigned, 'The Scot in Ulster', Scots Observer, 2 March 1889,p.402.

[36] Unsigned, 'Thorough', Scots Observer, 10 May 1890,p.692.

[37] WBY, Uncollected Prose, ed. by John P. Frayne and Colton Johnson, 2 vols (London: Macmillan, 1970-75), i, 24.

[38] WBY, Memoirs, p. 31.

[39] WBY, Letters to the New Island, ed. by George Bornstein and Hugh Witemeyer (Basingstoke: Macmillan, 1989), p. 57.

[40] New Island, p. 59.

[41] For a more detailed discussion of the genesis and production of the anthology, see James G. Nelson, The Early Nineties: A View from the Bodley Head (Cambridge, MA: Harvard University Press, 1971), pp. 151-83.

[42] Unsigned, 'Minors', National Observer, 7 May 1892,p.646.

[43] Elizabeth Sharp, William Sharp (Fiona Macleod): A Memoir (London: Heinemann, 1910), pp. 248-49.

[44] Lyra Celtica: An Anthology of Representative Celtic Poetry, ed. by Elizabeth Sharp (Edinburgh: Patrick Geddes and Colleagues, 1896), p. xix.

[45] The Collected Letters of W. B. Yeats, ed. by John Kelly and others, 3 vols (Oxford: Clarendon Press, 1997), ii, 37.

[46] Sharp was no doubt working from memory or from his own notes, as the published version of Brooke's lecture reads: 'When we have got them [Irish stories] into fine prose and verse, I believe we shall open out to English Poetry a new and exciting world, an immense range of subjects, entirely fresh and full of inspiration. Therefore, as I said, get them out into English, and then we may bring England and Ireland into a union which never can suffer separation'. See Lyra Celtica, p. xxxv and Stopford A. Brooke, The Need and Use of Getting Irish Literature into the English Tongue (London: Fisher Unwin, 1893), pp. 56-57, 62.

[47] Lyra Celtica, p. xxxv.

[48] Treasury, p. v. The first of many editions of Duffy's The Ballad Poetry of Ireland appeared in 1845.

[49] WBY, New Island, p. 59.

[50] The Book of the Rhymers' Club (London: Elkin Mathews, 1892), pp. 1-2, 92-93.

[51] On the page facing the half-title, the publisher noted: 'Four hundred and fifty copies of this edition printed, of which three hundred are for sale.' Mathews also put out fifty copies of a more stylish large-paper edition.

[52] WBY, New Island, p. 57.

[53] Uncollected Prose, i, 248.

[54] WBY, Uncollected Prose, ii, 413.

[55] The Oxford Book of Modern Verse, ed. by WBY (Oxford: Clarendon Press, 1936), p. ix. This view of the Rhymers' purism is echoed most famously in 'The Grey Rock'.

[56] Rhymers' Club, pp. 54-55.

[57] 'The Pathfinder', p. 18.

[58] Arnold, Lectures, p. 345.

[59] Lyra Celtica, pp. xxiii, xxvii, xlix, 421.

[60] CL, i, 468-69.

[61] Lyra Celtica, pp. xlvi and xlix-l. Arnold's epigraph, attributed to Ossian, was 'They went forth to the war, but they always fell'. But in The Poems of Ossian, translated by James Macpherson, and introduced by William Sharp, the lines, which come from 'Cath-loda: A Poem', are 'His race came forth, in their years; they came forth to war, but they always fell' (p. 14).

[62] Uncollected Prose, ii, 109.

[63] Lady Gregory's Diaries 1892-1902, ed. by James Pethica (Gerrards Cross: Colin Smythe, 1996), p. 156.

[64] Gregory claimed to have told Sharp 'that Yeats's friends were of the opinion the Celtic movement wd be injured by them merging into one camp -- that they should rather be allies like the Unionists and Tories' (Diaries, p. 157); for Yeats's letter to Sharp and a fuller discussion of the fracas, see CL, ii, 148-49.

[65] WBY, Essays and Introductions (London: Macmillan, 1955), p. 174.

[66] CL, ii, 568. The letter was dated 26 August 1900. Moran had attacked Yeats's Arnoldianisms in his article 'The Battle of Two Civilizations', New Ireland Review, August 1900,p.330.

[67] Fiona Macleod, 'A Group of Celtic Writers', The Fortnightly Review, January 1899, pp. 35, 46.

[68] Treasury, p. ix.

[69] Treasury, pp. ix-xiii, 113.

[70] The Treasury also included 'The Hosting of the Sidhe', 'Michael Robartes remembers Forgotten Beauty', 'When you are Old', and 'A Dream of a Blessed Spirit'.

[71] For the details of Moran's attack and Yeats's response to it, see CL, iii, 6, 10, 25; for his views on the alliance against Dowden, see Uncollected Prose, i, 347.

[72] Treasury, pp. 465, 486-87.

[73] Variorum Edition, pp. 117, 742.

[74] 'Morphology and the Book from an American Perspective', Printing History, 9 (1987), p. 4.

[75] See Bornstein, p. 348.