I will arise and go now, and go to Innisfree,
And a small cabin build there, of clay and wattles made:
Nine bean-rows will I have there, a hive for the honey-bee,
And live alone in the bee-loud glade.
And I shall have some peace there, for peace comes dropping slow,
Dropping from the veils of the morning to where the cricket sings;
There midnight’s all a glimmer, and noon a purple glow,
And evening full of the linnet’s wings.
I will arise and go now, for always night and day
I hear lake water lapping with low sounds by the shore;
While I stand on the roadway, or on the pavements grey,
I hear it in the deep heart’s core.
As much as any poet who ever lived, Yeats appears to have made a rule out of Valéry’s observation that poems are never completed, only abandoned. This is why Richard J. Finneran, co-editor of The Collected Works and editor of The Poems (Volume I), understandably points out that “the notion of a ‘final’ or ‘definitive’ text of Yeats’s poems is fundamentally illusory.” At issue is the fact that it takes extensive spadework to uncover the history of the composition, revision, publication, and subsequent reworking and resetting of many of the lyrics. The same is true of the prose. Passages appear, are revised, and then reappear interweaving in different contexts and under different titles throughout Yeats’s life. This is particularly true of the autobiographical material. Yeats would even on occasion rewrite his letters after having sent them, with posterity in mind. Further, genres interweave. “The Lake Isle of Innisfree” appears to have grown out of a childhood memory of Sligo and readings in various sources, was first recognizably expressed as the meditation of a fictional character in the early novel John Sherman, and only then became a poem that went through drafts before being published and then widely anthologized in the 1890s. The work’s 121 ultimately canonical words changed less after first publication than in many other Yeats poems, remaining “unusually stable,” as Peter McDonald points out, but Yeats’s practice and the poem’s publication history mean that the poem’s allusive and historical background are quite rich. 
Despite the fact that Yeats did not extensively revise “The Lake Isle” once it was in print, a further complexity arises in discussing any one Yeats work on its own, as John Unterecker argued many years ago in A Reader’s Guide to William Butler Yeats, which is that “Political, occult, autobiographical, historical material was all carefully organized and reorganized as Yeats’s master plan for his work evolved.”  Many, if not most poets, produce work that may be of a piece, but they do not seek to create an overall retrospective design for it. With Yeats, however, if we are to take the author’s intentions as any guide, trying to read one poem in isolation from his larger project of a modern Irish art rooted in history, national myth, and the occult arguably does that poem some violence.
Given their complex history as individual works and Yeats’s overarching intentions, it therefore may seem surprising that so many of the poems, very much including “The Lake Isle,” succeed so powerfully on their own terms (judging by their ubiquitous appearance in anthologies, with most if not all context stripped away). Yet, in similar situations with other artists who also conceived of their individual works as parts of a larger whole, we nonetheless often closely read one work at a time, and Whitman, Blake, Faulkner, and Shakespeare’s history plays seem not to suffer irreparable damage from this. This is not a small matter, for it gets quite quickly at large questions of what it means to read a poem.
My own feelings about the matter are clear enough: what such an apparent contradiction suggests for programmatic artists is that there are things going on in the best individual works of a strong writer that go well beyond anyone’s ability to manage the future. This is in no way to criticize the artist who would try to achieve such control; creativity has its prerogatives. Rather, it is a testimony to the power of art in the right hands, a power that goes well beyond any artist’s avowed organizational scheme.
Indeed, many readers respond viscerally to “The Lake Isle” while knowing nothing of Yeats’s life, his revisions, or his larger intentions. They may know little of the poem’s history and reception, which Peter D. McDonald, Warwick Gould, Russell K. Alspach and others (including Yeats himself) have examined so closely in this case. This background includes Yeats’s personal life, especially his childhood and family, and his time in London; his relation with William Morris; Irish history, myth, and politics; W. E. Henley’s poetics, jingoistic politics and periodicals (in one of which, National Observer, the poem first appeared in 1890); Yeats’s involvement with the Rhymer’s Club; and so on. They may well not even know where Innisfree lies. In my view, it is absolutely fine if most readers do not know this material.
The reason that much of the background material on “The Lake Isle” is of secondary importance is that it can obscure the extravagant power of this poem. For, although I emphatically do not wish to argue in favor of uninformed readings, I would argue that if millions of people respond deeply to a poem without knowing much about its background, it seems churlish to tell them their admiration is ill-informed. And, in the end, despite its rich background, the poem is not obscure in the slightest. Part of its fascination—at least for those who want to delve deeper into what poems are and how they work—lies, rather, in how Yeats made such a complex, powerful emotion, so difficult to evoke, appear to be merely a moment’s thought.
What, then, does the almost universally admiring reaction to “The Lake Isle” constitute? It is a visceral response to a profound and ancient theme—the desire for a simpler, vibrant, even charmed rural life, which contrasts with colorless urban existence—all executed with tremendous formal verve. We can connect it with the Epicurean garden poem, with pastoral and eclogue, with Romanticism (particularly Shelley, as Warwick Gould has demonstrated),  with certain hermetic religious impulses and many other expressions of the same inclination to idealize a mythically primitive, rural existence. The creative power of this tradition, focused in this particular place and way in this lyric, trumps the author’s own retrospective and larger intentions and the historical or biographical material. If anything, those 121 words, in that particular order, are part of what give the historical and biographical material any interest whatsoever to begin with. This is neither a purely formalist argument, nor a form of ingenuous or disingenuous aestheticism. It is simply to return to the poem as a poem, an artful web of words, a subject and approach so frequently lost in the contemporary roar of biographical, historical and cultural criticism and theory, to the point that obvious things about the poem are overlooked, as we shall see below.
And form does matter in such discussions of poetry as poetry, as Yeats himself deeply understood, especially in the case of “The Lake Isle.” He emphasized it again and again in his discussions of the poem, right up to the end of his life. In an oral introduction to a recording of several poems with the BBC in 1931, he said:
I am going to read my poems with great emphasis upon their rhythm, and that may seem strange if you are not used to it. I remember the great English poet, William Morris, coming in a rage out of some lecture hall where somebody had recited a passage out of his Sigurd the Volsung, ‘It gave me a devil of a lot of trouble’, said Morris, ‘to get that thing into verse’. It gave me the devil of a lot of trouble to get into verse the poems that I am going to read, and that is why I will not read them as if they were prose. 
I have added the italics in the passage above to indicate how forcefully Yeats says those two particular words, “will not.” He is still emphatic about the importance of the versification of the poem more than 45 years after he wrote it. The first poem he then introduces and reads is “The Lake Isle,” delivered in what could be characterized as a highly formal, almost tonal style.
Yeats was highly consistent in his feelings about the poem’s craft. In another recording of the poem for the BBC, made in 1937, Yeats introduces “The Lake Isle” by saying:
Perhaps you will think that I go too near singing it. That is because every poet who reads his own poetry gives as much importance to the rhythm as to the sense. A poem without its rhythm is not a poem.
Yeats’s delivery of the poems in both recordings, a direct outcome of his passionate feelings about its versecraft, still inspires debate. Burton Roscoe has said that “Yeats read in a dreary monotone with care only for the beat of a measure.” Yet we know both from descriptions, but also from listening to early recordings of Tennyson and Browning, that Yeats’s declamatory style of recitation had its roots in the common practice of his youth. The Americans were already quite different by that time, if Whitman’s recording of “America” is any indication, though one would expect him to be atypical. Seamus Heaney, on the other hand, has recently and unequivocally defended Yeats’s delivery, describing it as an “elevated chant.” At the very least, we are lucky to have such recordings as examples of that dying style.  On a larger canvas, disagreements over Yeats’s style of delivery point again to the poem’s versification and prosody as a crux of its composition and force.
This is as it should be in the case of any strong poem. In the context of the discussion of the individual poem and the poet’s larger project, versecraft, or, more properly, prosody, defined as generously as possible to include the widest possible range of aural artifice in the poem (verse form, rhythm, diction and all other phonological contours of natural speech), is part of a larger art that transcends individual intentions and local politics. In its refinement the poet must look inward, crafting the music of particular works and even lines to give them force. At the same time the product of that word-by-word craft inevitably moves outward again, like ripples from a stone, connecting the poem as an individual work to the larger tradition and to the language as a whole. It has always been thus, and Yeats knew it. As he wrote near the end of his life, looking beyond his own work, in Part V of “Under Ben Bulben”: “Irish poets learn your trade, / Sing whatever is well made.” Important as “The Lake Isle of Innisfree” may be as an early step in Yeats’s larger journey, and complex and embroiled as it may be in Yeats’s own development and in the politics and culture of its time, one of the reasons that this lyric stands so well on its own is that it gives so much pleasure in such a brief space. It is itself one of those things well made, even if the speaker’s stated intention to move to a deserted island in a lough apparently went unobserved.
Although Yeats on occasion showed annoyance that “The Lake Isle” remained his best-known lyric, he nonetheless recognized its significance in his own development. Widely anthologized in the decade after its initial appearance, it was one of a handful of early poems that began to advance his career, and he returned to the story of its composition again and again in later years. Indeed, there are several stories to tell here: one about the poem’s composition; another about its publication and reception; and a third, that will concern us most here, about the poem itself.
Following Yeats’s own observations in a number of places, Finneran’s note in The Poems points out that Innisfree (Inis Fraoigh, “Heather Island”) is a small island in Lough Gill, County Sligo. This large lough is just to the southeast of Sligo town, where Yeats spent quite a bit of time as a boy. It has a number of such small islands, especially at its western end. The stories of the poem and of its composition begin together in this memory of childhood, before bifurcating into two separate narratives. As Peter D. McDonald and Russell K. Alspach have pointed out, Yeats himself, in a letter of 1888 to Katherine Tynan (discussed below), shows where the poem grew out of a prose passage in a draft of his novel John Sherman, which was published under the pseudonym Ganconagh in 1891, though written several years earlier. 
Here is the Ur-passage from John Sherman. It includes the dramatic situation of the poem as a fantasy of the narrator, but also includes a description of the situation that inspires the fantasy:
Delayed by a crush in the Strand, he heard a faint trickling of water near by; it came from a shop window where a little waterjet balanced a wooden ball upon its point. The sound suggested a cataract with a long Gaelic name, that leaped crying into the Gate of the Winds at Ballah . . . He was set dreaming a whole day by walking down one Sunday morning to the border of the Thames—a few hundred yards from his house—and looking at the osier-covered Chiswick eyot. It made him remember an old day-dream of his. The source of the river that passed his garden at home was a certain wood-bordered and island lake, whither in childhood he had often gone blackberry-gathering. At the further end was a little islet called Inniscrewin. Its rocky centre, covered with many bushes, rose some forty feet above the lake. 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 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.
In this amalgamation of memory, fantasy, and fiction we can trace the outline of the poem’s pre-history: personal memory transformed in a difficult time (Yeats was very unhappy in London) to fiction, then excerpted as a spot of time and worked up into a poem, then fixed in the larger arc of the poetry and work. The material of the poem and the moment of its composition both enjoy roles in this story, drawing the life and work together, for Yeats repeatedly retold the story of the poem’s origins (though it changed over time).
As Yeats told and retold the story he added new pieces to it. One such tributary is the addition of accounts of his own reading and other aspects of his family life, along with explanations of the mythic background of Innisfree. It is no surprise to learn that Lough Gill and Innisfree figure in local myth and legend, and in various places Yeats provides describes the poem’s mythic background. In a note to “The Danaan Quicken Tree,” an uncollected poem that appeared in The Bookman in May, 1893, that is also set on Innisfree, he writes:
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. But legends say that the berries were the food of the Tuatha de Danaan, or faries.
In Autobiographies, he expanded on this, in a passage that first appeared in Reveries over Childhood and Youth, from 1916, connecting it to his reading and to his own life (particularly his erotic desires), incorporating the mythic material about the island, and transforming the fictional passage from John Sherman back into memoir:
My father had read to me some passage out of Walden, and I planned to live some day in a cottage on a little island called Innisfree…I thought that having conquered bodily desire and the inclination of my mind towards women and love, I should live, as Thoreau lived, seeking wisdom. There was a story in the county history of a tree that had once grown upon that island guarded by some terrible monster and borne the food of the gods. A young girl pined for the fruit and told her lover to kill the monster and carry the fruit away. He did as he had been told, but tasted the fruit; and when he reached the mainland where she had waited for him, was dying of its powerful virtue. And from sorrow and from remorse she too ate of it and died. I do not remember whether I chose the island because of its beauty or for the story’s sake, but I was twenty-two or three before I gave up the dream.
Warwick Gould shows how much of this may come from Yeats’s memory, but he also carefully shows that it closely echoes W. G. Wood-Martin’s History of Sligo, which includes language reminiscent of the poem itself. Also noteworthy, as Alspach points out, is that Yeats here and elsewhere revises the story of the poem’s composition, making it originate as a poem tout court, when it in fact began as a piece of prose, fiction at that. Interesting as well in this passage is the way that all the background elements of the poem seem to come together—with the exception of the composition of the novel, neatly elided—only to be described as abandoned at just about the time of the poem’s composition, as if writing the lyric were an exorcism of the very longing or dream it describes. And this makes sense, for of course Yeats did not in fact arise and go to live alone in a hut on an island in a lake in the west of Ireland, but rather delved ever more into the cosmopolitan and literary life of his time.
Also fascinating, given the poem’s complete absence of any overt reference to eroticism, is Yeats’s frank admission of his desires and his plan to vanquish “bodily desire and the inclination of my mind towards women and love.” We will return to this crucial point below. Yeats does make it clear elsewhere that the poem grew out of unhappiness at his situation as a young, penurious writer in London. As he writes in Chapter 15 of Four Years, 1887-1891, published in 1921, “I saw nothing good” in London, though he did work hard and made a number of important literary connections to advance his career, attending, among others, the regular salons of William Morris. His prose recollections in Four Years echo John Sherman and the passion of the poems in a number of places, as when he writes in the same chapter that, one day:
I grew suddenly oppressed by the great weight of stone, and thought, ‘There are miles and miles of stone and brick all round me,’ and presently added, ‘If John the Baptist, or his like, were to come again and had his mind set upon it, he could make all these people go out into some wilderness leaving their buildings empty,’ and that thought, which does not seem very valuable now, so enlightened the day that it is still vivid in the memory.
Yeats thought enough of this passage and others from this book to reprint them in various editions of what would become Autobiographies.
Most importantly for our purposes, in Four Years Yeats again retells the story of the poem’s composition, with some interesting additional comments:
…but with women, apart from their intimate exchanges of thought, I was timid and abashed. I was sitting on a seat in front of the British Museum feeding pigeons when a couple of girls sat near and began enticing my pigeons away, laughing and whispering to one another, and I looked straight in front of me, very indignant, and presently went into the Museum without turning my head towards them. Since then I have often wondered if they were pretty or merely very young. Sometimes I told myself very adventurous love-stories with myself for hero, and at other times I planned out a life of lonely austerity, and at other times mixed the ideals and planned a life of lonely austerity mitigated by periodical lapses. I had still the ambition, formed in Sligo in my teens, of living in imitation of Thoreau on Innisfree, a little island in Lough Gill, and when walking through Fleet Street very homesick I heard a little tinkle of water and saw a fountain in a shop-window which balanced a little ball upon its jet, and began to remember lake water. From the sudden remembrance came my poem “Innisfree,” my first lyric with anything in its rhythm of my own music. I had begun to loosen rhythm as an escape from rhetoric and from that emotion of the crowd that rhetoric brings, but I only understood vaguely and occasionally that I must for my special purpose use nothing but the common syntax. 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.
Notice that here, as elsewhere, the story of the poem’s composition interweaves with observations about its craft, in Yeats’s insistence on the importance to him of the poem’s rhythm, music (for both of which we can read verse craft), rhetoric and syntax, “my first lyric with anything in its rhythm of my own music.” Notice also, again, the background of erotic desire and the pain such desire caused the young poet, suggesting that the most important word in the entire poem may be “alone.” As the picture of the poem’s composition richens, we now find its sources in Yeats’s own life and contemporary situation, and a fusion of these elements in a moment of pain (or the recollection of one) with verse form and sentences. Shorter versions of this story (with all erotic background completely excised) appear in the both the 1931 and 1937 BBC introductions, with various shades of other detail.
Once it appeared in print, the story of the poem itself—its history as a more public object—takes on a life of its own, and inspired Yeats to revise and simplify his account of its creation. In the oral preface to the BBC recording of “The Lake Isle of Innisfree” cited above, Yeats says that he wrote the poem when “about 23,” which would be the second half of 1888 and the first half of 1889, which may be true enough, though he had already roughed out what would become the poem in prose fiction (in the letter he also retells the story of seeing the ball on the jet of water in London). At this point, while we have early versions of the poem in letters (to Katherine Tynan in 1888, to be discussed more closely below), once it was published further revision was minimal, extending only to a very few articles and prepositions (again suggesting great care in the poem’s versification).
“The Lake Isle” was first published in W. E. Henley’s National Observer in 1890. Its first book appearance came in 1892, in The Countess Cathleen and Various Legends and Lyrics, Yeats’s second sequence of published lyrics. As all critics point out, the major revision of “The Lake Isle” after it had been first published as verse had to do with the placement of the poem in the sequence of the early work as a whole. Yeats did not use the sequence-title The Rose, in which the poem always later appeared, until the first edition of Poems, in 1895. Finneran places “The Lake Isle” as poem 24 in The Poems under that later sequence heading, and one way to read it in a larger context is as part of that setting. Further, as McDonald carefully annotates, the poem was widely anthologized as soon as it was published, making no fewer than five such appearances in the 1890s alone, and each one of those appearances gives it a slightly different cast.  As McDonald points out after looking closely at all the early appearances of the poem:
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 [in Henley’s National Observer, 13 December 1890], or a short lyric asserting the new purist imperatives of a young group opposed to the English poetic establishment [The Book of the Rhymers Club, 1892], or a poetic accompaniment to an illustration of an ancient Irish legend [the Religious Tract Society’s Leisure Hour, August 1896], or a particularly clear expression of Arnoldian Celticism [William and Elizabeth Sharp’s anthology Lyra Celtica, 1896], or an especially ‘fine’, but also fortunately topographical, example of the latest achievements in the tradition of Irish poetry in English [Stopford Brooke and Rolleston’s A Treasury of Irish Poetry in the English Tongue, 1900]. Of course, whether or not these readers actually interpreted the poem in these ways is another question . . .
The interesting thing for our purposes is how popular the poem became as soon as it appeared, and how it seemed to fit so easily into so many different political, and cultural contexts. McDonald makes a strong case for the many ways in which the poem was likely to be viewed by the readership of the various journals and books where it appeared. No doubt some of this had to do with Yeats’s own political and entrepreneurial acumen in placing his work. At the same time, it is fascinating that the poem appeared to have satisfied such different groups of readers, or at least the editors serving them, in terms that are so different from each other. Something else to remember, however, is that we do now need painstaking scholarship such as McDonald’s to reconstruct this situation in the first place. Few contemporary readers, over a century later, are likely to read the poem in any of these contexts—and yet its popularity remains nothing if not strong. A Google search for “The Lake Isle of Innisfree” in July 2007 brought up almost 40,000 web pages, and an hour’s random sampling suggested many actually contain the text of the poem, often with blog encomia to its beauty. The explanation for this is more likely to lie in the words themselves than in any story of their composition or account of their cultural, social, and political import at the time. For real poetry, however rich its contexts, is a living thing, whenever it was written.
Crucial as all of the personal, cultural, political and historical background is in understanding how Yeats composed, revised, placed, and thought of “The Lake Isle” and the contexts in which it may have been read, none of this describes how the poem works as a poem, which is the very reason we do remember it to begin with. In exactly this context, the question of verse form and its interactions with diction and syntax comes up frequently in Yeats’s own discussions of the progression of his earlier work generally, sometimes quite forcefully. As he wrote in 1937, in the unpublished essay that would eventually appear (in 1961) as “A General Introduction for My Work,”
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.
This resonates with Yeats’s 1921 criticism (quoted above) of the archaisms and syntactical inversions in “The Lake Isle,” giving this train of thought a long lineage in the poet’s meditation on his craft. Still, in that earlier poem we do find some of the same nascent desire to create a poetic music in which powerful and passionate syntax correspond with traditional meters, and in which periods correspond with stanzas. Despite some archaic diction and the inversions of which Yeats was later critical, every line in the poem is endstopped and each stanza is punctuated as a single sentence.
Despite his later criticisms of the poem, Yeats clearly does still find seeds of his mature “rhythm” in it. The scope of Yeats’s project as a maker of verses that could contain great poetic passion and power, and the importance of “The Lake Isle” as an early step in that project, is borne out, as we saw above, in the recordings he made for the BBC. And now we come to what appears to be an overlooked and crucial fact, for the allusion to William Morris that Yeats makes so forcefully in his oral introduction to his own reading is telling.
There has been a good deal of writing about Yeats and William Morris, whom Yeats always admired in a number of ways despite his later criticism of the poetry and the politics. During his time in the late 1880s and 1890s in London, the young Yeats spent quite a bit of time with Morris while working on the poems that would become The Rose. We know that Yeats met Morris in April of 1886, at the Contemporary Club, and that in mid-1887 he started to attend Morris’s regular Sunday night salon. Peter Faulkner has written that “on the evidence of the poetry it is impossible to trace a direct influence from Morris on Yeats: the most that can be said is that we are aware of their common participation in the Romantic tradition at a stage when this had lost its initial freshness and become a literary convention.”   Faulkner is nothing if not thorough in annotating Yeats’s many interactions with Morris and his writing about him; others, such as D. M. Hoare and even Eliot, have pointed out the connections as well. Yet no one, so far as I know, has traced Yeats’s reference, in the BBC introduction to his reading, to Morris’s discussion of “rhythm” with respect to Sigurd the Volsung, which had appeared in 1876. And this is interesting, for “The Lake Isle” is one of Yeats’s very few poems in hexameter—which is the identical meter to Sigurd.
Morris’s long poem opens as follows:
There was a dwelling of Kings ere the world was waxen old;
Dukes were the door-wards there, and the roofs were thatched with gold:
Earls were the wrights that wrought it, and silver nailed its doors;
Earls’ wives were the weaving-women, queens’ daughters strewed its floors,
And the masters of its song-craft were the mightiest men that cast
The sails of the storm of battle adown the bickering blast.
There dwelt men merry-hearted, and in hope exceeding great
Met the good days and the evil as they went the way of fate:
There the Gods were unforgotten, yea whiles they walked with men,
Though e’en in that world’s beginning rose a murmur now and again
Of the midward time and the fading and the last of the latter days,
And the entering in of the terror, and the death of the People’s Praise.
This long book includes some of Morris’s most memorable verse, not least because of the mastery of the end-stopped hexameters for the long narrative job.
There are a number of things to point out about Morris’s verse in Sigurd. First, while we could call the lines hexameters, it is important to note that the lines fit only uncomfortably into an accentual-syllabic scheme. Rather, they scan most easily as ballad-like stress-based lines. Morris is counting stresses and not worrying too much about strict syllabic regularity. Further, the lines divide clearly into three-beat hemistiches. While many of the lines offer an obvious caesura at the mid-point with punctuation, in others the distinction is more subtle, involving, for example, only the placement of a conjunction. Morris, however, provides another key marker to his verse by almost always making the syllable before the hemistich break unstressed:
There was a dwelling of Kings | ere the world was waxen old;
Dukes were the door-wards there, | and the roofs were thatched with gold:
Earls were the wrights that wrought it, | and silver nailed its doors;
Earls’ wives were the weaving-women, | queens’ daughters strewed its floors,
And the masters of its song-craft | were the mightiest men that cast
The sails of the storm of battle | adown the bickering blast.
And so on. In this passage, as is typical throughout the poem, the first hemistiches of all but the first line end in unstressed syllables. Yeats’s versification in “The Lake Isle” is similar in almost all respects, as we will see in detail below.
There are a number of other things to say about the similarities of the verse lines between Yeats’s lyric and Morris’s epic. All of Morris’s line-end rhymes—for many hundreds of pages—are end-stopped primary stresses with no trailing unstressed syllables. While Yeats uses an abab stanza, he follows this practice faithfully. In Morris, there is an echo of alliterative metric throughout, though it is at most a secondary organizing principle. Many lines contain paired alliterating stresses such as world/waxen, dukes/door, wrights/wrought, wives/weaving/women, mightiest/men, and so on. The alliteration rarely spills over the hemistich marker, thus only reemphasizing that boundary within the line. Yeats uses less of this, and also employs repetition, which creates identity and alliteration across the hemistich, but we do see alliteration in “The Lake Isle” in a number of places, e.g., “go/go,” “have/hive/honey,” “live/alone,” “peace/peace” “glimmer/glow,” “lapping/low,” and “hear/heart’s.” (some of these are discussed further below).
Morris’s choice of meter makes sense given his sources in Icelandic and his knowledge of and admiration for the classics and Old English forms as well. He is self-consciously crafting a form that draws on ancient sources to create a powerful verse form for his mythic narrative, which is a key to its success. He was of course not alone in his admiration for these forms at this time; Longfellow had similar interests which he put to use in his own narratives, such as Evangeline and Hiawatha. At any rate, it seems fair to say, given Yeats’s own reference to Morris, that this had something to do with his own choice of meter for “The Lake Isle.”
All of the foregoing prepares us finally to look as closely as possible at how “The Lake Isle” itself works as verse. When we turn to the early version of the poem that Yeats sent to Tynan in that letter of 1888, what we see is instructive and brings us as close as we are likely to get to understanding what happened to make the poem so memorable in its process of revision:
I will arise and go now and go to the island of Innisfree
And live in a dwelling of wattles, of woven wattles and wood work made.
Nine bean-rows will I have there, a yellow hive for the honey-bee,
And this old care shall fade.
There from the dawn above me peace will come down dropping slow,
Dropping from the veils of the morning to where the house hold cricket sings;
And noontide there be all a glimmer, and midnight be a purple glow,
And evening full of the linnets’ wings.
Not bad, but not great. And yet much of the imagery is the same—and Yeats was surely the same person who wrote the famous revision, if a bit younger here. In the far tighter revision that has remained so popular, what has changed is not the person who wrote the poem, or the poem’s general semantic content, but rather the poem itself: a few of the words. And the signal thing that Yeats has done is to conform his versecraft to what appears to be his model in Morris, reducing from seven beats to six the number of strong stresses in each of the first three lines of each of his stanzas. More specifically, we see a reduction from a 3+4 stress pattern across the hemistich caesura, to 3+3. It is impossible to come much closer than this in viewing a poet’s verse-compositional process, but if we simply pull extraneous beats from the second half of the lines in the earlier version, we wind up with something quite close to the final in many lines:
I will arise and go now and go to [the island of] Innisfree . . .
Nine bean-rows will I have there, a [yellow] hive for the honey-bee . . .
Dropping from the veils of the morning to where the [house hold] cricket sings . . .
We can see a similar, if somewhat more complex process in the remaining lines:
And noontide there be all a glimmer, and midnight be a purple glow
becomes the much improved
There midnight’s all a glimmer, and noon a purple glow
This reduces the syllable count from 17 to 13, and again cuts out the fourth stress in the second half of the line.
It is well nigh impossible to know how this revision process occurred, how many drafts were discarded, and so on, but in looking at these two versions, given the patterning of the process, we can allow ourselves to speculate a bit. The interesting thing about the glimmer/glow line is the fungibility of “midnight” and “noon.” In the earlier draft, noontide is a glimmer and midnight a purple glow; in the final draft, this is reversed. Why? Well, there is no single-syllable reduction available for “midnight” as there is for “noontide.” So, assuming Yeats had determined to reduce the lines to 3+3 accentual hexameters, he might have begun with
There noontide’s all a glimmer, and midnight a purple glow
But then, if he wanted to use a somewhat less anachronistic diction and fewer syllables, that would lead to
There noon’s all a glimmer, and midnight a purple glow
From here the only question would be: why reverse “noon” and “midnight”? It seems unlikely that he suddenly decided that he had the images “wrong.” After all, this is the writer who, in “An Introduction for My Plays” (from 1937, but not published until 1961), wrote that “I have spent my life in clearing out of poetry every phrase written for the eye, and bringing all back to the syntax that is for the ear alone.” This is a revision for the ear, not the eye.
Taking Yeats’s statement of his prosodical and rhetorical intentions as a kind of license, I would speculate—and I hasten to reemphasize that this is the purest kind of speculation—that Yeats may have thought, rightly, I believe, that “noon’s all” crowds the stresses very closely together and the line would be better with an intervening unstressed syllable there, leading to:
There midnight’s all a glimmer, and noon a purple glow.
While Yeats did crowd stresses together in a few other lines, e.g., “a small cabin” “Nine bean-rows,” ”low sounds,” perhaps he felt that in this case such crowding didn’t work as well.
An analysis that takes into account the descriptive categories of IPA (The International Phonetic Alphabet) may help to explain Yeats’s intuition here, for “noon’s all” not only crowds the stresses, but those two stresses, with all their consonants, are also quite difficult to pronounce in succession. In IPA terms, |n| is an alveolar nasal consonant. It requires placing the tip of the tongue against the gum ridge just behind the upper teeth, on the upper alveolar ridge; the consonant is vocalized (the vocal chords must sound) and air escapes through the nose, not the mouth. This motion must occur twice in quick succession in the word “noon,” hard enough to do. But then, in “noon’s all,” the next consonant, after the apostrophe, is |z|, a voiced alveolar sibilant, creating the consonantal cluster |nz|, followed by |l|, which is an alveolar lateral approximant, all three of them in the context of strong prosodical stresses. The final two consonants |z| + |l|, are oral, not nasal, so while the tip of the tongue is in the same place as with the two |n|’s, the middle of the tongue must now move downward to allow air to escape off its sides. Readers unfamiliar with IPA can get some sense of how this works by vocalizing |l| and |n| in one vocal stream while holding the back of the hand very close to the mouth, directly in front of it—on |l| you will feel a breath; on |n|, none. The bottom line: pronouncing |n| + |nz| + |l| in quick succession is a tongue-twister, all that much more difficult when all the syllables involved are stressed.
In contrast with “noon’s all,” “midnight’s all” utterly changes the phonetic contour of the line. |m| and |n| may stand next to each other in our alphabet, and look similar, but in terms of their phonetic production they are utterly different. |m| is a bilabial nasal; the tongue can be anywhere in the mouth without affecting the sound much. This makes it far easier (even though the action is unconscious) to set up for the next consonants, whatever they may be. In this case, that is two consonantal clusters, |dn| followed by |ts|, leading finally back to |l|, as in the earlier version of the poem. Further, in the final version, “-night’s” is unstressed, and the reduced volume, pitch and duration allow easier movement across that syllable. It is worth noting, however, that “-night’s” ends with the unvoiced consonant cluster |ts|, at the end of an unstressed syllable. The unstressed, unvoiced cluster is, I believe, far easier to move off of than the voiced, stressed cluster |nz|. In the end, I think there is good support for the notion that the earlier version, “noon’s all aglow,” trips the tongue, whereas the latter, “midnight’s all aglow” trips off it.
In both of his BBC recordings of the poem in the 1930s, Yeats refers to the glimmer/glow line in a similar way. In the 1931 introduction, he says: “I think there is only one obscurity in the poem; I speak of noon as a ‘purple glow’: I must have meant by that the reflection of heather in the water.” Perhaps so, yet his own revision process and many meditations on his own poetics suggest other things mattered more. The image works well enough – whatever he really meant was intended for the ear more than the eye.
Here it is worth noting that the one significant variation in the poem that was anthologized in Leisure Hour (October 1896) is in line 10, which in that version runs “I hear the lake water lapping with low sounds by the shore,” inserting “the” before “lake.” In his thoughtful discussion of this appearance of the poem, Warwick Gould writes that “The addition . . . destroys the rhythm of the line and seems unlikely to be authorial.” Perhaps true—it never appeared again. But the choice does appear to have a purpose. In the canonical version, the strong stresses fall as follows:
I héar lake wáter lápping | with lów sóunds by the shóre
“Lake” bears some stress, but it is secondary at most. The Leisure Hour version of the line, however, would scan as follows:
I héar the láke wáter lápping | with lów sóunds by the shóre
No doubt whoever added that “the,” whether it was Yeats or not, was seeking to promote the stress-value of “lake” to foreground the alliteration with “lapping” and “low.” This occurs because the intervening unstressed syllable allows the next one to increase in prominence. The problem is that it doesn’t quite do enough to demote the stress on “water,” giving the line seven stresses instead of six, or, arguably, if “water” is demoted a bit to give the line six stresses, still slowing the line down significantly. Whatever scansion one chooses, that, I would argue, is why the revision was not retained in subsequent publication.
My reconstruction of Yeats’s revision process is, as such things always must be, a prosodic and linguistic fantasia. Still, I would like to think that the poet who didn’t give a visual hoot whether or not it were “midnight” or “noon” that’s “a glimmer,” and who relineated Pater as poetry for the Oxford Book of Modern Verse, would not feel too put out by such speculation. After all, as Yeats himself repeatedly points out, the rhythm and music of the poem were crucial to him. He tightened the screws, found the right rhythm, and the poem became the one we know. The poem itself ramifies this; the speaker does not so much see a vision as “hear it in the deep heart’s core,” a verb he repeats and places at the lyric’s climax, “bringing all back to the syntax that is for the ear alone.”
There is still much more to discuss about prosody, broadly defined, in this poem. At times the patterns begin to seem endless. Consider the extraordinary amount of repeated lexical material, which works with the verse patterning to emphasize it even further. Look at the repeated words, such as “and go,” “there,” “peace,” “dropping,” and of course “I.” Interestingly, every single one of these repetitions occurs across a hemistich boundary, or in a position to emphasize a parallel occurrence in a contiguous line. For example, the repetition of “and go,” in the two subsequent beats of line one surrounds hemistich boundary within the line: “. . . and go now, | and go . . .” It therefore serves to emphasize that boundary immediately, syntactically foregrounding the metrical pattern that Yeats then follows throughout the poem. This happens yet again in the first line of stanza two, “some peace there, | for peace.” As discussed above, these repetitions resonate with other alliterative patterns, but they also resonate in the syntax, strengthening its reiterative patterns as well.
A similar alliterative phenomenon occurs in line three, by alliterating on the strong stresses of “I have there” with “a hive for”; and; and even in the alliteration of the second line of the last stanza, although the stresses don’t quite line up here: “lake water lapping | with low sounds” (hence the contemplated addition of “the” in the Leisure Hour version, discussed above). And the repetition of “dropping” in the first and second lines of the second stanza (“. . . dropping slow, / Dropping . . .”) serves a similar function in that once again it is first used near the end of a hemistich, then at the very beginning of the next, although in this case there is a line boundary as well.
Perhaps the most important of these repetitions occurs in the final stanza, with the phrase “I hear,” at the beginning of the second and fourth lines of the stanza. This repetition is not only the verb that closes out the poem, but is so powerful and famous it seems hardly to be read clearly by most who encounter the poem. For in this case the repetition creates an extended metaphor, identifying the sound of the water lapping with what one hears in “the deep heart’s core.” The final phrase remains odd—for presumably it would make more sense to refer to “the heart’s deep core.” Within the body, what one might hear in the heart is only the heartbeat, and to compare that sound to the “lake water lapping” retains its power to enchant, as if the lake were coursing through the speaker’s very veins, not just his memory and imagination. But then what remains is the question of what the “deep heart” might mean. We will return to this below.
All in all, there is an astonishing amount of repetition in the poem, not only in the inevitable reiterations of the verse form, but also lexically, syntactically, and rhetorically. In all of this, Yeats presumably wanted us to feel the hemistich structure very strongly, for he not only built it systematically into each line, but emphasized it again and again with echoes and repetitions, and by ending lines with identical prosodic contours, even repeating words at the end of hemistiches (“there”) and actually repeating words across the hemistich boundary. The placement of “build there” (line 2), “have there” (line 3), and “peace there” (line 5) at the ends of initial hemistiches serves to reinforce, lexically, metrically, rhythmically, syntactically, and rhetorically, the very strong juncture mid‑line in each of the first three lines of each stanza. It is a five-fold formal redundancy, stronger than which there is none. Further, it is significant that the repetition of “there” in this position occurs at the ends of the first hemistiches of three of the first four lines in the poem containing six beats:
And a small cabin build there, of clay and wattles made:
Nine bean-rows will I have there, a hive for the honey-bee;
And I shall have some peace there, for peace comes dropping slow,
This drives the pattern home as hard as possible quite early on in the mind of reader or listener. Doing all of this at once is not an easy trick, when it must also appear a mere moment’s thought—and make sense. No wonder even Yeats was willing to say, almost half a century after writing it, that it was difficult to do and he would not pretend otherwise in his performance of it for the BBC.
Just a bit more. Notice that, as in Morris’s Sigurd the Volsung, the syllable that begins each second hemistich in every line of “The Lake Isle” is unstressed, and is often repeated, creating both reiterated syntactic patterns and aural similarities. In order and grouped by stanza, these words are: and, of, a, in; for, to, and, of; for, with, or, in. In Yeats’s poem, the pattern is even more pronounced than in Morris, as each unstressed syllable is a single syllable word. All are prepositions or conjunctions, with one article (“a”). In the second part of each fourth line the pattern is particularly tight, as each is a prepositional phrase:
And live alone | in the bee-loud glade
And evening full | of the linnet’s wings
I hear it | in the deep heart’s core
Notice that the first two stanza-end lines divide into two groups of two stresses each, but the final line divides into one stress followed by three, giving it a different and final cadence. In any event, the simple noting of the caesura reveals yet more patterning as we examine syntax throughout the poem, for it shows us how Yeats has lined up syntactic segments with prosodical ones. This could not possibly occur by chance.
Indeed, throughout the poem each second hemistich in some sense echoes or extends the first. Look at the first three lines of each stanza—the second hemistich almost always extends the meaning of the first, either repeating the verb or extending the meaning already articulated in the first half of the line, in a number of ways (prepositional phrases, conjunctions with parallelism, adverbial phrases). Note that many of the second hemistiches throughout the poem have no verb at all.
Indeed, we have language that makes good sense, even a kind of poem, if we simply eliminate all those second hemistiches and repunctuate accordingly:
“The Lake Isle of Innisfree”
I will arise and go now,
And a small cabin build there.
Nine bean-rows will I have there,
And live alone.
And I shall have some peace there,
Dropping from the veils of the morning.
There midnight’s all a glimmer
And evening full.
I will arise and go now, [for]
I hear lake water lapping.
While I stand on the roadway,
I hear it.
The point is not that the second hemistiches are superfluous; they are of course utterly necessary. Indeed, they make the poem. The point is that this rewriting exercise helps to show us some of how they work, as codas, or echoes, or extensions of each active statement. To extend the “hear”/”heart” metaphor of the final stanza (and the orthographic similarity of the words hardly seems to be a coincidence, now, as does the fact that “hear it” and “heart” sound almost identical), the second hemistich of each line acts like a diastole—a relaxation and filling of the ventricles—to the more active systolic beating of the first hemistich in each line.
The four-beat final lines of the three stanzas show other interesting echoes and reiterations. The first two begin with the word “And”: “And live alone in the bee-loud glade”; “And evening full of the linnet’s wings.” The last, “I hear it in the deep heart’s core,” repeats the word “I,” which stands at or near the beginning of each of the three previous lines in that stanza, along with the beginning of lines 1 and 5 of the poem as a whole. In a poem full of codas they are the codas of codas, standing at stanza end. Indeed, the entire poem could be read as a terrifically powerful first hemistich, followed by an eleven-and-a-half line coda or series of codas. And of course the beginning of the third stanza’s first line repeats that opening clause, the poem’s central statement. This poet knew exactly what he was doing, and these particular words in this particular order is what it is about.
It takes patience to evoke the prosodical complexity of this apparently simple poem, its astonishing qualities of reiteration and repetition, the way that Yeats overdetermines them in every linguistic dimension available. Obviously, there is much more to say—one could go on at great, great length. The larger point is that this Jakobsonian dissection and cataloguing of effects should make it clear—as he emphasized throughout his life—that Yeats had to be very much after them; they are too self‑reinforcing and carefully repeated to be accidental or arbitrary. He has assembled every linguistic resource on which he can lay his hands to build a world of words.
The fascinating thing that finally emerges from all this is, to borrow a term from Amittai Aviram, that that the poem’s versification, prosody, diction, syntax and rhetoric are tremendously redundant and “overdetermined.”  For not only are there particular patterns of stress, but rhetorical and syntactical repetitions and echoes within them as well. When we anatomize them, we wind up with an astonishingly complex piece of language, so highly stylized and with so many apparent rules that it only seems that much more miraculous how effortless it appears in the end. And we could keep going, for the multiple patterns foreground every word. For example, we could look at the potential differences between “will” and “shall,” and consider Yeats’s choice of prepositions . . . but we will not.
Given all the above, if someone were to try to create a generative menu of rules that Yeats followed in creating “The Lake Isle,” it would include at the very least the following:
1) Four line stanzas;
2) Rhyme scheme abab;
3) All lines end-stopped;
4) First three lines of each stanza are stress-based hexameters;
5) Final line of each stanza is stress-based tetrameter;
6) Most lines have some strong stress alliteration (even if it rests on repetition);
7) All lines divided into hemistiches;
8) Each first hemistich ends with an unaccented syllable;
9) Each second hemistich begins with an unaccented, single-syllable word;
10) Each stanza is a complete sentence;
11) Most first hemistiches have active verbs;
12) Most second hemistiches are prepositional phrases or syntactical codas of some kind;
13) Key verbs and nouns repeated throughout (arise, go, peace, have, dropping, hear);
14) First line of first stanza and first line of last stanza share the same first hemistich.
This is not an exhaustive list, just a prosodical skeleton. And I hope the reader doesn’t think I am suggesting Yeats made such a formal list. At the same time, I hope the reader doesn’t think that such a high level of order occurs by accident—that would be like discovering a perfect circle in nature. And I do think that these are some of the rules that, in this case, helped Yeats to discover the “rhythm of my own music.”
So now is the time to speculate on a larger canvas. Surely the verse, prosodic, and syntactical structure of this poem, embodied in its tightly controlled stanzas, and which, once set and followed so closely, had to precede the actual words in most of the lines, has something to do with the lasting power of those words. For considering the astonishing amount of structure and reiteration in the poem, it seems on the one hand understandable that Yeats should have, throughout his life, made such strong statements about how hard he worked to create it; and, on the other, somehow impossible that it should embody such a specific feeling of languid remembering and desire, such profound, expressive longing for peacefulness. How does one follow such a tight pattern and yet still create an emotion that seems so particular, so precise, so calm and clear? Why is it that language actually seems more alive, not less, when placed under such strict rules? And not only alive, but highly memorable in its specifics? And why, in this case, these particular rules?
After all, we have to be careful about associating this particular meter, set of rhythms, diction, syntax and rhetorical tactics with particular emotions—for, as in a famous experiment, we could minimally alter just a few syllables, retaining virtually all the prosodic and syntactical structures intact, and suggest something utterly different, even nonsense:
I will arise and blow now, and blow upon her knee,
And a small rabbit kill there, of gray and mottled shade:
Et cetera. So how can the specific emotions of “The Lake Isle of Innisfree” be said to lie in the specifics of its prosodic structure, or other linguistic patterns? For they only work when manifest in specific words. In particular: these words.
Similarly, the alternation from the lines of two groups of 3 beats to two groups of 2 at the end of each stanza has a powerful effect (including the final line variation of 1+3), but as to why—well, why nine bean rows? I don’t think Yeats knew himself, and the arbitrary, or at any rate occult, quality of that “nine” may be something of a clue to his awareness of that, or at least a bow in its direction.
Perhaps one of the reasons the poem is so powerful is that the poet simply gives himself over to its rhythms. Who knows why they liberated him to write such a powerful lyric? But they did. Something about this verse form, fused with a highly foregrounded set of lexical, syntactical and rhetorical repetitions and reiterations, makes the words memorable. Perhaps in this form Yeats found a way to let rhetoric follow verse structure so closely that the two seem to become one—a great Romantic goal. In its concatenation of codas, Yeats makes the syntax reiterate itself obsessively again and again, just as verse forms do. Indeed, as we began to examine above, “and” is one of the most important words in the poem. Notice, for example, that of the 24 hemistiches in the poem, fully six begin with “and,” a very high ratio, comparable to the anaphoric structures of Biblical poetry. Further, all six of the “and” hemistiches occur in the first two stanzas, making the ratio there 6:16. Yet Yeats is careful not to allow the poem to begin to have too much the feeling of dense asyndeton; no single line has more than one “and.” This means that six out of the poem’s first eight lines have one of these hemistiches. I doubt this is a coincidence.
Still, the notion of prosody simply liberating creativity is an unsatisfying answer, that tends toward the mystical: that certain rhythms unleash or free creativity in certain cases. This would reduce poetry to a shamanistic ritual in which one calls up certain rhythms and voilà, lyricism drips like water from stone. But there is too much conscious craft at work here to convince me of this, Yeats’ own views about spiritualism notwithstanding. I think much of the answer lies elsewhere and has to more to do with the quality of relatedness that the metrical abstraction heightens, and that Yeats knew how to intensify, with purpose, to dizzying heights—some of it learned for the first time in the making of this poem. Remember that this poem describes a fantasy; the speaker “stand[s] on the roadway” dreaming of Innisfree, but he is not in that place of which he dreams. In effect, he is nowhere, or rather, the place he is can only be fully described abstractly, as a kind of imaginative mixing between “the pavements grey” and “the bee-loud glade.” He, and therefore his reader as well, is in a world of words that creates resonances and relations. At the same time, as short as the poem is, it forestalls the very action the speaker describes he will take in the future, arising and going to Innisfree, for as long as it continues as a poem. The highly self-conscious and artful language of the poem measures and strains against itself, not merely by activating some prosodic squeeze-box, but by forcing us to contemplate both what it is—a description of the future—and what it is not—that future action itself.
The problem with treating the poem in this way, as a kind of description, is that it diminishes it to a mere thing, which it is not. So perhaps it is better not to see the poem as a description of an action at all. Instead, it is probably best understood as an action itself, a speech act, the kind of action possible only in words. In this case, the kind of action it would be is a vow. It is a vow that the speaker reiterates again and again, in almost every verb, saying that he will arise, and go, and build, and have, and live, and so on. The tumbling reiterations of those passionate, redundant vows, are not reinforced or described, but rather embodied in a language that is itself passionately redundant to the point of a chant, a chant of vows in which the speaker strives to make the words conjure what they promise. And the longing, the strains of music, emerge in the separation between reality “on the roadway” where the speaker is not going anywhere, but only “stands,” motionless, ungoing, and the place he would go, Innisfree. So the poem is not an action of going, but rather an action of desiring and vowing to go. The rhythms and reiterations of the poem’s vows echo each other relentlessly and their beauty comes from the way such reiterations echo within us in our own lonely desires.
This is where we might begin to understand “the deep heart’s core” somewhat better. For if what one hears in the heart is the heartbeat, perhaps what one hears in the “deep heart’s core” is the even more profound rhythm of desire, without which no heart could exist in the first place. The low sound of lake water lapping that he hears is his heartbeat and also his spirit. By returning in memory and imagination to an enchanted and peaceful place, and giving form to the longing to be there alone, Yeats conjures one shape of that desire, one of its forms. The reality of his life—of any life—is not the fact, or not only the fact of where he may be; it is also the rhythm of desire. Yoking, through art, his own heartbeat to the deeper rhythms of memory and longing, Yeats elevates mere versecraft to another realm, where the verses, as if by magic—a very real magic, requiring no occult explanation—enact what they say. In other words, what Yeats hears “in the deep heart’s core”—and what his artifice, syllable by syllable in this poem, ultimately helps us to hear in our own “deep heart’s core”—is this poem. That is why it remains so justly popular. It is not “about” enchantment; it enchants.
Now we are somewhat closer to the source. For if the poem does enchant, and I think it does, I must directly contradict what I suggested earlier when treating it more as a description, and say that its vow is fulfilled—we do indeed “go” to Innisfree, albeit not the real place. For, to state the obvious, if the poem were merely a travelogue no one would still be reading it. Skeptics at this point should note another movement in the poem’s syntax: that it begins in the future tense . . . and then in each subsequent stanza, moves towards the present, where it ends. And that is a delightful paradox: to begin in the future but end in the present. How could this be an accident? A work of artifice, the poem draws so deeply on the language that it purposefully becomes what it expresses—both the vow and, in a way, the fulfillment of the vow. The poem is not about “Innisfree,” it is about the desire to be there, and it does not describe that desire, but rather enacts it, as verse form, syntax, rhetoric, and speech act; as art. In this sense, we have all been to Innisfree. We have been there at least insofar as we desire what it represents to this speaker and thereby to us.
One great irony in this expression of desire for peace, also clearly part of Yeats’s designs upon us, is the anguish that this poem about peace embodies. It may appear to articulate a desire for peacefulness and even fulfill a vow to go where it is, yet it is written out of the opposite of peace—it is filled with “a devil of a lot of trouble.” The cry of desire to be alone and at peace comes from a crowded place characterized by the endless workings of desire—roadways and pavement. One imagines the speaker standing on the roadway staring into space and thereby risking injury. It is a chant of desire against itself, and it is that tension, that anxiety, that terrific internal dilemma, that argument with itself that ultimately calls forth the strains of this poem. For if not-desiring were simple, we would have no poem. Instead, what we have is a fully realized conflict between desire – very much including erotic desire – and peace, suggesting that what the poet wants may be something quite different: both the peacefulness beyond desire, yet also desire itself, and desire’s ends, explicitly meaning women and love, as embodied in this poem where they are conspicuous by their absence. For one cannot desire to “[conquer] bodily desire and the inclination of my mind towards women and love” unless one feels it powerfully to begin with. In this, “The Lake Isle” reminds me of, for example, Van Gogh’s “Starry Night,” a peaceful scene filled with turbulent emotion, where even the stars seem uncomfortable with the emptiness that should surround them.
Real poetry is rarer than we think. We may murder to dissect, but “The Lake Isle” is so vital a poem—in particular, a poem of erotic longing and its contradictions—that I have little doubt it will survive any analysis. It is, to appropriate a phrase from Harold Bloom, the achievement and prolongation of an anxiety rather than its resolution. That is why, as long as desire thrives—alongside its apparent opposite, the desire for peace—and people read about it all in English, this poem has a good chance to remain alive. And at the heart of that vitality, as Yeats understood, is the fact that the poem’s expression of erotic desire, and simultaneously his desire for release from it, are both grounded in a self-conscious and intense focus on verse form, prosody and linguistic material at every level, however languid or hypnotic an effect that formal flow may appear to have on us. In his oral introduction to the recording cited near the beginning of this essay, Yeats explicitly referred to the creation of the poem’s prosodic contours as work, or trouble—“a devil of a lot of trouble”—and I think, far more than any historical material, however interesting it may be, it is exactly that visionary trouble with language that enabled Yeats to articulate a poem out of the whirlwind of language and memory, anxiety and desire, balancing them the way the movement of a heart balances itself.
Many contemplate the time in which they live—few can make a poem like “The Lake Isle” in it. This is hardly because of a lack of intelligence, but rather because art at this level requires an obsessive skill that very few possess. Yet only in the possession of such fierce skill can poets speak for us, rather than merely to us. The question of which form Yeats used is therefore less important than the way in which he worked it. For in the end, it is artifice that poets use to forge relatedness from the chaotic forces of nature, however recurrent those forces may appear when we use words and numbers to describe them. And that successful artifice, rampant on a wave of intense reiteration, is exactly why “The Lake Isle” will continue to survive, like all strong poetry, long after the social, cultural, and political forces that surrounded its author, and in which he played his own part, have faded into mere scholarly fodder.
  Peter McDonald, “A Poem for all Seasons: Yeats, Meaning, and the Publishing History of ‘The Lake Isle of Innisfree’ in the 1890s.” Yearbook of English Studies 29 (1999): 202-30.
  John Unterecker, A Reader’s Guide to William Butler Yeats. New York: The Noonday Press, 1959.
  Warwick Gould, “Yeats as Aborigine.” Four Decades of Poetry, 1890-1930 2.2 (1978); 65-76.
  Documentation I have seen is surprisingly unclear about the dates of Yeats’s BBC recordings. The Collected Works of W. B. Yeats, Volume X, Later Articles and Reviews, edited by Colton Johnson, is the most reliable and gives the dates of the two I have cited, in order, as September 8, 1931, and October 29, 1937. I am grateful to Patrick Muckleroy, of Savage Library at Western State College of Colorado in Gunnison, for invaluable assistance in tracking down information on these recordings and for other research support.
  “Seamus Heaney on William Butler Yeats,” in Poetry Speaks. Ed. Elise Paschen and Rebeka Presson Mosby. Naperville, IL: Sourcebooks, Inc., 2001.
  Russell K. Alspach, “Yeats and Innisfree.” Yeats Centenary Papers, III. Dublin, The Dolmen Press, 1965.
  The only significant change seems to have been in the 1896 anthology appearance in Leisure Hour, which was limited to some punctuation and one additional word, later omitted. This is discussed below.
  Amittai F. Aviram, Telling Rhythm. Ann Arbor: University of Michigan Press, 1994.