[Reprinted with permission from The Atlantic Monthly]

THERE is only one full collection of Frost's poems in print- The Poetry of Robert Frost, edited by Edward Connery Lathem -and the text is corrupt: the editor has altered the rhythm of Frost's poems by repunctuating them. Although several of Frost's critics have complained of these alterations, no one has noted their full extent and nature; most anthologists and many critics now reprint the corrupt text. Lathem has removed commas, added commas, removed hyphens, added hyphens, made words compound, added question marks, and altered dashes. Besides regularizing quotation marks (double quotes for single seems a tolerable change), the editor makes by my count 1,364 emendations, of which his notes justify 247 by reference to earlier printings. Thus he makes 1,117 changes for which he offers no textual sources, an average of 3.4 for each poem.

Frost wrote the line 'The woods are lovely, dark and deep'. We do not find this line in The Poetry of Robert Frost. Instead we find: 'The woods are lovely, dark, and deep'. To say that the woods are 1) lovely, 2) dark, and 3) deep differs considerably from claiming that they are lovely in that they are dark and deep. In Frost's line, the general adjective 'lovely' is explained by the more particular modifiers 'dark' and 'deep'. In the editor's line, the egalitarian threesome appears to be parallel, but of course it is not-it is as if we proclaimed that a farmer grew apples, Mclntoshes, and Northern Spies.

Before The Poetry of Robert Frost was published, our texts were the books that Frost saw into print during his lifetime. Most of the poems were reprinted without change while the poet lived, and at his death, in 1963, were available in Complete Poems of Robert Frost and the 1962 collection In the Clearing. In 1969, Holt, Rienhart and Winston published Lathem's edition, which is now available in paperback. (Two volumes of selections use Lathem's text; there remains a Pocket Books selection edited by Louis Untermeyer with Frost's punctuation.) In England, Jonathan Cape published Lathem's Frost in 1971, and Ian Hamilton edited a Selected Poems for Penguin in 1973, which uses the corrupt text.

It would seem axiomatic that an editor's task is to represent the author's intent insofar as the editor can establish it. When the author is long dead, when manuscripts or printed sources are absent, when a variety of evidence lacks single authority, an editor must rely on historical scholarship to inspire guesses of authenticity and to mediate a readable, probable text. The matter of Robert Frost, however, differs from the matter of Shakespeare or of Keats. Frost lived long and spoke his poems aloud on thousands of occasions, reading from one of his printed texts. If he had wished to sprinkle his lines with new commas, as one might salt a roast, he would have pencilled them into his reading copy. If he had wished to add a question mark, or to delete a hyphen, it would have been simple to do so. In the absence of alterations, his repeatedly printed texts suggest intention.

His inconsistencies in punctuation and his deviations from standard practice bother a tidy mind. So a tidy mind will find relief in Lathem's revision, which insists on consistency to the exclusion of other criteria. But poets are notoriously innovative in punctuation, inventing combinations such as'-:' and ';-?'. E. E. Cummings would make an extreme example, but we need not go so far. William Butler Yeats arranged colons and dashes and semi-colons among his lines not according to the conventions of prose but as notation for pause and pitch; T. S. Eliot's punctuation and capitalization were eccentric and expressive; among Frost's American contemporaries, Wallace Stevens most nearly abided by conventions of punctuation, but he was partial to an unorthodox colon; Marianne Moore was both scrupulous and inventive. Even Edwin Arlington Robinson, the one contemporary whom Frost praised, was given to coinages such as ',-'.

We await a history of punctuation in modern poetry. In the meantime, eccentric punctuation has perhaps been purposeful. Possibly some punctuation fails its purpose, but deliberate intent remains clear-to serve as notation of sound, a pale cousin of musical notation. Frost cared for the sound of verse. He went so far as to claim that words existed in order to make noises: 'Words are only valuable in writing as they serve to indicate particular sentence sounds.' Frost seemed not to have cared much for assonance, lush vowels rubbing against each other. He cared most for the cadence of talk, with the nudge and thrust of intelligence in pace and pitch. In his work he continually referred to a semantics of noise. 'Remember,' he told us, 'that the sentence sound often says more than the words . . . . 'There are tones of voice that mean more than words.' Another phrase he liked was 'the sound of sense', the way cadence makes sense and sense makes cadence.'. . . if one is to be a poet he must learn to get cadences by skilfully breaking the sounds of sense with all their irregularity of accent across the regular beat of the metre.'

Anyone who has read Frost's letters and essays is familiar with the theme of 'sentence sounds' and 'the sound of sense', Frost's contribution to modern poetics. Doubtless, both critical idea and poetic practice derive from Wordsworth; they are an extension of the desire to write with the material of the spoken idiom. And in his best poems Frost exemplifies again and again the miraculous wedding of speech and metrical line. If ever one feels puzzled by a phrase such as 'breaking the sounds of sense . . . across the regular beat', one can refresh one's understanding by listening to a stanza such as this one, from 'A Patch of Old Snow':

There's a patch of old snow in a corner
  That I should have guessed
Was a blow-away paper the rain
  Had brought to rest.

Frost listens to speech and repeats it like a mimic. With a line break the poet expects some vocal indication (almost always a slight pause; often a change of pitch) to match the line structure visible on the printed page. Set as prose, Frost's lines make a sentence without tension: 'There's a patch of old snow in a corner that I should have guessed was a blow-away paper the rain had brought to rest.' Breaking the sounds of sense across the regular beat, Frost gives us 'the rain/Had brought', making a pause both unnatural and wonderful, raising pitch on 'rain' and hesitating between subject and verb.

Lathem was largely merciful with this poem; but he put a comma at the end of the first line, after 'corner', where the line-ending pause is all the pause Frost wanted. (The added comma is of course grammatically incorrect.) Put a comma after 'corner' and you pile pause on pause, slowing the poem down, making Frost's cadence sluggish.

I cite this alteration not because it destroys the poem but because it is typical. Unlike the comma in 'lovely, dark, and deep', it is not confusing; it does not alter Frost's thought. Lathem adds punctuation that Frost omitted, and by so doing he slows things down. In the second line of 'Stopping by Woods', Lathem changes Frost's 'His house is in the village though' to 'His house is in the village, though'. Besides adding pause, Lathem adds pitch variety: we drop our voices when the comma isolates 'though'. It is no calamity, perhaps, to drop the voice-but we have Frost's word for it that 'The living part of a poem is the intonation entangled somehow in the syntax idiom and meaning of a sentence.' By intonation we may know character. Maybe the speaker who runs over the pause-'His house is in the village though'-tosses away his notion more lightly than the speaker who pauses and lowers pitch: 'His house is in the village, though'. This latter fellow seems more calculating.

Doubtless many readers will find these objections quibbles, and think they resemble the discomfort of the princess troubled by a single pea under twenty mattresses. Many readers find a poem's identity in its paraphrase or summary, 'what the poet is trying to say', instead of in its articulate body, its bulk and its shapely dance, its speech and color and tone and resolution, its wholeness, of which meaning is part. But a poem resembles a Henry Moore reclining figure at least as much as it resembles philosophical disquisition. Frost defined poetry as what gets lost in translation; the meaning, in the vulgar sense, is what gets translated. When you change the movement of a poem by adding pause, you alter the reclining figure's dimensions. Lathem adds 443 commas to Robert Frost's poems, thirteen colons, 156 long dashes (some replacing short dashes), and twenty-two semicolons. He adds three parentheses; he adds ellipses three times and deletes them twice. In most categories he deletes some instances of the punctuation mark; but when he deletes one he often adds another that indicates greater pause-a semicolon or a long dash replaces a comma.

Lathem also adds twelve question marks, altering not time but pitch and meaning. On many occasions, especially in narrative poems," Frost wrote sentences in the grammatical form of questions, but concluded with periods instead of question marks. In 'Home Burial', the wife gazes out a window from which she can see her son's grave; Frost has the husband say: 'What is it you see/From up there always-for I want to know.' The idiom, grammar and pacing embody the husband's state of mind. He feels his wife estranged and he wants the sensitivity he lacks. As Frost punctuated it, the husband begins the speech as a question, which he interrupts with a demand:'-for I want to know.' Lathem alters and diminishes Frost's characterization of the husband by supplying the question mark. The question mark raises the pitch of 'From up there always', sentence-sound is altered, and the husband's voice becomes gentler and less demanding.

Later in the poem, the wife remembers that her husband, after digging their child's grave, remarked casually how little time it took a birch fence to rot. She cannot forgive 'talk like that at such a time!' 'What had how long it takes a birch to rot/To do with what was in the darkened parlour.' Her sentence is not a question inviting an answer but a statement of outrage. Lathem's question mark raises pitch and diffuses anger: 'What had how long it takes a birch to rot/To do with what was in the darkened parlour?' Correctness makes politeness. For one more example, 'The Witch of Coös' cackles at her son, 'We'll never let them, will we, son! We'll never!' For Frost the exclamation is the thing. Lathem has the witch inquire: 'We'll never let them, will we, son?'

Elsewhere, Lathem adds eighty-one hyphens and deletes 101. These alterations are as destructive as added commas and question marks, but the changes are more various and more difficult to describe. When Lathem emends 'lower chamber' in 'Storm Fear' to 'lower-chamber', he alters the rhythm of the line: 'low-' becomes louder than 'chamb-'. In 'The Trial by Existence', Lathem turns 'cliff-top' into 'cliff top', slowing the line down by adding the percussion of equal volume. When Lathem changes 'sun-burned' into 'sunburned' ('Pan With Us'), and 'tip-toe' into 'tiptoe' ('The Death of the Hired Man'), he follows modern usage but he does not represent the sentence-sounds of A Boy's Will and North of Boston, Frost's first two books of poems. In 'Hired Man', 'barn-door' becomes 'barn door', 'harp-like' becomes 'harplike', and, on the other hand, 'college boy's' becomes 'college-boy's'. For Frost's 'up hill' we have Lathem's 'uphill' ('The Bonfire') and Frost's 'down-hill' becomes Lathem's 'downhill' ('The Gum-Gatherer'). In 'A Lone Striker', 'many, many eyed' travels a long distance to become 'many-many-eyed', and Frost's pace is considerably quickened. In 'Birches', Lathem removes five hyphens from Frost's fifty-nine lines. 'Ice-storm' and 'ice-storms' become 'ice storm' and 'ice storms', and 'snow-crust' becomes 'snow crust'. One line loses three hyphens: 'With all her matter-of-fact about the ice-storm' becomes 'With all her matter of fact about the ice storm', with considerable change to the sentence-sound; 'matter-of-fact' speeded the line up. In 'The Last Mowing', Frost wrote the touching lines: 'There's a place called Far-away Meadow/We never shall mow in again . . .' which are destroyed in this edition. Surely that hesitation, that mini-mini-pause of the hyphen in 'Far-away', makes delicate mimicry of emotion. Lathem's 'Faraway Meadow' sounds like a real-estate agent's name for a subdivision.

For one more example let me cite a small change in a poem that some critics consider Frost's best work. 'The Most of It' does not escape revision. 'And then in the far distant water splashed' finds itself changed into 'And then in the far-distant water splashed'. Doubtless this addition of a hyphen is a small matter; its alteration of sound is slight. But if the hyphen does not change the sound of the line, why bother to add it? If it does change the sound of the line, how dare we alter Frost's sentence-sound?

Edward Connery Lathem was graduated from Dartmouth College in 1951, took a D. Phil at Oxford in 1961, and was librarian of Dartmouth College from 1968 to 1978. A friend of Robert Frost's for many years, he is author or co-author of several publications connected with Frost, beginning with Robert Frost: Farm-Poultryman, edited with Frost's biographer, Lawrance Thompson, in 1963. They also edited Robert Frost and the Lawrence, Massachusetts 'High School Bulletin', in 1966. Also in 1966, Lathem and Hyde Cox edited the useful Selected Prose of Robert Frost; Frost's frequent omission of the expected comma goes uncorrected in his prose. In the same year, Lathem compiled Interviews with Robert Frost and in 1969 edited the complete poems.

It should be acknowledged that Lathem's notes, at the back of the text, record his emendations. And, of course, some of them are justified. Lathem discovered that broken type in an old edition had deleted a comma, a deletion scrupulously followed in subsequent editions; he restores this comma, clearly intended by the author. Lathem replaces English spelling with American; evidently Frost had expressed annoyance at the English spellings surviving in some of his poems. Also, Lathem standardizes practice in the use of double and single quotation marks. Although I cannot say that the old inconsistency bothered me, I find this alteration innocuous, because it is silent. These alterations are the editor's proper business.

In 1971, two years after his edition was greeted with some criticism, Lathem defended himself in an introduction to the Imprint Society's reprint of The Poetry of Robert Frost. He began with the statement that Frost requested help with a new collection which the poet did not live to make. 'Following the appearance of In the Clearing in 1962,' Lathem said, 'Mr Frost spoke with the current editor about such an undertaking and requested his assistance with parts of the overall task.' When Frost died, in 1963, 'the responsibility [for the undertaking] devolved upon his publishers and estate'. (The executor of the Frost estate is Alfred Edwards, Frost's publisher at Holt, Rinehart and Winston; he came to Holt from the National City Bank of New York. It is understood that Lathem will succeed him as executor.) Lathem, chosen to be editor, arrived at the decision that 'the general reader would oftentimes be helpfully served by some degree of editorial attention to the poems. . . . Thus, the desired objective could be attained of editorially enhancing textual clarity . . .'

It would be difficult to object to the motives acknowledged in these passive sentences; but in looking back from the perspective of the completed text, 'enhancing textual clarity' seems understatement and 'some degree of editorial attention' gross euphemism. Lathem nowhere alleges that Frost authorized him to change the punctuation of his poetry. He quotes from a letter in which Frost spoke of his pride in omitting punctuation. As Lathem puts it: '[Frost] replied disarmingly [to the correspondent who complained about the 1930 Collected Poems]: "I indulge a sort of indifference to punctuation. I don't mean I despise it. I value it. But I seem rather willing to let other people look after it for me." ' At first glance, these sentences seem almost the permission Lathem requires. But the whole letter is not so much disarming as enraged. Leonidas W. Payne, Jr., chairman of the English Department at the University of Texas, had sent Frost a list of 'errors' in the 1930 volume. After speaking of his indifference, Frost boasted about being able to write a telegram without using the word 'stop', and continued: 'I'll have those commas and hyphens tended to, though, if only for your peace of mind.' Enjoying his sarcasm, he went on: 'You must remember I am not writing school-girl English'; defending an inversion, he said, '. . . the order should remind you of a very ancient figure of speech. Your friends of the Classical Department will tell you about it.' If I had received this letter, I would not have characterized it as 'disarming'.

Although Frost appeared to give way to Payne on punctuation -'Fortunately it turned out you were wrong in all your findings of errors except the punctuational'-it is noteworthy that he lived thirty-three years after writing this letter without [having] those commas and hyphens tended to . . .'.

Acknowledging that Frost hated to be corrected, Lathem tells of three occasions on which Frost accepted correction during his lifetime; two of the suggestions were offered by Lathem himself. These corrections add commas to remove ambiguity. Because of a friend's puzzlement, Frost altered 'To err is human, not to animal' to 'To err is human, not to, animal'. Lathem's suggestions were similar; but if Lathem, after years of association with Robert Frost, has only two corrected commas to tell us about, how can we accept his addition of 441 others? If we cannot accept these revisions, we may at least understand why Lathem hesitated to propose them to the living poet. As Lathem puts it,'. . . Frost did not actually relish having individuals challenge him regarding the punctuation of his verse . . .'

Of course the great majority of Lathem's changes do not remove ambiguity. Lathem's unauthorized repunctuations exist for the purpose of consistency-and they impose a consistency on Frost's poems that Frost gave no evidence of desiring. 'Mr Frost was,' says Lathem, 'in many ways, a very inconsistent person-as Emerson proclaims great souls are wont to be. (It was, withal, one of the myriad fascinating aspects of his personality.)' Fascinating as Lathem found Frost, surely the two men differed in character as thoroughly as they differed in prose style.

For Lathem is passionate on the subject of the consistency that his friend lacked. Consistency is his pride, as Frost's was self-reliance. Although Lathem acknowledges that 'Diverse spellings and irregularity of practice in punctuation' are not 'apt to render text unintelligible . . .', he insists that 'they can distract, puzzle, and indeed annoy readers, undesirably intruding upon an assimilation of what the author has wished to communicate'.

Here I must object. The inconsistency of omitting question marks in 'Home Burial' was itself communicative; the inconsistency of grouping 'dark and deep' together rather than separating them by a comma was itself communicative. The poet who argued that sentence-sounds carry the thrust of meaning would not agree that altering these sounds facilitates communication: these sentence-sounds are themselves the communication.

As evidence supporting his revision of Frost's punctuation, Lathem cites Frost's performance of the poems preserved on records and tapes. He uses this device to support only 'lovely, dark, and deep': '. . . literally scores of voice recordings exist of Frost saying his poems. Over and over again he is heard to give the three adjectives approximately equal stress, with no vocal suggestion that punctuation [might be otherwise] .' Well, I also have listened to Frost and I do not agree; perhaps we hear what we expect to hear. In reading his poems, Frost was as inconsistent as he was in other matters, and he often hurried through them as if he wanted to get the reading done with. He kept to the meter, usually kept to the line break-but he would sometimes rush through the pauses his punctuation indicated; if we were to punctuate according to Frost's performance, we would find ourselves omitting many commas in his true text-and omitting full stops, as well. His performance recalls another Frost comment on punctuation, quoted by Lathem: 'I hate to depend on punctuation at all. I hate to end with a word in one sentence that might well belong to the next sentence.'

But editors must not punctuate by performance. In 'Ash-Wednesday', Eliot's text omits expected commas in 'my legs my heart my liver . . .'. But in performance he paused where we would expect commas in ordinary punctuation; perhaps the ear needs help the eye does not require. Should Eliot's future editor add commas to 'Ash-Wednesday'? Wallace Stevens, equally skilled in meter and in free verse, had the most consistently beautiful ear among modern poets; but if his editor relied on his ghastly poetry readings, we could find full stops between words in a single phrase, or the lines broken after every word, to represent the rhythm-destroying pauses of his speaking voice.

I raise an old question, never to be answered: What is a text? If a text is the product of consistent copy-editing designated as official by the poet's estate, then Lathem's Frost is Frost's text. But I would rather a poet's text represented a poet's probable intentions, even if those intentions were inconsistent. My editor's task is picayune, difficult, and humble; it serves the poet and the poetry. I would have my editor study manuscripts to discover printer's errors and unauthorized emendations, and study the poet's own texts and correspondence for clues to intended revision. Perhaps I would have him perform mechanical (preferably mandated) changes such as making English spelling American, or regularizing double and single quotes. Perhaps I would have him repair the spellings of proper names, and footnote the original misspellings. To discover an author's intention is often difficult, since the evidence is often contradictory; we are grateful to editors who serve reader and poet in the cause of authenticity. Other modern poets-W. B. Yeats and Marianne Moore-have been edited with a scrupulous regard for their latest known intention.

It seems strange that there has not been more complaint about Lathem's Frost. However, some voices have been raised. Richard Poirier is perhaps Frost's best critic; in Robert Frost: The Work of Knowing, he complains of Lathem. William Pritchard, who teaches at Amherst and was acquainted with Frost, has spoken out on several occasions, including an Atlantic review of the Lathem book in October, 1970. Gerald Burns, in the Southwest Review, complained of 'Lathem's hyperconservative commas, largely unnecessary in a poet who prided himself on writing lines you couldn't misread . . .' The most outraged commentary was published in Partisan Review by Frank Bidart, a poet whose control of speech tempo depends on ingenious, expressive, and eccentric punctuation.

But Lathem's corrupted text of Robert Frost is increasingly taken as the true text. When a critic or anthologist writes Holt for permission now, permission is granted to reprint from Lathem's edition. Not everyone does as he is told. Pritchard quoted the old text in Lives of the Modern Poets, as did Robert Pinsky in The Situation of Poetry. The first volume of Thompson's biography, published in 1966, uses Frost's punctuation; the second volume (dedicated to Lathem) and the third use Lathem's punctuation. Anthologists by and large do poorly. William Harmon's The Oxford Book of American Light Verse preserves the corrupt Lathem text. Richard Ellmann's The New Oxford Book of American Verse prints thirty-three poems, almost thirty pages' worth, and uses Lathem's version every time. These books seem especially regrettable, because Oxford books have the half-life of strontium 90. Ellmann's The Norton Anthology of Modern Poetry (edited in collaboration with Robert O'Clair) is possibly worse: here the text follows Lathem most of the time but occasionally follows both, as when 'Stopping by Woods' has the 'though' comma but not the 'dark and deep' comma; occasionally it provides punctuation that neither Frost nor Lathem anticipated. In time, if we do nothing, sounds that are not Frost's sounds will be the only sounds one can hear- and Frost's own punctuation, when it turns up in an eccentric edition, will look like a misprint.

A responsible literary scholar should be commissioned to edit a variorum edition of Frost's poems. A variorum Frost should reestablish Frost's intended punctuation while a subtext records variations-including broken letters, English spellings, single or double quotes, and even Lathem's corruptions, for all I care. But the poet's sentence-sounds must return to the poet's page.