The Language of Poetry

Haj RossDepartamento de Lingüística,

Universidade Federal de Minas Gerais&

Circle - Noetic Services

This article will come at poetry in a way that may strike some readers as unusual. It is traditional to say (and it is obviously somewhat true, in a sense) that poems are written in a particular language. But the shift of perspective that I want to suggest that the reader try on, for the duration of this paper, is to view poetry itself as a kind of language. To imagine that what poets do to everyday language is to convert it into a universal language, one to plunge the reader into contrasts, and to give the experience of the unity to be found by going beyondapparent contrast. Poetry is a language to express total messages, ones which are not readily susceptible to segmentation and linearization, a language used to walk the edge of the sayable and point beyond, to what lies outside of the limits of languages. Let us agree to refer to this universal language as POETESE.

This “language” of poetry is, I believe, also used in the construction of music, or of film, or dance, of their fusion in video art, and probably even in forms of art not presented in time, like painting or sculpture or conceptual art. I will not attempt this ambitious semiotic linking here, however. Instead, I would like to summarize, for thegeneral reader who has no previous experience with the studyof poetry, some of the most salient ways of communicating that are made use of by poets – the “speakers” of this language of poetry.

The first feature of this language that I will call attention seems too obvious to mention: the existence of stanzas. But stanzas are not made just by leaving a space between two lines. Some of the other properties that a

group of lines must have to make up a stanza are what will constitute our first focus. My research on poetry thus farhas suggested to me the existence of a small number of typesof formal “building blocks,” with which poets “section” their works. I will list four of these sectioning devices below.

A. Splittings: a poet may demarcate the sections of a poem (often, its stanzas, but not always) by putting all of the occurrences of some type of linguistic element in one section, and none in another section. Thus, in the following William Carlos Williams poem, a beautiful and stark one, taken from the collection titled “Spring and All (1921),”

I. By the road to the contagious hospitalunder the surge of the bluemottled clouds driven from thenortheast a cold wind. Beyond, thewaste of broad, muddy fields

5brown with dried weeds, standing and fallen

patches of standing waterthe scattering of tall trees

All along the road the reddishpurplish, forked, upstanding, twiggy

10stuff of bushes and small treeswith dead brown under themleafless vines–

II. Lifeless in appearance, sluggishdazed spring approaches– 15

They enter the new world naked,cold, uncertain of allsave that they enter. All about themthe cold, familiar wind–

Now the grass, tomorrow 20the stiff curl of wildcarrot leafOne by one objects are defined–It quickens: clarity, outline of leaf

But now the stark dignity ofentrance. Still, the profound change

25has come upon them: rooted, theygrip down and begin to awaken.

I suggest segmenting these lines into two great sections: the first fourteen lines (ending with the line which, importantly, begins with the word Lifeless), and the last thirteen, which introduce the topic (sluggish / dazed spring), and, not coincidentally, follow it with the first tensed verb: approaches.

Williams has painted the inertness, the dead motionlessness, of the retreating winter with a linguistic correlate – verblessness. That is, there are no verbs in the poem’s first section. It is natural to associate verbswith action, and thus with birth and life. After the middle couplet of the poem,

Lifeless in appearance, sluggishdazed spring approaches_

we find a number of changes, not only the presence of eight finite verbs (15approaches, 16enter, 18enter, 22are [auxiliary] (defined), 23quickens, 26has [auxiliary] (come), 27grip (down) and27begin). Note also the slow increments of verbal complexity in the two auxiliary verbs (are, and has) and theverbal particle down ), but also the poem’s first and only subordinate clause (18that they enter) and finally, the poem’s first and only infinitival complement: 27to awaken.

I think, though, that the splitter – that is, the device for sectioning the poem – that is most apparent is probably Williams’ use of “sudden” pronouns – the two they’sthat are the subjects of the two occurrences of the verb enter (and of the only verb to indicate downward motion: 27grip down), and the two objects of prepositions: 18about them and upon them. These three pronouns are without antecedents;only later do we surmise that they are perhaps plants. Williams introduces them as having only enough form to be able to refer – they have identity, but no content, they seem skeletal. The most sudden of all, of course, is the itwhich is the subject of 23quickens – can we ever really know for sure what this tiniest it refers to?...

And finally, after the middle couplet, instead of the irregular pattern of number of lines per stanza (6 in the first, 2 in the second, 5 in the third) which the poem openswith, the poem settles down into a steady rhythm of 4 lines per stanza, possibly as a suggestion of the pulsing of life in the nascent spring, the throbbing of the heart of Mother Earth.

Any one of these “split” distributions would have divided the poem into two sections; the fact that so many of them cooccur to mark the same boundary is a measure of its poetic importance in Williams’ painting of the rising, unfolding, deepening ascension of this most very special of the year’s faces - Spring.

•

For another example of a splitting, this time, one effected in part by the distribution of some phonetic elements, let us inspect Gerard Manley Hopkins’ famous sonnet, “The Windhover.”

The Windhover: To Christ our Lord

I caught this morning morning’s minion, king-dom of daylight’s dauphin, dapple-dawn-drawn Falcon,

in his ridingOf the underneath him rolling steady air, and striding

High there, how he rung upon the rein of a wimpling wing In his ecstasy! then off, off, forth on swing,

As a skate’s heel sweeps smooth on a bow-bend:the hurl and gliding

Rebuffed the big wind. My heart in hidingStirred for a bird, the achieve of, the mastery of the thing!

Brute beauty and valor and act, oh, air, pride, plume, hereBuckle! AND the fire that breaks from thee then, a

billionTimes told lovelier, more dangerous, O my chevalier!

No wonder of it: shéer plód makes plough down sillion Shine, and blue-bleak embers, ah my dear,

Fall, gall themselves, and gash gold vermilion.

Gerard Manley Hopkins

The traditional division of a sonnet into eight lines and six lines, the octet and the sestet, is obviously relevant here - the boundary is not only supported by the typography, but also by the AbbAAbbA CdC dCd rhyme scheme.

What I would like to point out here are some rather simple phonetic correlations with this traditional boundary:in this poem, Hopkins has carefully used the consonant whichends all the rhyme words in the octet, namely [ŋ] (I have indicated these by boldfacing the <ng>’s), only in the octet: there are no occurrences in the sestet. And another kind of phonetic event – syllables beginning with an[s] followed by another consonant (it must be a phonetic consonant – thus I do not include <sh> groups here) – are likewise only found in the octet (these I have indicated by underlinings).

Hopkins also splits the sestet off from the octet in other phonetic ways: only in the last six lines do we find the sound [ʃ] (orthographically, <sh> [and <ch>, in chevalier ] – I indicate the four occurrences of [ʃ] here withitalics). Furthermore, Hopkins salientizes word-initial consonant clusters that start with [p] or [b], because the sestet begins with two words that start with [b] and are followed by a resonant: [r] for brute, and [y] for beauty. And later in the same line, we encounter another pair of adjacent alliterants, this time in [p]: pride, plume. I have indicated all nine of the poem’s word-initial labial stop clusters by underlining them and putting them in boldface. All are in the sestet.

And one last way we might note that Hopkins splits sestet from octet – the distribution of the poem’s words in [v]. Except for the preposition of, all of whose six occurrences except the last, in line 12, are in the octet, all of the poem’s six lexical words which contain [v] are inthe sestet, except the first of these – the striking verb-

used-as-noun achieve, which helps to bring the octet to a crashing close. And when we examine the sestet’s five [v]-words a bit more closely – valor, lovelier, chevalier, themselves, vermilion – we see that Hopkins has created a small “gang” of consonants –[v, l, r] – and that [v] only occurs in the company of at least one of its gang members. This idiosyncrasy helps to highlight with sound the brilliant visual image of the poem’s last word - the gold-red Christfire of vermilion.

There are many more ways in the language of poetry in which Hopkins highlights the traditional octet-sestet boundary. And if we ask why, in experiential terms, Hopkins should have wished to mark this boundary so clearly,we can, I think, find good reasons for doing so. For the octet is seen in Hopkins’ mind’s eye: he tells us of “catching,” on the morning of the day on which he writes thepoem, the glory of the windhover – a small English hawk, also called a kestrel. The octet is the dominion of the windhover – all the images directly concern the world of airand sun (daylight, dawn, air, gliding, wind) - except for the “rhyming” motion from the world of ice-skating: Hopkins lets us feel the acceleration of the hawk’s swoop kinesthetically, by literally bringing it down to earth, so that we may translate it into our bodily feelings as we round a curve on skates. The bird is mentioned only in theoctet, as are also the four occurrences of the past tense (caught, rung, rebuffed, stirred) and the five verbs or nouns with progressive meanings: riding, rolling, striding, gliding, andhiding. Hopkins marks the end of this vividly verbal scene with the poem’s only striking zero-nominalization – a verbnoun : : the achieve.

The images of the sestet – the fiery daring of the windhover, the iridescent shine of the freshly plowed earth,the sudden flash of light from the gray-dusted coals in a fire – are all concerned with unexpected revelations – the revelation of the Christ in all of us, in each thing of the phenomenal world, the possibility for this divine light to break forth from any thing, at any time. The octet is the example, the poetic merging of bird and watcher (my heart in

hiding / Stirred for a bird) – which lets Hopkins feel his own divinity, and each reader theirs. The sestet is the domainof recollection, reflection, of realizing what a gift we have been given. At first our thoughts are confused, unordered – we are so close to the epiphany that we have nosyntax - look at the list of nouns in the first line of the sestet. Syntax slowly reasserts itself: the fire, which immediately follows the charged and deliberately ambiguous verb Buckle!, though it is modified by the poem’s only relative clause, is not the subject of any expressed finite verb. Are we to hear the second and third lines of the sestet as a copular sentence, as if there were an is before the two adjectives – lovelier, dangerous ? – has the power of the windhover deleted this small verb, as it has deleted the occurrences of the conjunction and between the last three nouns which precede Buckle! Only in the last trio of lines does Hopkins come back, with the help of some soothing, unexceptional clauses, to (a part of) himself - the Philosopher.

I think, then, that the transition from the reliving ofan epiphany to “getting a grip” on this experience is a gigantic experiential leap – and that it is this leap that Hopkins marks so clearly in his dazzling poem.

•

B, Balancings: A poet may also demarcate sections of a poem by putting equal numbers of linguistic elements of a given type in each section. For instance, a ballad of fifty verses might be balanced by exhibiting exactly three nouns in each stanza. Such balancings are demonstrated brilliantly in Roman Jakobson’s analysis of William Blake’s “Infant Sorrow” (Jakobson 1970).

INFANT SORROW

My mother groand! my father wept.Into the dangerous world I leapt:

Helpless, naked, piping loud:Like a fiend hid in a cloud.

Struggling in my father’s hands:Striving against my swadling bands:Bound and weary I thought best:To sulk upon my mother’s breast.

William Blake

Jakobson shows that one stanza balances the other, in that each mentions both parents once - in the second stanza, the parents occur in the reverse order of that of the first, with the poem opening and closing with references to the mother (Cf. Jakobson (1970)). This reversal of order mirrors a thematic aspect of the poem: the infant’s motion,as it enters the world, is outward, away from the mother, inthe first stanza, as opposed to a return, back toward the mother, in the second stanza. Jakobson also points out a less obvious fact: that this thematic mirroring is formallyechoed by each stanza having exactly five lexical nouns. Jakobson goes on to show that there is also a simultaneous balancing of the two inner couplets (lines 3 - 6) as opposedto the two peripheral couplets (lines 1 - 2 and 7 - 8), in that each of these two pairs of couplets also has five lexical nouns. Jakobson is a pastmaster of this art – I will leave to the interested reader the joy of reading through Jakobson’s own masterful exposition, and pass on to a brief discussion of another beautiful poem, this one by Yeats.

When you are old and grey and full of sleep,And nodding by the fire, take down this book,And slowly read, and dream of the soft lookYour eyes had once, and of their shadows deep:

How many loved your moments of glad grace,

And loved your beauty with love false or true,But one man loved the pilgrim soul in you,And loved the sorrows of your changing face;

And bending down beside the glowing bars, Murmur, a little sadly, how Love fled,And paced upon the mountains overheadAnd hid his face amid a crowd of stars.

William Butler Yeats

Before we look at the structures which the poet uses totransport us, let us talk very briefly about a kind of experience that the poem evokes.

I think that one way to feel this poem is as a meditation on two kinds of love. The middle stanza shows us two clauses, all of whose four verbs are loved. The subject of the first clause is some indefinite number of persons, who loved the person to whom the poem is addressed in a way we might call “easy.” These “lovers” loved her grace and beauty, but the very reversal, in the last three words of the first half, of the standardly fixed expression,true or false, suggests strongly that the poet had no great regard for those whose love is based only on such superficialities.

For there was another, a different man, possibly a loner? - who loved his mate’s soul, who loved her for being a pilgrim, who even loved her in her times of sorrow, loved the deep shadows of her eyes...

And what was the fate of these two kinds of love, the superficial, the transitory, on the one hand, and the eternal, on the other?

Here I think that we can only say that it is too early to tell. The mood of the poem is sad – there is a suggestion that unless the poet’s beloved wakes up, and learns to choose that lone eternal lover, she will end up in

the days when she is old and grey (am I right in hearing in full of sleep a suggestion of the nearness of death?), aloneby the fire, reading slowly, murmuring, wondering about the way the true love that she once knew in her life has gone. Her motion is downwards: she takes down a book from above (where the poet, who wrote it, is), her head moves down, as she nods, and she bends down to look at the fire. It is not a fire blazing as with passion, but only glowing, a fireof coals. Her murmur is only a little sad – it is far fromthe anguished cry of the poet, whose path is upwards. His love for her is a fierce one; a loss of his love would force him to flee, first to the heights of the mountains, and then higher yet, up to the eternal realms, where the infinity of the timeless stars would be the ones to console him in his loss. I think that the poem allows for this saddest of fates to be averted, but only if the poem’s addressee awakes, and hears the truth of the poet’s love forher. And stops fooling around with the inconsequential how many.

This basic dialectic of high vs. low is paralleled in the realm of meaning by the contrast in intensity in the acts of the poet’s love, as opposed to those of the poet. Thus the poem’s you is said to be full of sleep, to nod, to read slowly, to dream of her lost beauty, to murmur – and onthe other hand, the poem’s one man flees, paces, and hides his face, to conceal the depths of his grief.

I think Yeats wants us to take the mirroring in the gross syntactic structure of the poem’s one long sentence asan index of the importance of this set of contrasts. Thus we note that the poem begins with a subordinate clause that starts with a wh-word (when), and that the subject of this clause, you, has three adjectives predicated of it: old and grey and full of sleep. These are separated by the poem’s first two and’s. And the poem ends with another subordinate clause, introduced by the poem’s last wh-word, how, whose personified subject, Love also heads a coordinate verb phrase. The three volitional verbs that Love is the subject of flank the poem’s last two occurrences of and: fled, / and paced upon the mountains overhead / and hid his face amid a

crowd of stars. The fact that you is associated with adjectives, and Love with agentive verbs, in this mirroring structure, is another semantic rhyme of the high/low, intense/muted contrasts that we have already noted.

With this suggestion of my general understanding of thespace of feeling that the poem may have grown out of, and may take some readers like me back to, let me now look at some of the structural properties of its twelve lines.

I would like to show that as in the analysis of “InfantSorrow,” where Jakobson argues that there are grounds for postulating two alternative, simultaneous sectionings - one of first versus second verse, and one of outer versus inner couplets - there are also alternative sectionings in Yeats’ poem.

The first is the obvious tripartitioning into three verses, which I will refer to as A, B, and C.

In addition to the typographical device of leaving a line blank between the verses, and the repetition of the ABBA rhyme scheme, both of which being characterizable as kinds of balancings, Yeats equates the three verses with a series of different balancings, of a number of different kinds of linguistic elements. The positioning of the poem’s three major punctuations is balanced: the one long sentence of the poem is interrupted every four lines, the first two times by two semicolons, and then by the final period. Also, each of the verses has just one clause that starts with a wh-word – when, in the first line of A, then how, inthe corresponding position in B, and again how, in the second line of C.

We may note that not only are the poem’s rhyme words selected in such a way that their sounds conform to an ABBA rhyme scheme, which thus provides a phonetic balancing in support of the tripartitioning, but that their parts of speech are also chosen in a way that lends support to the tripartitioning. Thus in each verse, we find an instance

of what Jakobson calls grammatical rhyme – a pair of nouns that rhyme: in A, book and look; in B, grace and face; and in C, bars and stars. This strong grammatical rhyme is bolstered by a weaker one: in each verse, there is one further line which ends in a noun, or noun-like, element. Thus in A, we find the noun sleep; in B, we find the pronounyou; and in C, we find the adverb overhead, a fused prepositional phrase which still obviously ends with the noun head. Both of these uses of grammatical rhyme are of course additional examples of balancings.

Furthermore, when we look at what kinds of parts of speech are represented within each line, we find that each of the three verses has exactly four prepositional phrases -[PP’s] - (that is, phrases consisting of a preposition followed by a noun phrase), which come in a steady rhythm ofone per line until line ten, which has none. This “missing” prepositional phrase is then picked up by the last line, special in many ways, which has two PP’s, one inside another. The twelve PP’s are shown in (1) below. (Note that down, although it has prepositional uses, is here, in lines 2 and 9, used as a directional particle, having no nominal object. I have thus excluded it for this tabulation)

(1) A: of sleep B: of glad graceby the fire with love false or

trueof the soft look in youof their shadows deep of your changing

face

C: beside the glowing barsupon the mountains overhead[amid a crowd [of stars]PP]PP

And another grammatical element that Yeats uses to establish a poetic equation of the three verses is the definite article. There are just two the’s per verse: in

A, in lines 2 and 3; in B, in lines 7 and 8; and in C, in lines 9 and 11.

There is one lexicosemantic pattern of striking regularity which I will mention here. There are only two words that are repeated in the poem: love(d), which I will return to shortly, and face. Both of these repetitions have important structural roles to play. The latter repetition fits into a pattern of placing a body part on thefourth line of each of the three verses: eyes, face, and face.

I will leave the subject of the ways in which Yeats supports the poem’s tripartite structure after an examination of just three phonological patterns: the distributions of words which begin in initial consonant clusters, the distribution of poem’s six stressed long [ow]’s; and the poem’s six occurrences of metrically strong words which contain [f] (full, fire, [soft is metrically weak]; false, face; fled, face).

There are twelve words in initial consonant clusters, and each verse has just four. These cluster-initial words,which are all placed in metrically strong positions, start out in a pattern of pairs: two per odd line. In line 1, grey and sleep; in line 3, slowly and dream; and in line 5, in the alliterating noun phrase glad grace. The two-per-line rhythm continues, but moves to an even line; the last line of the first half, line 6, which has beauty ([byúwDiy]) and true. The four initial clusters of C start out with glowing in its first line, fled in its second, and return to the two-per-line pattern for the poem’s last phrase: crowd of stars.

The second segment that dances to the structural rhythmof the tripartition is the metrically strong long vowel [ow]. There are just six such vowels, and each occurs in an odd line: 1old, 3slowly, 5moments, 7soul, 9glowing, and finally, in the poem’s only trisyllable [another ripple in an established superior rhythm], 11overhead

And finally, we find the same number of [f]’s which introduce accented syllables – 6. There are two in each verse: in A, full and fire; in B, false and face; and in C, fledand face. Interestingly, the last five of these six [f]’s

are intercalated between the odd-line [ow]’s - only the first [f]. in full, is on an odd line.

In sum,I think we can see fairly clearly that there arestrong structural concomitants to go with the three-part structure indicated by this poem’s typography. What I would like to do now is to argue that there are just as strong structural reasons for seeing the poem as having, at the same time, two equal halves, of six lines each.

A strong indication is provided by the poem’s only but,which opens line 7 and the second half. Immediately after this but, we find the poem’s only other agent, aside from the you to whom the poem is addressed – one man. Furthermore, there are just two occurrences of you, and under the hypothesis of a two-part structure, they are located in a way that makes sense: there is one in the first line of each half – the first stressed syllable of thefirst line of the first half, and the last stressed syllableof the second half’s first line.

We have already noted the importance of the contrast ofthe two kinds of love – the superficial and the eternal – inour discussion of the experiential ground of the poem. I think that the halving of the whole poem really grows out ofthe strength of the conceptual halving of the central verse.Since love is so fundamental a concept for the poem, it is important to note that the halving structure I am proposing here makes sense out of the distribution of the word love; ineach half, there are two occurrences of the verb loved and one occurrence of the noun love. If we had only the tripartite structure to refer to, the distribution of this fundamental word would remain mysterious.

As would the distribution of the poem’s 20 non-pronominal nouns, which makes no sense if viewed in terms ofthe poem’s three verses: there are 6 in A [sleep, fire, book, look, eyes, shadows], 8 in B [moments, grace, beauty, love, man, soul, sorrows, face], and 6 in C [bars, Love, mountains, face, crowd, stars]. But there are ten in each half of the poem.

There are many other points of interest that could be discussed in connection with the halving of this poem – for instance, the fact that the poem’s four [p]’s and six [b]’s are balanced around its midline – and other pieces of phonetic evidence. But there is not space enough here to do more than sketching in the briefest of ways the most important points of this poem’s structure, which, as we haveseen, exhibits simultaneous three- and two-part sectioning,using the formal device of balancing in various ways for both of these kinds of divisions.

I think that the evidence that emerges from the study of even these two poems - one by Blake and one by Yeats - issolid enough for us to be able to conclude that balancing isanother of the formal devices that is available to poets whowant to section their poems. And it may also seem reasonable that the same poem can be sectioned in more than one way. But we might want to ask a more specific questionwith regard to the Yeats poem: why is it that the poet chose to divide the poem in exactly the two ways that we have seen - that is, into two and three parts, instead of, say, into three and seven parts, or into four, five and eight parts, etc.? The answer I would like to suggest is avery tentative one, as I do not even know how to go about backing it up, so I hope that it will be heard as a possibility, rather than as a claim. I wonder if the simultaneous presence in the poem of threeness and twoness is not a kind of semantic, or thematic, “rhyme” with the nature of the two kinds of love with which we started our discussion. That is, the addressee of the poem is at the same time involved in what we might call a triangular, unstable, relationship, involving herself, the poet, and an indefinite set of the how many others, and also in the dyadic union of two true hearts – in the eternal, transcendent love whose true symbolis Two.

•

C, Lonerings: The third type of formal device that I would like to suggest as a building block that poets may take advantage of is one that I will refer to as lonering. Lonering is most closely related to splitting; the difference is that while in a splitting, we find all of the elements of a given type in one section, and none in the other, in lonering, the division is not all-or-none, but what we might call “all-or-one.” Thus the first section ofa poem may be full of elements of one kind, and the second section all but empty – one element alone, as a kind of echo, may appear in the second section. I think that one of the clearest examples that I have found of a poem which makes use of this principle of structuration is one of Robert Frost’s most famous poems.

The Road Not Taken

Two roads diverged in a yellow wood,And sorry I could not travel bothAnd be one traveler, long I stoodAnd looked down one as far as I couldTo where it bent in the undergrowth;

Then took the other, as just as fair,And having perhaps the better claim,Because it was grassy and wanted wear;Though as for that, the passing thereHad worn them really about the same,

And both that morning equally layIn leaves no step had trodden black.Oh, I kept the first for another day!Yet knowing how way leads on to way,I doubted that I should ever come back.

I shall be telling this with a sighSomewhere ages and ages hence:Two roads diverged in a wood, and I—

I took the one less traveled by,And that has made all the difference.

•

I think that it is hard not to hear this poem as allegorical, as an invoking of a metaphorical theme: Life is a road. The decisions that we make, the forks in the road, may be subtle: we may never know what the effect of “taking” one turning will be. And yet choose we must. The poem is more than about the indecision which is our constant companion on our path – the poet puts us into his indecision, we live his quandary as our own.

To start with, it is fall – the woods are yellow. There is no time to lose on this trip – there never is. Each decision is crucial.

And here we are before one. We stand for a long time,hesitating – almost taking one of the roads, after looking as far as we can down it – then abruptly, we take the other,which we are told is “as just as fair” – so our choice did not have much riding on it, not much in terms of beauty, at any rate.

But then we are told that the one we have taken was in fact preferable: because it was grassy and wanted wear. Sothere was a good basis for our choice after all. But no sooner are we freed from our indecision than we are plunged again back into the middle of it:

Though as for that the passing there Had worn them really about the same.

And as if to tamp down the absence of a basis for a rationalchoice, we are told once more that the roads “lay there equally.” I think it fair to say that Frost, the great story-teller, has succeeded masterfully in these first twelve lines - the poem’s first sentence - in awakening in us the experience of being of two minds about something.

And we are not to be let off the hook yet - the second sentence, line 13, suggests that we will have another chance, that we will be able to go back, to resolve our doubts, to find out whether we have made “the right” choice.But the third sentence dashes that hope – there will never be another today. We choose, with the slimmest of bases fora choice, we hope that we are making the right decision. Paradoxically, we know that it does make a difference. It may not be easy to find the road less traveled by, we may never really know for sure that we have found it . . . but it will be the most important find of our lives. Frost, inhis typical poker-faced way, describes the game to the reader as if it were just once in our lives that we had a right choice to make. But we know that this less traveled-by road is only found through a lifetime of tough-to-call choices. That what in the end does make a difference is the seeking, the hanging in there, the searching for our road.

Given this general experiential ballpark, let me now turn to pointing out the way in which Frost uses the device of lonering in service of awakening in us the feel of the insecurities of our search. In this poem, the first element that I would like to call attention to is the grammatical one of transitive verbs, which for our present purposes I will define as verb followed by nominal direct objects. Thus I will not be considering doubted as a transitive verb, though I would see telling as one, even though its object, this, is not a lexical one.

Most crucially, I would suggest that while trodden is clearly notionally transitive, in the context of its use in line 12, where it is the main verb of a relative clause, since its underlying object, leaves, does not follow it, I will not include it among the poem’s (surface) transitives.

Granting, for the purposes of discussion, the validity of the above analytic decisions, we discover that the poem’s22 main verbs are arranged stanza by stanza in an alternating pattern. In the odd stanzas, we find that all the verbs are intransitive, with one lonered exception in A

and one in C. In the even stanzas, we find the opposite pattern: all transitives, with one lonered intransitive in each of B and D. The distribution is shown in the table below.

Stanza Transitives Intransitives

A divergedcould not travel

bestoodlookedbent

B tookhaving

waswantedhad worn

C layhad trodden

keptknowingleadsdoubtedshould (ever) come (back)

D shall be tellingdiverged

tookhas made

I would like to suggest that it is not accidental that this poem has the syntactic property of transitivity highlighted in this pattern of alternating complementary lonerings, rather than, say, the property of questions, or prepositional phrases of direction, or any other of a numberof features that we might mention.

But before I can explain this feeling, I would like to present some evidence for the hypothesis that this poem is another in which there are simultaneous alternative sectionings. The quadripartite sectioning of the four verses is supported not only by the typography and the ABAABrhyme scheme, and by the alternating lonerings that we have just noted above, but also by a pattern of lexical repetitions: each verse has one and only one pair of lexical items repeated. In A, we find travel - traveler; in B,we find wear - worn; in C, we find two occurrences of the noun way; and in D, we encounter the repetition of ages. Iwill assume, then, that it is not controversial to postulatethe existence of a quadripartitioning in this poem.The alternative sectioning is what we might call the inversion of the quadripartite one. In addition to having four sections of five lines each, as can be seen in the following diagram, Frost has also given this poem five sections of four lines each (I will refer to them as “quatrains.”).

The Road Not Taken

Q1 Two roads diverged in a yellow wood,And sorry I could not travel bothAnd be one traveler, long I stoodAnd looked down one as far as I could

Q2 To where it bent in the undergrowth;

Then took the other, as just as fair,

And having perhaps the better claim,Because it was grassy and wanted wear;

Q3 Though as for that, the passing thereHad worn them really about the same,

And both that morning equally layIn leaves no step had trodden black.

Q4 Oh, I kept the first for another day!Yet knowing how way leads on to way,I doubted that I should ever come back.

I shall be telling this with a sigh

Q5 Somewhere ages and ages hence:Two roads diverged in a wood, and I_ I took the one less traveled by,And that has made all the difference.

Probably a good place to start in providing evidence for the existence of these five quatrains is to note that they provide a basis for an explanation for the positioning of the poem’s five trisyllabic words. It is rare, in a poem this long, to find so few longer words. Thus we should pay all the more attention to them. Note first thatin terms of the four verses, the five trisyllabic words are not distributed in any symmetrical way – there are two trisyllables in the first and third verses, and one in the fourth. But if we look at their distribution with respect to the five quatrains, we see first of all that there is a regular rhythm of one per quatrain, and secondly, we note that in the first four quatrains, the trisyllables are placed in a regular way with respect to the lines of each quatrain. That is to say, in the first and third quatrains, the trisyllable is placed in the quatrain’s third

line, and in the even quatrains, the trisyllables are in thequatrains’ first lines. The final quatrain breaks this rhythm of alternating line-placement. It is in the last line that we find the fifth trisyllable: difference.

The last word says different and it also is different. Thisis the essence of the poetic urge – for language to reach beyond the desire and ability to say, to attain to the ability to do, and ultimately, supremely, simply to be.

So now look what we have inside the rhyme scheme of each verse - a phonetic pattern, which repeats itself four times –

wood fair A lay sigh travelerboth claim B black hence

undergrowthstood wear A day I equallycould there A way by

anotherundergrowth same B back difference difference

And what is striking is that we find just exactly this formal pattern - ABAAB - repeated in the distribution of thetrisyllables: just as the second and the fifth lines rhyme in the ABAAB rhyme scheme, so we find that it is the second and fifth of the trisyllabic words which are line-final – which are then themselves rhyme words. Diagrammatically,

Rhyme scheme Distribution of trisyllabicity

A travelerB undergrowth/A equallyA anotherB difference/

So the structure of the whole is reflected in the structure of each verse: as above, so below.

, probably for dramatic effect. Many poems establish a rhythm in their earlier portions, a rhythm which the poet will later perturb: it is exactly at these broken symmetries that our attentiveness, our openness, is maximized. Broken rhythms are the homes of poetic epiphanies. As in this poem – there is a great sweep of the soul in the weight of the final trisyllable - difference. Frost has made it a perturbation among the poem’s rhyme words, too - it is the only less than perfect rhyme, the only feminine rhyme word in the poem.

Such cases of multiperspectivality arise in other areas of poetics than in the relatively formal one of sectioning, andraise important questions about the correctness of the current insistence in much of linguistic theory, as in fact in most of current science in general, on one single “true” account of any domain of fact. The close study of poetry gives us a clear message about the viability of any such restricted view: tain’t so. One of the goals of my research is the tracing of the implications of pluralistic models for the structure(s) of a poetic text for a new kind of linguistic theory, in which “competing” analyses are not in fact seen as such. The parallels to the principle of complementarity that we have learned from the physics of this century are striking. (Cf. Capra (1977))The first salient feature of the language of poetry being sectioning, the second is a feature one might call its diametric opposite: what I have called “hologramming.” (Cf. Ross (1982a,b)) Briefly, poems exhibit holographic structure to the extent that the structure of the whole poemis recapitulated in the structure of some subpart. An example: Robert Frost’s famous “Stopping By Woods On A Snowy Evening” [Cf. Appendix G] exhibits an (almost) totally regular AABA rhyme scheme: the third lines are lonered by this scheme. In Ross (1982a), I suggest that ofthe four stanzas, the third is also lonered conceptually andpoetically. This stanza is the one in which the “stopping”that the poem concerns – namely, in my view, the momentary experience, with a calm inevitability, of the truth of one’sown tiny transience against the vast backdrop of cosmic time– is most strongly felt by the reader. Note that this stanza is the only one which does not mention either the first person or the woods, or give any adverbial reference to space or time. It is also the only stanza all of whose rhyme words are of the same part of speech. The other three stanzas all have at least two different types of word - stanza 3 has just one. And the one kind it does have is the stoppedest of all parts of speech, the most opposite from verbs and life: stanza 3’s lines all end in nouns, and

two of these are even nominalized, or desiccated, verbs - mistake and sweep. Just as the A-rhyme stops in each third line (except in line 15, an exception that is consonant withthe work of the fourth verse, which is to get us going again[note how the two final verbs, which are the goingest of allparts of speech, are the first return to a verbal rhyme-wordsince line 1]), so the third stanza stops us cold. Each stanza holograms the whole poem.

The poems that I have studied manifest a number of types of holographic structures. I am constructing a typology of hologrammings, with an eye to coming to an understanding of what kinds of messages seem to require thiskind of structuration. Until quite recently, most structures postulated in linguistic theory have been quite atomistic, though the network-like representations of some current syntactic theories (cf. Johnson and Postal (1980)) and the kinds of phonological patterning that have begun to be studied as an outgrowth of John Goldsmith’s pioneering work on autosegmental phonology (cf. Goldsmith (1976)) herald a change in emphasis. However, to my knowledge, semantics continues to be studied almost entirely within atomistic perspectives. It is generally felt that words (0r morphemes) have meanings, and that the meanings of sentences and of larger units are to be seen as combinatorial functions of these atoms. (Cf. Reddy (1979),for a brilliant analysis of the conduit metaphor, which predominates as a perspective for current thought on the nature of communication.)

The single most salient fact about poems is that they are fused into wholes by such processes as hologramming, despite the atomistic effect of such processes as sectioning. The investigation of the dialectic between holism and atomism, not only within poetics but within linguistics in its entirety, is a question that must needs concern anyone who has studied even a single poem.

The third feature of poetic language that I will touch on here involves what I will refer to, for want of a better term, as semantically irrelevant patterning. To convey this notionmore clearly, it may be helpful to first inspect a poem

which manifests a pattern which is highly similar to those that I wish to call attention to, but whose patterning is relevant semantically. Jakobson (1981) points out that theclear thematic ternariness of Paul Klee’s “Zwei Berge Gibt Es” (cf. Appendix H) is supported by a wealth of triadic structurings: there are three main clauses in the poem; three occurrences of the central noun Berg; three nach’s; three und’s; three negatives – and on and on and on. We have little difficulty in concluding, with Jakobson that this triadicity is indeed semantically relevant.

But there are poems in which equally clear numerical patternings exist, but where I, at least, can find no semantic/thematic relevance. For example, in Wallace Stevens’ “Domination Of Black,” (cf. Appendix I), there is an astonishing wealth of five-fold structurations. For instance, the only pair of different rhyme-words, hemlocks and peacocks, occurs five times; the first person pronoun I is used five times as a subject, each time with a verb of perception or remembering, each time in the past tense; thefive subordinate clauses are all introduced by manner-adverb-related conjunctions (as or how), and so forth. While I will defer a full presentation to the poem’s fivingsuntil Chapter 4 below, I think that it is already possible to see that it is too pervasive to be dismissed as mere accident. Thus we see here another contribution to the “numerical music” of a poem, but a music that here finds no conceptual or thematic echo that I have been able to discern. This suggests to me that there are parts of poetry that are akin to the prelinguistic babbling of infants, or to the semantically irrelevant nonsense syllables, like tra la la, or oodley pop a chow pop a chow pop a chow chow or With a riteful la, tiddy fie day, Riteful, riteful, tiddy fie day, which fill the musical spaces between the verses of a song.

I consider the thesis of semantically irrelevant patternings to be sufficiently bold as to warrant the most detailed scrutiny. Some questions that immediately arise are: what is the extent of such numerical music? What kinds of linguistic elements can be used as “notes,” to continue the analogy? Most significantly for linguistic

theory, is such semantically irrelevant patterning restricted to poetic texts? If not, as I suspect – if we can document the existence of such patternings in ordinary language as well, the consequences for our conception of therelationship between sound and meaning that makes up the linguistic code will be profound indeed.

To sum up, the language of poetry depends upon an intricate set of combinatorial signals. of the abstract patterns which poets use to section poems with, to hologram with, and to make “numerical music” with. As I have said above, this is an apparent paradox in a communication systemwhich seems to simultaneously segment structures and to unify them. The paradox is intensified when we consider the implications of the existence of numerical music. For here we encounter a dialectic of the most shattering kind: some formal features of a poem are to be interpreted by a reader, while others would appear to be merely there to be listened to. But which are which? When does the reader have “the” right understanding of the poem?

The answer is, I think, clear: never. Just as the language of poetry is made to allow artists to break free ofthe bondage that would be imposed by any fixed system of categories, so the great poem is great precisely because of its boundless capacity to resist analysis. To work with a poem is, for me, a humbling experience: it takes all of theanalytic energy that I can muster to come to the frontiers of analysis itself. At that frontier, I stand face to facewith the enigma which underlies, and motivates, all art.

J

The most beautiful experience we can have is the mysterious.It is the fundamental emotionwhich stands at the cradleof true art and true science.

Albert EinsteinThe World As I See It