I

I often find surprising the weight that some stylists are willing to lay on technical evidence that has been reduced to quantitative or statistical summary. Some issues certainly respond well to statistical investigation—decisions about how to attribute literary works, for example, about authenticating newly discovered correspondence, or about dating historical fragments. But the general reader should beware whenever a stylist combines an explication of this particular specialization with the chauvinistic claim that that branch of stylistics (itself only a subspecialty, it seems to me, within technical analysis) represents all that is worthy of attention in this marvelously fertile discipline.

Typical of such a claim would be the assertion that “[when] authors opt for the same transformations, their styles are said to be similar.”¹ In this chapter, I hope to show, to the contrary, that if we seek a concept of style that will be easily recognized by the general reader, our own researches will have to range considerably more widely. A mere count of rule applications may satisfactorily “fingerprint an author,”² but to claim that it reveals all that needs to be considered in a stylistic analysis is surely to indulge in the critical equivalent of palmistry.³ In the pages that follow, I shall illustrate how notions such as “similar” and “different” can depend on aspects of syntactic form widely divorced even from less quantitatively rigid technical analysis, and I shall introduce a second class of statements about style which will begin to give us the flexibility that the narrowly technical approach so obviously denies.

In John Dryden’s elegiac ode “To the Pious Memory of . . . Mrs. Anne Killigrew” we find the lines:⁴

The Sacred Poets first shall hear the Sound, . . .

And streight, with in-born Vigour, on the Wing,

Like mounting Larkes, to the New Morning sing.

(WJD III, 115: 188, 191-192)

These striking lines occur as this poem reaches its extraordinary emotional climax. In them, Dryden seeks not only to convey to his readers his vision of Anne Killigrew’s apotheosis as a “Harbinger of Heav’n” (line 194), but also to express the intensity of his personal commitment to that vision, a double focus referred to by the editors of WJD as Dryden’s combination of “irrepressible lyricism and fervent conviction” (WJD III, 317). A review of the syntactic structure of this passage reveals one of several ways in which Dryden achieves the second of his two ambitious goals, injecting an air of excitement and fervency into the narration of these altogether mythical events.

Underlying lines 191-192 of this poem a transformational syntactician would in all probability posit something like this deep structure:

(The sacred poets shall) straight sing to the new morning with inborn vigor like mounting larks on the wing.

The four PPs in this clause would all, that is, be generated underlyingly at the righthand end of their dominating VP. Their eventual position in Dryden’s surface structure would be attributed to a subsequent wholesale application of PP-Fronting. But such remarks about the technical background of this particular passage are of little interest when compared with other comments that one may make about the surface configuration itself. For what surely strikes the reader most forcibly about the linguistic form of this couplet is that, by distorting the underlying word order, Dryden delays for nineteen syllables (almost the full two lines of verse available) his revelation of what exactly the “Sacred Poets . . . shall” do on the Judgment Day.

Let us for a moment trace the steps of the reader as he encounters this passage and attempts to analyze it for the first time.⁵ He must first read and file away, so to speak, the subject and auxiliary verb of this clause (Poets and shall respectively). After completing work on the first of its conjoined predicates, which is, as it happens, comparatively straightforward, he reaches the second (. . . and streight . . .). At this point, however, he must continue to hold in abeyance both subject and auxiliary while he interprets the four increasingly evocative preposed PPs, for only after he has negotiated these obstacles will Dryden finally grant him the all-important verb, sing. The culminative force that this small word carries is due in part, no doubt, to its metrical and phonological role in completing the wing/sing rhyme, and in part to the fact that it suddenly clarifies semantically the relevance of the prominent simile of the “mounting Larkes.” Just as important, though, is the fact that sing resolves the suspended syntactic structure of this second VP, the effectiveness of this resolution being heightened by the sheer length of the preceding delay.⁶

Two problems arise if one tries to describe the temporary frustration experienced in reading this couplet simply as a technical syntactic phenomenon. First, transformational grammarians have arbitrarily ruled such matters outside the domain of theoretical linguistics. Were the interruption to the forward progress of interpretation at this point itself a center-embedded construction and thus, as we saw in the course of our discussion of Shelley’s style, liable to halt parsing altogether, the syntactician might show some interest. If, alternatively, that same material were to encourage premature (and false) parsing by providing constituents capable of being accidentally misanalyzed as the predicate to “Sacred Poets . . . shall,” a psycholinguist at least might have relevant observations to make about it as a kind of syntactic “garden path” sentence.⁷ In general, however, matters involving linguistic memory and its limitations have consistently been viewed as facets of linguistic performance and virtually excluded from serious theoretical study.⁸

In itself, this first obstacle to integrating the concept of syntactic delay into technical stylistic analysis is largely procedural. Liberalizing the definition of syntactic competence to include such factors could be easily accomplished and might well prove beneficial. But there remains a second, altogether more significant, reason for avoiding this innovation. Consider in this connection a second textual example. Lines 77-98 of George Gordon, Lord Byron’s “Childish Recollections” seem at first to flow from the pen of a contrite poet ashamed of his early “childish” vituperations:⁹

Away with themes like this! not mine the task

From flattering friends to tear the hateful mask; . . .

Or, if my muse a pedant’s portrait drew,

POMPOSUS’ virtues are but known to few:

(BPW, 34: 77-78, 89–90)

Leavening this rather stodgy and self-righteous fare the reader is delighted to find flashes of temper that show Byron’s more characteristic reluctance to capitulate without putting up a good fight. Lines 91-92 in particular betray just such a strong sense of assertive indignation:

I never fear’d the young usurper’s nod,

And he who wields must sometimes feel the rod.

(BPW, 34: 91–92)

In the bland context just described, the rod, as archetypal symbol of just, even divine, chastisement, stands out boldly. As in the passage from Dryden’s elegy discussed previously, the concomitant stylistic prominence of the word “rod” is supported in part by Byron’s exploitation of the rhyme-slot in the second line of the couplet. Again as in the earlier case, however, this culminative constituent also completes a syntactic construction that had been left hanging, object-less, in mid-line (“and he who wields . . .”).

In both these passages, then, poets achieve stylistic emphasis by creating a syntactic distance between parts of a single construction, temporarily suspending or delaying the reader’s ability to parse the language of the text before him. Yet—and this is the most important observation of all—Byron’s technical method in his couplet differs radically from Dryden’s in the lines from the Killigrew ode. Where Dryden, as we saw, used PP-Fronting, a movement rule, to interpose syntactic material between his subject and its verb, Byron employs Right Node Raising simply to delete what would otherwise have been the first occurrence of rod, postponing until the second any chance for a complete interpretation of either conjunct (see Figure 10). Any technical analysis of these two passages, then, will inevitably and quite properly concentrate on isolating, in linguistic terms, this important difference between their styles. The contrapuntal similarity that we detect between them despite their technical disparity would seem to be of an altogether different order, directly related to syntactic structure by all means, but encompassing also nonsyntactic factors such as distance (or delay) and suspension. It will be the general aim of this chapter to explore further the definition and application of such concepts in literary criticism—concepts which I categorize under the heading “aesthetic.”¹⁰

FIG. 10

II

Richard Cureton undertakes in his dissertation “to present, illustrate, and document a typology of syntactic aesthetic effects,”¹¹ a step which, he claims, will permit stylists to break out of their blinkered commitment to either highly formalistic, interpretively contentless linguistics or technically sloppy, albeit insightful, literary criticism. As my remarks throughout the preceding pages should have made clear, I strongly sympathize with Cureton’s general goal. When his “typology” itself arrives, furthermore, I find it disappointing only in that Cureton repeatedly uses the single term “aesthetic” to cover altogether too heterogeneous a class of stylistic effects, effects which range from “semantic tension” to “syntactic parallelism.” Such a lack of internal discrimination detracts, I feel, from what nevertheless remains an extremely stimulating first pass over this infuriatingly elusive material.

My own employment of the term “aesthetic” will therefore be more narrowly restricted. In this study I shall use it to refer only to those features located in the surface form of any work of art that are experienced by the reader as functions of time or space. That aesthetic features exist only at the surface level divorces them, clearly, from much of the technical material discussed in chapter 2. That they exist as “functions of time or space” places them equally firmly outside the legitimate domain of interpretation—a point to which I shall return in due course. But a more distinct impression of what I do in fact intend by this definition may be gained from considering a third observation: that aesthetic features themselves exist independently of any one artistic medium, being applicable with equal success to music, the visual arts, or poetry.

Let us consider a couple of examples. At this particular level of abstraction a classical fugue may be said to instantiate musically the same aesthetic effect as the iterated visual pattern in a rug or on a ceramic tile. A similar correspondence relates the temporal syncopation of jazz rhythms in the musical sphere to the spatial and metrical off-lineation of Gwendolyn Brooks’ “We Real Cool”:¹²

We real cool. We

Left school. We

Lurk late. We

Strike straight. We . . .

This kind of broad aesthetic equivalence between different artistic media has of course been noted before, even by stylists, who are often poorly versed in aesthetic theory. Insufficient care has been devoted, however, to noting its scope, especially with respect to the verbal arts; as a result, a clear picture of its implications for literature has failed to emerge. Donald Davie, for example, cites approvingly a very suggestive remark from Susanne Langer’s Philosophy in a New Key:

[The] tension which music achieves through dissonance, and the reorientation in each new resolution to harmony, find their equivalents in the suspensions and periodic decisions of propositional sense in poetry.¹³

On the one hand, I too am happy to accept the broad correspondence that Langer is trying to establish here. But on the other, I cannot help regretting that her actual formulation of the equivalence effectively equates two very different kinds of phenomenon. “Suspension” assuredly belongs in any glossary of aesthetic terms, applicable alike to music and to verbal expression. It differs crucially from the cognate term “suspense”¹⁴ precisely in its perceptual vividness and its relative lack of emotional connotations. We can effectively perceive something being suspended either in time (as with the arrival of a thunderclap) or in space (as with the arc of a half-built highway overpass). The term with which Langer links suspension in her discussion, however, does not describe an aesthetic phenomenon at all under my definition. “Tension” constitutes just one of several interpretive reactions that we might (or might not) in a given instance correlate with a suspended aesthetic structure. To compare the two is something of an apples-and-oranges exercise.

An aesthetic effect, then, is created by and in the surface features of a work of art. Its existence is not uniquely related to a particular artistic medium, nor should it be confused with the various interpretations that we may subsequently place upon it. Related by definition to a work’s form rather than to its interpretation, aesthetic effects are indeed as devoid of a priori meaning as phrase-structure rules or the Tensed S Condition.

Naturally, such a narrow definition couched in such strong terms will entail all kinds of testable consequences for a theory of stylistic analysis that makes use of “aesthetic features” in this sense. Let us briefly review one of its more obvious predictions. Dryden’s technical means of achieving the aesthetic effect of distance or delay in the passages studied early in this chapter were seen to differ radically from Byron’s means of producing precisely the same effect in “Childish Recollections.” My assertion that aesthetic effects are also meaning-independent now suggests in addition that the interpretations readers place on the delay that each poet creates might likewise differ. I do not wish here to preempt the business of chapter 4, where I shall discuss in detail the interpretations of stylistic choices. Still, it does seem to me that whereas Dryden seeks an effect of “emotional climax” in his ode, Byron rather contents himself in his couplet with a simple sense of “rhetorical finality.” To the extent that I am correct in this assessment, my assertion that the level of aesthetic effects constitutes an entirely autonomous arena for syntactic manipulation receives preliminary support. For it is only at that one rather abstract level of analysis, we now see, that these two passages unmistakeably share any stylistic feature: their common reliance upon a strategy of delay.

Since aesthetic features are equally relevant to all kinds of art criticism, stylistics itself may benefit from a distinct term uniquely applicable to aesthetic judgments about verbal style. Wherever a particular aesthetic effect is realized in poetry using syntactic material, therefore, I shall apply the term perceptual to the statements that stylists use to describe that phenomenon. My remarks throughout the opening section of this chapter about the respective styles of Dryden and of Byron fall, of course, within this domain. In the remaining pages of the chapter I shall offer a number of further examples which will, I hope, clarify my theoretical assertions in the preceding paragraphs and at the same time give some hint of the variety and subtlety of readers’ judgments in this area as a whole.

As the discussion thus far may already have revealed, I am myself an enthusiastic amateur as an aesthetician. It seems to me however that aesthetic effects as I have defined them fall into at least three classes:

(a) Features of proportion define our perception of what is relatively large or small, long or short within a given work. Syntactic distance or delay would thus represent a typical perceptual stylistic effect taking its lead from a feature of aesthetic proportion.

(b) Aesthetic scale refers to the overall or absolute dimensions within which a work of art or its component parts are constructed. As a mural demands an appreciation different in some respects from that appropriate to a miniature, so in poetry a couplet and an epic will be differently perceived. The need to characterize these distinct perceptions will result in stylists formulating specifically perceptual judgments about scale.

(c) Pattern, finally, is the aesthetic effect whose presence in literary texts has been most widely discussed and accepted by stylists in the past. (Indeed much of Cureton’s dissatisfaction with stylistics as a discipline is due to what he sees as its almost obsessive concern with this aspect of stylistic analysis—to the virtual exclusion, he claims, of sensitive interpretation.) This area may, then, be a good one from which to begin our exploration of perceptual stylistic analysis.

III

The stylistic literature is indeed long on books and articles that examine various poets’ exploitation of abstract syntactic patterning in their works.¹⁵ There remains nevertheless a great deal to be said about the many subtle variations that may be played on this simple theme and about the connections between perceptual patterning on the one hand and matters such as genre and period styles on the other. Let us begin by outlining the basic concepts involved.

Patterns of various sorts, I take it, form part of mankind’s general aesthetic code. They are perhaps most easily recognized when instantiated as sequences of meaningless symbols such as letters of the alphabet. In this study, I shall rely on only three such pattern types: chiastic patterns (a b . . . b a), involving simple inversion of just two elements;concentric patterns [a b c (d) . . . (d) c b a], mirror-image sequences of more than two elements; and parallel patterns [a b c . . . a b c (. . . a b c)], repetitions of sequences of elements in uninverted order. Stylists who have done any work at all at what I am calling the perceptual level of analysis will immediately recognize these aesthetic patterns as the basis for arrangements of syntactic material that are common in literary texts, as in these rather straightforward cases:

Chiasmus

Watch all their Ways, and all their Actions guide:

(“The Rape of the Lock” PAP, 225: 88)

Concentricity

As you meet it, the Land approacheth you.

(“Astraea Redux”: WJD I, 29: 253)

Parallelism

Squeezed in ‘Fop’s Alley,’ jostled by the beaux,

Teased with his hat, and trembling for his toes;

(“Hints from Horace”: BPW, 133: 311-312)

Naturally, such simple cases do not exhaust by any means the many configurations actually employed by poets, but they will provide a sufficient basis for making several useful preliminary observations.

It is no coincidence, for example, that I selected as my three paradigm citations lines from the works of two major Augustan poets, Pope and Dryden, and a couplet from a poem of Byron’s which he referred to in his correspondence as “that Popean poem.”¹⁶ Syntactic patterning has long been recognized as a major stylistic trademark of the Augustan period of English poetry, and taught as such in English literature classes. But many pitfalls await the teacher who advances this generalization too incautiously, for equating patterned syntax too glibly with Augustan style may lead one to overlook a more truly characteristic feature of Augustan poetic syntax, as we shall see. So simple a rule of thumb may thus equip the student with an analytical tool that is in practice prone to serious malfunction.¹⁷

For a start, it takes no great effort to uncover examples of perceptual patterning in works that belong both chronologically and philosophically to the consciously anti-Augustan reaction of the early nineteenth century. As the first of two such examples, let us consider the style of one of Shelley’s early works, Queen Mab. Seldom read except by Shelley scholars, Queen Mab suffers from a number of serious faults, not the least of which are its grandiose ambition and cosmic perspectives. Carlos Baker sums up the opinion of many critics when he describes this poem as “a distempered and unoriginal vision, mediocre as verse, and something less than mediocre as history.”¹⁸ It is also Baker, however, who seems almost to surprise himself when he notes of the “synthetic poem that results”:

Shelley does a better job with it than one would have supposed possible.

. . . By the use of . . . the device of surveying past, present, and future from a suprahistorical plane, he achieves a certain unity.¹⁹

Shelley himself does nothing to conceal his dependence on the rather transparent expository ploy that Baker here applauds:

. . . Spirit, come!

This is thine high reward:—the past shall rise;

Thou shalt behold the present; I will teach

The secrets of the future.

(CWS I, 77: 64-67)

But other poets have structured narrative works along similar lines without letting that choice affect their stylistic behavior (Dryden’s “The Hind and the Panther” is such a poem, as Earl Miner argues²⁰). I suspect, therefore, that Baker has happened upon a factor of more than trivial importance when he draws attention to triple-ness as a source of unity in Queen Mab. Triadic structure, I suggest, offers Shelley in this poem an organizing principle that supplies, not only narratively but at various levels in the text, the coherence that his philosophical and historical material would otherwise totally lack.

Triads (parallel perceptual patterns, in the terminology of this chapter, with exactly three repetitions of the syntactic sequences involved) certainly occur at every syntactic level within this poem. Almost every verse paragraph, for example, yields at least one phrase composed of three conjoined members of the same lexical category, many of these combinations being bolstered by alliteration to counter their natural tendency to slip into weak tautology:

lovely, wild and grand (CWS I, 69: 70)

astonishes, enraptures, elevates (CWS I, 69: 71)

Rots, perishes, and passes (CWS I, 71: 156)

passions, prejudices, interests (CWS I, 78: 103)

Strangers, and ships, and merchandize (CWS I, 81: 201)

morals, law and custom (CWS I, 94: 130)

quake, believe, and cringe (CWS I, 96: 219)

judgment, hope, or love (CWS I, 97: 256)

frozen, unimpassioned, spiritless (CWS I, 98: 25)

weak, unstable and precarious (CWS I, 118: 206)

Nor can Shelley’s reliance on lexical triads of this kind be dismissed out of hand as a compulsive structural “twitch,” since there is, at least sometimes, method to this apparently scatter-fire technique. Both T. S. Eliot and Melvin T. Solve have noted the general importance of “catchwords” in Queen Mab,²¹ recurrent words and phrases associated with both the virtues Shelley sought to recommend to his readers and the vices he so vigorously condemned. Neither critic comments, though, on the distinct likelihood that these catchwords will turn up in those persistent (and again sometimes alliterating) triadic constructions:

Virtues

youth, integrity, and loveliness (CWS I, 98: 14)

peaceful, and serene, and self-enshrined (CWS I, 120: 256)

Love, freedom, health (CWS I, 121: 15)

virtue, love, and pleasure (CWS I, 129: 75)

Vices

kings, and priests, and statesmen (CWS I, 92: 80)

Kings, priests, and statesmen (CWS I, 93: 104)

priest, conqueror, or prince (CWS I, 97: 237)

Ruin, and death, and woe (CWS I, 92: 85)

ruin, vice, and slavery (CWS I, 93: 99)

The ruin, the disgrace, the woe of war (CWS I, 99: 68)

As these and many other similar examples show, Shelley’s employment of triadic series often correlates with and accentuates major oppositions he wished particularly to pinpoint for discussion in Queen Mab. At the very least, the discovery of this correlation between syntactic patterning and thematic emphases invites us to study a good deal more closely a stylistic feature that at first seemed trivially mechanical.

Shelley’s preference for syntactic triads is no less evident if we shift our focus from the level of lexical categories to that of phrasal and clausal constructions; triple parallel patterns continue to leap from every page:

So bright, so fair, so wild a shape (CWS I, 69: 74)

all virtue, all delight, all love (CWS I, 98: 19)

Blighting all prospect but of selfish gain,

Withering all passion but of slavish fear,

Extinguishing all free and generous love

Of enterprize and daring, . . .

(CWS I, 100: 84-87)

This commerce of sincerest virtue needs

No mediative signs of selfishness,

No jealous intercourse of wretched gain,

No balancings of prudence, cold and long;

(CWS I, 104: 231-234)

The larger the syntactic units of which the triad is composed, of course, the greater the opportunity to embellish the whole by varying details of the internal structure of individual members. At the same time, the underlying perceptual reality remains the same in each case, triadic parallelism running through the syntactic fabric of Queen Mab as insistently as the beat of a Viennese waltz.

Perhaps the most beautiful and certainly the most complex example of syntactic patterning in this poem occurs, appropriately, at one of the poem’s several thematic turning points. For 213 lines at the beginning of Section V, Shelley has ruthlessly attacked the coercive power of “commerce,” an aspect of “civilization” that he implausibly blames for having caused virtually every evil known to man. At line 214, however, this ranting abruptly gives way to a calmer tone as the poet prepares his readers to consider his own (highly idealistic) alternative goal for inspiring human actions, “[the] consciousness of good” (line 223). It is this nine-line transitional passage to which we may now turn our attention, noting first that it displays a particularly heavy density of “local” syntactic triads such as those exemplified in the preceding paragraphs.

A simple lexical triad in line 222, for instance, appears positively normal, given the bias in favor of such constructions throughout this poem:

There is a nobler glory which . . .

Imbues his lineaments with dauntlessness,

Even when, from power’s avenging hand, he takes

Its sweetest, last and noblest title—death;

(CWS I, 104: 214, 220–222; emphases mine)

Equally unsurprising, though slightly richer for its pleasing thesis-antithesis-synthesis sequence, is the placement of three finite verbs, one each in lines 214-216 of this passage:

There is a nobler glory which survives

Until our being fades, and, solacing

All human care, accompanies its change;

(CWS I, 104: 214-216; emphases again mine)

Shelley neatly captures in these verbs, first, the lifelong strength of our “nobler glory” wherever absolute good rather than pecuniary gain is our goal; second, his awareness of man’s mortality; but also, third, his conviction that the passing of life will not necessarily entail the total loss of that glory, which may in some undefined way “accompany” our “change.”

We may conclude our survey of relatively local triadic patterning in this pivotal passage by observing a slightly more complex arrangement of phrasal triplets in lines 217-219. I transcribe the lines themselves on the next page, and indicate in Figure 11 the particular parallel pattern that I have in mind.

FIG. 11

There is a nobler glory which . . .

Deserts not virtue in the dungeon’s gloom,

And, in the precincts of the palace, guides

Its footsteps through that labyrinth of crime;

(CWS I, 104: 214, 217-219; emphases my own)

(We may note in passing that this syntactic parallelism is again more than merely decorative. The word palace here finds itself in distinctly unsavory company, dungeon and crime being the corresponding modifiers in the flanking NPs. Such an association, of course, is perfectly in tune with the general derogation of traditional power structures that forms the major topic of this poem.)

All of the observations that we have made so far about parallel triadic patterning in this important transitional section of Queen Mab pale, however, when one steps back sufficiently far to view the nine-line passage as a whole:

There is a nobler glory which [S1 survives

Until our being fades, and, solacing

All human care, accompanies its change;]

[S2 Deserts not virtue in the dungeon’s gloom,

And, in the precincts of the palace, guides

Its footsteps through that labyrinth of crime;]

[S3 Imbues his lineaments with dauntlessness,

Even when, from power’s avenging hand, he takes

Its sweetest, last and noblest title—death;]

We have already seen that each of the major clauses constituting this passage (S1, S2, and S3 as indicated above) contains its own internal triadic pattern. Now, however, from a higher vantage point, we can isolate three further correspondences linking, externally, their respective sentential structures:

(a) each matrix clause consists of two smaller clauses linked by a conjunction;

(b) the second clause in each case contains fronted (prepositional or participial) material; and

(c) the second clause in each case also contains in its object NP an anaphoric pronoun (its) which refers to a thematically prominent NP in the first part of the construction.

The complete pattern is perhaps best appreciated when the structures of the three clauses are simplified so as to permit close comparison, as in Figure 12.

In all, therefore, we find in this important passage from Section V of Queen Mab three three-line constructions whose internal syntactic structures embody three independent triadic parallelisms, but which also share the three significant structural features defined in points (a) through (c) above.

Viewed dispassionately, Shelley’s valiant attempt in Queen Mab to persuade his readers of the attainability of a world governed only by people’s pursuit of abstract good seems doomed to failure from the very beginning. Could man operate at all, we find ourselves asking, if deprived of all the major traditional motives for success, “gold,” “sordid fame,” and even “hope of heavenly bliss” (another triad!)? It is this extreme naivete, certainly, that has led critics to reject out of hand any philosophical claim to fame for the poem. Yet, as we saw at the beginning of this discussion, critics do grudgingly acknowledge some poetic merit in Queen Mab. Our perceptual analysis of Shelley’s style demonstrates, I think, how this may be. For it is in the rhetorical force of his perceptual style that Shelley comes closest to convincing us that his vision may have some chance of success, however slight. Only the relentless pressure of his hierarchically ranged triads, that is, shows any sign of carrying the day (it too, of course, falling short as soon as we recognize the enormity of the issues to which Shelley is seeking our essentially irrational, emotional assent).²²

FIG. 12

I have examined at some length the role of syntactic patterning in Queen Mab partly, to be sure, because I have not yet found any reference to that feature of this particular poem’s style in standard critical commentaries. I shall also have more to say in due course about what its presence implies for our conception of Augustan and Romantic syntactic styles. Let us first, however, devote a little more space to considering certain consequences of these perceptual patterns themselves. In particular, let us examine the way in which such patterns may predispose readers toward certain interpretations of quasiambiguous syntactic structures even where those interpretations involve a certain amount of deviance from standard analytical procedures.

The same section of Queen Mab that furnished us with so much material for analysis in the preceding paragraphs—Section V—also includes the lines:

. . . gold:

Before whose image bow the vulgar great,

The vainly rich, the miserable proud,

The mob of peasants, nobles, priests, and kings,

(CWS I, 99: 55-58)

Shelley’s catalog of those who “bow” to commerce’s “all-enslaving power” ranges widely; a technical analysis would take note of seven independent NPs. The task of assigning an internal structure to the larger NP that dominates them and constitutes the ultimate subject of “bow” would still be a simple one, however, under normal circumstances. The overt presence of an interpretively crucial conjunction (and) between the two rightmost conjuncts (“. . . priests and kings”) would dictate that, as in the (a) structure of the figures that follow, all seven coordinate NPs be accorded coequal status, since the syntax of standard English provides that conjunctions between coordinated constituents of equal rank other than the rightmost such conjunction may be, and usually will be, deleted. This technically unremarkable conclusion runs into trouble only because structure (a), shown in Figure 13, at least by my reckoning, fails to represent readers’ most likely interpretation of this passage. In what follows, I shall show that our awareness of Shelley’s characteristic employment of perceptual patterning throughout this poem probably influences us to reject structure (a) and to accept instead a somewhat deviant syntactic analysis.

FIG. 13

The first three NPs of the seven specified by Shelley for censure clearly form a triad, linked by the congruence of their internal syntactic structure in a manner now thoroughly familiar to us and illustrated in the structure (b), shown in Figure 14. Let us hypothesize, then, that the reader relies on his perception of this parallelism to assign these three NPs to a single node within the dominating phrase. Such an assumption gives us in turn structure (c), shown in Figure 15, to represent his overall analysis of these lines.

Even that structure, however, fails to reflect what I see as the most probable interpretation of this passage. The three nouns nobles, priests and kings belong, after all, to one of those classes of “catchwords” that Shelley employs so often as a special code in this poem (see note 21 and the examples to which that note is appended). The noun peasants, by contrast, certainly does not fall into that category. Thus the reader finally opts, I think, for yet another parsing, as illustrated in structure (d), shown in Figure 16. Significantly, hypothesizing triadic internal structure for both NPA and NPB in this analysis also creates an additional triple division within the highest phrase, NPO, thus compounding the already rich hierarchy of triads that makes up this construction.

Let us be quite clear about what is involved in the reanalysis that I am proposing here. In an everyday context, Shelley’s sentence as written could be assigned only structure (a) of the three described, since standard syntactic practice would require in the other two cases inclusion of additional conjunctions to assist readers in identifying their respective more complex structural configurations:

FIG. 14

FIG. 15

FIG. 16

for structure (c): the vulgar great, the vainly rich, and the miserable proud, and the mob of peasants, nobles, priests, and kings.

for structure (d): the vulgar great, the vainly rich, and the miserable proud, the mob of peasants, and nobles, priests, and kings.

In assigning to Shelley’s words the structure shown in (d), then, the reader is not just choosing one of a number of competing structural analyses. He is positively rejecting the most straightforward available analysis in favor of another that is at least mildly deviant in its omission of structurally decisive coordinating conjunctions. His initial willingness to entertain this possibility depends, I take it, on a perceptually induced predisposition to seek out syntactic triads wherever possible in this particular text. Application of such a general heuristic strategy to these specific lines then receives support from other features of the poetry [in the form of the syntactic parallelisms between NPs 1, 2, and 3; in the semantic relatedness of NPs 5, 6, and 7; and finally also in the “triple triad” structure that emerges from the (d) structure].

A stylist who argues along these or similar lines risks being misunderstood on several counts. His readers may assume, first, that he proposes to base his technical account of a given passage on perceptual features in its immediate context. This is certainly not the case. Technical analysis deals with matters of linguistic competence, strictly interpreted. If readers assign structure (d) to Shelley’s lines from Queen Mab as I have suggested, then they do so in outright defiance of their standard linguistic competence, a fact which it is the business of the stylist to make perfectly clear. By other detractors, the stylist may be accused of allowing his own pattern-seeking predilections to dictate structural subtleties that real readers could not, or at least do not, appreciate. This is an altogether more complicated challenge, which we shall have to consider in depth in later chapters. Certainly my openness here to the consideration of perceptual patterns and their consequences has enabled me to advance some intriguing empirical claims about real readers’ behavior. It will take the assessment of a number of such predictions to refine our understanding of the many ways in which perceptual and technical features of style may interact and of how both may then influence reading strategies (if indeed they do so at all). But in any case, as I shall argue in detail in chapter 5, the outcome of such “applied” research projects will in no way affect the perceptual component of the theory of stylistics as such, since stylists’ aims are critical rather than psychological, geared to evaluation by the criterion of insight-fulness rather than empirical accuracy.

Last, and most dangerous of all, is the related accusation that analyses such as these cross my own firmly drawn demarcation line separating “perception” from “interpretation.” Let us briefly reexamine, therefore, the way in which matters of interpretation were introduced into the preceding discussion. It is a fundamental belief of the theoretical linguist that native speakers base their interpretations of each sentence, to a large extent, on the structure that their syntactic competences assign to it. Transformational and even structuralist syntacticians have always depended heavily on eliciting native speakers’ intuitive interpretations when attempting to verify the details of their syntactic analyses. This dependence has never been held, however, to entail the existence of an incestuous relationship between syntax and semantics in linguistic theory; each component of the grammar can be terminologically and methodologically autonomous even though the linguist may make crucial use of the simple, and indeed unavoidable, fact that natural languages exist as means for conveying human thoughts. My stylistic analysis of the three lines from Queen Mab in the preceding paragraphs analogously involved an attempt to construct for them a syntactic structure that would accurately describe the interpretations that readers do seem to place upon them. While the aim was absolutely to mirror those interpretations structurally, though, the argument itself was formal and at no point depended on nonperceptual terms, assumptions, or conclusions.

IV

As an example of intricate syntactic patterning in the work of an English early Romantic poet, the triadic parallelisms of Queen Mab are by no means unique. Unfortunately, considerations of space will prevent me from devoting all the attention I might have wished to the major role played by concentric perceptual patterns in the poetry of another major representative of the same school, Samuel Taylor Coleridge. To demonstrate in full just how typical of his general style such perceptual devices are would demand a textual survey of major proportions. The three passages that I shall have time to discuss have instead been selected principally as further illustrations of some of the theoretical stylistic claims advanced earlier in this chapter and as starting points for the development of additional, more detailed observations on certain aspects of perceptual stylistic analysis.

Coleridge’s “Reflections on Having Left a Place of Retirement,” a poem originally subtitled “A Poem which affects not to be Poetry,” opens with a deceptively simple descriptive passage:²³

Low was our pretty Cot: our tallest Rose

Peep’d at the chamber-window. We could hear

At silent noon, and eve, and early morn,

The Sea’s faint murmur. In the open air

Our Myrtles blossom’d; and across the porch

Thick Jasmins twined: the little landscape round

Was green and woody, . . .

(CPW I, 106: 1-7)

Analysis of the surface structure of these lines reveals just five independent clauses, each set off with either a period or a colon, none with more than four major constituents, as illustrated in this chart:

Nor does transformational analysis reveal for any of these sentences a particularly noteworthy derivational history. PP-Fronting has applied twice, once to each conjunct of sentence 4; a related movement rule sometimes referred to as Niching²⁴ has moved the PP in sentence 3 to a position in front of the direct object (admittedly not usually a very acceptable location for such shifted material); and the transformation of Copula Inversion has exactly reversed the underlying NP-V-ADJ order of sentence 1. None of these processes, though, is uncommon in English, especially in poetry, and on technical evidence alone one might be inclined to accept at face value the claim to an “unaffected” poetic style implicit in Coleridge’s original subtitle.²⁵

A full perceptual analysis, however, reveals unsuspected subtleties in the apparent artlessness of this opening passage. The surface syntactic form of sentence 1, we note, neatly reverses that of sentence 5. Sentences 2 and 4 contain, respectively, clause-final and clause-initial locative PPs. Sentence 3, finally, places its (temporal) PP medially, devoting to it a whole line of verse. If we now concatenate these structures, therefore, there emerges a strikingly symmetrical syntactic form for the passage as a whole, as illustrated in Figure 17.

It would be extremely simple at this stage to attempt to draw a variety of exciting “conclusions” from the discovery of this patterning in Coleridge’s poem. One is tempted, for example, to argue that the mere existence of this concentric mapping gives the lie to the implications of prosaicness and lack of artifice that may or may not have been intended by, but are often drawn from, Coleridge’s use of the term “Conversation Poems” to describe this and other similar works. In a different vein, as I myself have argued at length elsewhere, one might allege that a proper understanding of the perceptual form dominating this passage can help the student of Coleridge’s work to appreciate more fully the peculiar attraction that “dells” such as the one he describes here held for the poet at this point in his career; their natural symmetry, reflected in the contrived syntactic symmetry of the poetry, brought order, coherence, and “relation” to a world that, Coleridge felt sure, was rushing headlong toward disintegration.²⁶ Most intriguing of all, perhaps, since Coleridge is not a poet noted for his humor, our analysis of this passage opens up the possibility of a play on words in line 6. Here Coleridge rejects the standard adverbial form around in favor of the more poetic round. Is this scene, we wonder as a result, a “round” or rounded (or concentric) landscape, besides being of course the “little landscape around”? Actively to pursue any of these hares at this point, however, would be to play right into the hands of those skeptics who already doubt the ability of stylists to conform to their own procedural guidelines. Such observations as those noted in this paragraph are, of course, primarily interpretive and hence out of place at this juncture. Having noted their promise as avenues for future exploration, therefore, let us instead focus rather more narrowly on the perceptual pattern that we have brought to light in “Reflections . . .” itself.

FIG. 17

Given sufficient ingenuity on the part of the investigator, it is often alleged by both friends and foes of stylistics, at least one perceptual pattern—whether parallel, concentric, or chiastic—could be unearthed in the work of almost any poet.²⁷ Our response must probably be in the affirmative, but this suggests only that our next step should be to clarify, if we can, what exactly it is that makes some of those patterns worthy of notice, others mere “technicalities” within the perceptual field. Two factors, I suggest, may reasonably prompt the stylist to assume perceptual patterning in a given passage to be stylistically salient (although the patterning itself will, of course, be present whether or not it turns out to have very much importance in our overall stylistic analysis of the work in question). One such factor, the coincidence of syntactic patterning with some crucial aspect of an interpretation of the text—some such coincidence in fact as the ones I unsportingly alluded to and then abandoned in the preceding paragraph—will not be discussed in detail until we reach chapter 4. A second justification for according perceptual discoveries further consideration, however, and one that is equally relevant to our analysis of Coleridge’s concentric patterning early in “Reflections . . .,” turns out to be entirely proper even at this juncture in our deliberations: this is the fact that very similar perceptual patterns may occur in other compositions by the same poet. Where specific perceptual strategies recur, they become in essence legitimate topics for stylistic commentary by virtue of that simple fact alone.

The tricky task of devising dedicatory verses was not one at which the highly egocentric Coleridge excelled. “To the Rev. George Coleridge” is a deservedly little-read poem in this genre that combines exaggerated panegyric with bombastic but often self-pitying autobiography. Conspicuous, therefore, for their moderation and comparative emotional detachment are a few lines toward the end of the poem in which Coleridge recalls for his brother those moments when “ ‘tis to me an ever-new delight/To talk of thee and thine” (CPW I, 175: 52-53). Among those moments Coleridge numbers, in particular, times

. . . when, as now, on some delicious eve,

We in our sweet sequester’d orchard-plot

Sit on the tree crook’d earth-ward; whose old boughs,

That hang above us in an arborous roof,

Stirr’d by the faint gale of departing May,

Send their loose blossoms slanting o’er our heads!

(CPW I, 175: 56-61)

A simplified diagram of the surface syntactic structure of this lengthy clause, as shown in structure (a) of the two that follow, provides the basis for several statements about perceptual patterning as a feature of Coleridge’s poetic style.

Structure (a) (Figure 18) differs from its most likely deep-structure source, structure (b) (Figure 19), primarily in that two PPs, on some delicious eve and in our sweet sequester’d orchard-plot, have been preposed, presumably by applications of PP-Fronting or the Niching transformation. In surface structure, as a result, the subject NP of S1 (we) occupies the pivotal position in a local concentric arrangement of like syntactic constituents, PPa-NP-PPb. This pattern, however, by no means exhausts the symmetries in this passage. S1, it will be noted, also contains a relative clause (S2) modifying its rightmost NP [the tree (crook’d earth-ward) whose old boughs . . .]. The NPe-S2-NPf substructure that effects this subordination may also, I propose, be considered as lying at the center of a symmetrical pattern. The syntactic evidence supporting my claim comes this time from the observation of a remarkable degree of congruence between the internal configurations of S1 and S2.²⁸ Consider these correspondences:

(a) both clauses are introduced by wh- words (when and whose respectively);

(b) the subject of each clause is separated from its verb by lengthy modifiers (by PPb in the first sentence; by both S3 and S4 in the second);

(c) two major modifiers are embedded preverbally in each sentence (PPa and PPb in S1; S3 and S4 in S2); and

(d) both clauses close with rather simple VPs whose final constituents are PPs (on the tree and o’er our heads respectively).

FIG. 18

FIG. 19

S1 as a whole thus closely resembles S2; all that remains to reveal yet another concentric syntactic pattern is to imagine them balancing one another across the phrase the tree whose boughs. [While logic might perhaps dictate that the center of this pattern should be perceived to lie somewhere in the structural “space,’ so to speak, between the NPe and S2 nodes in structure (a), I suspect that readers are disposed to place such cruxes at nodes or within lexical items rather than between them. Hence my selection of the tree . . . old boughs as the core of this particular symmetry.]

The fact that this clausal concentric pattern appears so shortly after the phrasal PPa-NP-PPb symmetry at the beginning of the same construction (and indeed may be said properly to contain it) leads one in turn to consider the intriguing possibility that Coleridge is seeking to suggest by this perceptual device a relationship between the respective “foci” of the two patterns—the Coleridge family (we) and the old tree on which they sit. Although apparently speculative, this hypothesis receives support from the observation that exactly the same association of the tree with the family figures also emerges from a separate perceptual parallelism in this passage: that noted above as correspondence (d) between the prominent VP-final PPs of S1 and S2. My detection of such a delicate interaction between a pair of concentric perceptual patterns on the one hand and a single parallel pattern on the other should, I hope, satisfy even the strictest formalist. For those less puristic in their approach, however, I will include two further brief observations that stray a little outside the announced bailiwick for this chapter.

If we allow ourselves for a moment to take semantic as well as syntactic features into account as we develop and substantiate our analysis of this passage, then lines 57 and 61 of the poem will be found to define the time at which Coleridge’s scene is set; lines 58 and 60, correspondingly, describe the place in which it is supposed to occur. As a result, we may use the lineation of this passage to demonstrate an additional A-B-C-B-A semantic symmetry. (At the center of this pattern, incidentally, there appear yet again the seated forms of Coleridge and his companions and the tree on which they rest.) My second ancillary observation concerns the relationship of the syntactic patterning in this passage to Coleridge’s repeated use, sometimes even in nonpoetic contexts, of the metaphor of a tree to represent himself or his fortunes. In a 1798 letter, Coleridge wrote:

I am like that Tree, which fronts me. . . . The beings who know how to sympathize with me are my foliage.²⁹

Still more to the point, in an earlier passage from the very poem that we have been studying he exploits the same unusual comparison:

Some have preserv’d me from life’s pelting ills;

But, like a tree with leaves of feeble stem,

If the clouds lasted, and a sudden breeze

Ruffled the boughs, they on my head at once

Dropped the collected shower; . . .

(CPW I, 174: 21-25; emphases mine)

Coleridge then goes on to identify brother George as the solid English oak amid the malevolent forest of the world:

. . . But, all praise to Him

Who gives us all things, . . .

Beneath the impervious covert of one oak,

I’ve rais’d a lowly shed, and know the names

Of Husband and of Father;

(CPW I, 174: 30-31, 33-35)

Looking ahead to the interpretive stage of the critical process, therefore, we might close the discussion of this passage from “To the Rev. George Coleridge” by suggesting that Coleridge’s use of syntactic patterning to link to one another certain of its NPs constitutes an intriguing perceptual continuation and reinforcement of the “arboreal typology” introduced explicitly earlier in the text of the poem.

It would be foolish to conclude this section without making some comments on that most circular of all Coleridge’s poems, “Kubla Khan.” From the “stately pleasure-dome” itself to the “twice five miles” that Kubla “girdled round,”³⁰ and from the mystical invocation at the end of the poem to “Weave a circle round” the inspired poet back again to its striking, phonologically concentric first line, illustrated in Figure 20, every aspect of this amazing work contributes to a network of interlocking rings reminiscent of the famous Olympic emblem.

FIG. 20

I have, however, offered elsewhere a fairly exhaustive stylistic analysis of this poem,³¹ so I shall contribute here only one further brief example of the kind of patterning with which this chapter has been specifically concerned.

The final couplet of “Kubla Khan” reads:

For he on honey-dew hath fed,

And drunk the milk of Paradise.

(CPW I, 298: 53-54)

In the first of these lines, Coleridge creates the lexical compound honey-dew, though the modificational force of honey- might have been just as easily expressed as a postnominal of-prepositional-phrase (compare, for example, the compound apple-juice and its phrasal counterpart juice of apples).³² At the same time he has also preposed the whole PP on honey-dew from its underlying postverbal position. Since the verbs feed and drink constitute a standard semantic collocation, we may use this fact, together with those just uncovered about the syntactic structure of line 53, to construct for the entire couplet a frame of a now very familiar kind as in Figure 21.

FIG. 21

Simple syntactic symmetries of this sort are, I allege, Coleridge’s stylistic “fingerprint”—a perceptually defined fingerprint as vivid, certainly, as those that Fish hopes to extract from technical analyses (see note 2 to this chapter). To recognize such perceptual fingerprints reliably the stylist will need to make himself adept in analysis of the kind developed throughout this chapter; in particular, he must be able to combine real syntactic precision with the less obviously linguistic skills of detecting and concatenating the subtle aesthetic patterns that are the raw materials for this kind of verbal artistry. But not to invest in acquiring these skills and limit oneself strictly to technical remarks would be to miss the opportunity to pinpoint stylistic excellences such as those achieved by Coleridge in the passages that we have discussed, since theories of linguistic competence do not by definition embrace such concepts as structural concentricity or congruence.

V

Even stylistic studies that do take fully into account the existence of a perceptual dimension to syntactic form in poetry are not immune from serious distortions. The commonly accepted view, alluded to earlier in this chapter, that Augustan style is abnormally or even exclusively susceptible to analysis in terms of syntactic patterning is just such a case of mistaken emphasis (see note 17 and the associated discussion above). A variety of theoretical issues were addressed in the course of the preceding pages; taken together, however, they also develop what I take to be a strong case for believing that patterned syntax constitutes a powerful stylistic weapon in the expressive arsenal of at least two major Romantic poets. This being the case, any attempt to base a theory of how the styles of Augustan poets are “similar to” (and, mutatis mutandis, of how they “differ from”) those of their successors in the nineteenth century on the mere presence of syntactic patterning in their poems is surely doomed to failure.

But what can then account for the way in which succeeding generations of critical scholars have fastened on patterning as the quintessential ingredient of characteristically Augustan verse? If patterning itself fails as a touchstone, where should we turn instead for a more adequate standard by which to identify this style which clearly has been felt by many accomplished readers to possess a distinct and discrete identity? In chapter 4, I shall argue that some discussion of favored interpretive strategies is indispensable to the formulation of a fully satisfactory answer to this question. We need not wait until then, however, to make significant improvements, at least, in our understanding of the differences between typically Augustan and Romantic modes of realizing perceptual patterns.

Let us begin by bringing the matter down to the level of specific cases. Early in this chapter, I quoted two lines from Byron’s “Popean poem,” “Hints from Horace”:

Squeezed in ‘Fop’s Alley,’ jostled by the beaux,

Teased with his hat, and trembling for his toes.

Shortly thereafter, I also discussed perceptual parallelism as a dominant feature of Shelley’s style in Queen Mab. What, we now need to ask, distinguishes these two samples of syntactic parallelism, making one Augustan, the other neither more nor less patterned yet equally distinctively Romantic? Equivalently, we might ask what separates concentric patterning in line 253 of Dryden’s “Astraea Redux” (also cited above),

As you meet it, the Land approacheth you,

from the very same perceptual feature when it is repeatedly employed in Coleridge’s poetry of the Romantic school.

Part of the answer lies, I believe, in the very special relationship established by Augustan poetic convention between perceptual patterning on the one hand and perceptual scale on the other. Statements about the aesthetic scale of a work, it will be recalled, define the dimensions within which each artist chooses to work his particular brand of creative magic—sonatina or concerto, summerhouse or cathedral. What is noticeable about perceptual patterns in Romantic poetry generally is that those dimensions appear to be very much a matter for individual choice. In this chapter alone, we have seen parallelisms and concentricities created by Romantic poets within the space of a few words (Shelley’s lexical triads, for instance), within a line or two (the closing couplet of “Kubla Khan” is such a case), or over the course of an entire verse paragraph (as with the opening section of the conversation poem “Reflections . . .”). In many of those cases, the pattern has occupied a rather awkward fraction of a poetic line; in others, we have had to link half lines or even odd words with the lines preceding or following them in order to allow a given perceptual configuration to emerge. As a practical matter, in fact, it is altogether simpler to concentrate on grammatical units when examining a Romantic text for perceptual patterning than on conventional poetic units of whatever size or type; the orderliness implied by poetic structures such as the line, the couplet, or the stanza is as likely to obscure syntactic arrangements present in Romantic poetry as to reveal them.

The overall effect of Romantic perceptual patterning might usefully be compared to that created when a child builds a wall with a set of old-fashioned wooden building blocks. Each individual block may be right-angled, to be sure, but the considerable variety of their specific dimensions and colors makes for a final appearance of complexity and “irregularity.” Where a pattern emerges from the child’s construction, its parts may or may not correspond to individual blocks built into the fabric. In Augustan poems, by contrast, we find patterns far more akin to the “regular” brickwork of a Victorian building, since the sizes of the stylistic building blocks available to Augustan poets were strictly regulated by convention. The basic verse unit employed was set rigidly by metrical requirements at ten syllables, larger units almost always turning out to be even-number multiples of that basic structure because of the demands of the simple couplet rhyme-scheme. At the same time, poetic convention further dictated for the Augustan poet that these basic verse units provide a perfectly fitted frame for any stylistic effects he might also wish to introduce. It thus becomes a rather simple matter to isolate and describe perceptual patterns in such poems since their size and location relative to the poetic line have been predetermined by the poet’s initial stylistic choice (or that of his “school”), a governing choice made initially in the subfield of perceptual scale.

Like all analogies, my comparison of the role of scale in poetic syntax to techniques for laying bricks will probably not survive close scrutiny. What I hope to have clarified by using it is that the constant recurrence of syntactic patterns within segments of poetry of the same arbitrary length, rather than their mere presence in the text, is what makes them so prominent in Augustan verse.³³ This theoretical claim of course has implications for an area far broader than the specific problem that suggested it here. My general conclusion should transfer rather smoothly, in fact, to cases in which the “segment of poetry” involved is not the heroic couplet; a similarly recurrent perceptual pattern within each separate stanza of a ballad, for instance, or a pattern that appears repeatedly within the triplets of a poem in terza rima should, I predict, benefit from precisely similar perceptual prominence. It is only the extraordinary dominance of one particular closed form throughout more than a century of English poetry, in fact—the absolute dictatorship, so to speak, of the heroic couplet—that has led us to associate too simplistically the syntactic patterning that it highlights with the genre itself. Stylistic analysis can now do literary criticism a considerable service by once again dissociating features of pattern from features of scale. As a result, readers may be taught to see perceptual patterning as an integral feature of many, potentially extremely varied poetic styles—the Romantic, in particular, as much as the Augustan—and to view Augustan poetic syntax itself as adequately described in terms of perceptual patterning only if the analysis in question also includes details of the characteristic scale of those patterns. In poetry of periods and genres totally unrelated to Neoclassical heroic couplet satire, readers will continue to expect varying degrees of syntactic patterning; they will also, however, appreciate fully its somewhat less obvious and less conventional character. Such a revaluation, I feel, is sorely needed in scholarship, in our textbooks, and in the classroom instruction of new and naive readers.

VI

One section of Pope’s tongue-in-cheek apologia for the fickle behavior of Belinda, his heroine in “The Rape of the Lock,” catalogs the “varying Vanities” with which the “Sylphs” contrive to stock “the moving Toyshop” that is every young belle’s heart (PAP, 221: 91-104). Pope describes a “giddy Circle,”

Where Wigs with Wigs, with Sword-knots Sword-knots strive,

Beaus banish Beaus, and Coaches Coaches drive.

(PAP, 221: 101-102)

Pope’s choice of syntactic form here displays far less perceptual patterning than would have existed had he left the four clauses that constitute this couplet in their underlying forms. Metrical considerations aside, in fact, Pope could easily have used two neatly matched pairs of clauses, the first repeating the syntactic configuration NP-V-PP, the second employing parallel NP-V-NP structures:

Where wigs strive with wigs, sword-knots strive with sword-knots, beaus banish beaus, and coaches drive (out) coaches.

Instead, the attested surface form shows that he has allowed Gapping, PP-Fronting, and some such rule as Dillon’s Verb Final to eradicate almost all signs of parallelism from the surface structure of his sentence.³⁴

By adopting this ostensibly un-Augustan tactic, Pope succeeds in achieving a stylistic effect of an altogether different kind. Perceptually, the surface structure that results (and perceptual effects are, it will be recalled, surface structure related by definition) jumbles together nouns and prepositions in an unruly and confusing mess. Instead of orderly NP-V-NP or NP-V-PP clauses, the reader is confronted in line 101, for instance, with the thoroughly improbable syntactic sequence NP-P-NP-P-NP-NP and may legitimately feel that he is being openly challenged to uncover some halfway plausible analysis. We will address the issue of why Pope chooses this particular approach in just a moment.

But contrast this case first with two lines from Dryden’s “Mac Flecknoe,” in which the “aged Prince” of “Non-sense,” Flecknoe, “burst[s] out” in an absurd mock-panegyric to his heir-apparent, Shadwell:

Heavens bless my Son, from Ireland let him reign

To farr Barbadoes on the Western main;

(WJD II, 58: 139-140)

As Pope did in “The Rape of the Lock,” Dryden here uses PP-Fronting to distort the underlying word order of his clause’s VP [underlyingly (let him) reign from Ireland to Barbadoes . . .]. In this case, however, the surface configuration that results separates from one another constituents that had in deep structure been juxtaposed, an effect precisely the reverse of the excessive compression witnessed in the lines from Pope’s poem.

As might be expected, neither of these great artists misses his own interpretive mark in thus selecting contrasted perceptual strategies. Dryden certainly achieves his purpose—to emphasize (or to have Flecknoe emphasize) the magnificent extent of the “Dominion” that will be Shadwell’s inheritance—just as effectively as Pope, in “The Rape of the Lock,” recreates syntactically the bewildering “mystick Mazes” of court society. To appreciate fully the achievements of each poet in this area, though, it is clear that we shall first need to be able to discriminate properly at the perceptual level itself between Dryden’s stressing of syntactic distance as his major stylistic effect and Pope’s reliance instead on close proximity between syntactic constituents in selecting his surface form.

With the mention of syntactic distance (and its opposite, so to speak, proximity) we return to the category of perceptual effects with which this chapter began—effects dependent on our sense of aesthetic proportion. The examples of stylistic delay in Dryden’s Killigrew ode and Byron’s “Childish Recollections” discussed then, after all, depended on precisely the same fundamental attributes of syntactic structure as the cases cited here from “The Rape of the Lock” and “Mac Flecknoe.” All four passages exploit particularly the temporal and spatial aspects of the linguistic code. (It is, by the way, not clear to me that these are separate syntactic concepts. In practice, it is certainly clear that readers will tend to associate some examples of suspension with temporal perception, since they will be vividly aware of the actual delay caused in their processing of a long or involved sentence. Cases such as Dryden’s separation of “Ireland” from “Barbadoes,” by contrast, will be related most easily to the spatial dimension, since the topic being addressed is itself geographical. These biases result, however, from features built into the semantic contexts of the passages concerned;syntactically speaking, such cases remain indistinguishable. In each, the relatively close or relatively remote structural placement of syntactically related material is what contributes primarily to the reader’s perception of the passage as a whole.) No abstract, formal pattern results from these poets’ syntactic choices (replacing like constituents with letters of the alphabet, for example, elicits no internal regularities suggestive of deliberate patterning). Neither the extrinsic size of the constructions involved, furthermore, nor the absolute distance between their component parts has any bearing on our appreciation of the style of any of the passages (as would be anticipated in cases dependent on perceptual scale). Rather, as all of these examples illustrate, the description of proportional effects, in stylistics as elsewhere, differs from other perceptual activities in that it demands the use of open-ended scales and depends on purely relativistic assessments of form.

That various combinations of proximity and distance may be (and have been) interwoven by poets to yield fascinatingly subtle syntactic fabrics has been beautifully illustrated by Cureton in chapter 4 of his dissertation. In general, I have little to add to his illuminating discussion. The pages that have intervened since my last mention of his work, though, do allow me now to clarify somewhat the few reservations I have with regard to his taxonomic principles. My own belief that spatial and temporal distance are syntactically indistinguishable, in particular, accounts for my rejection of his use of this dichotomy as a taxonomic prime. His classification obscures what I see as very important syntactic similarities between cases of what he calls (spatial) “inclusion” on the one hand, and cases of (temporal) “simultaneity” on the other.³⁵ Especially after my comments in previous sections of this chapter, it should also be simple to predict that I would be uneasy with Cureton’s inclusion of categories with titles such as “symmetry” in a list of what are for the most part proportional stylistic effects. My own preferences are for a descriptive machinery that emphasizes the very different natures of perceptual effects based on aesthetic pattern and on aesthetic proportion respectively. Neither of these theoretical reservations, however, should be seen to detract at all from the variety, informativeness, and insightfulness of the material that Cureton has amassed. His textual examples suggest some of the richness that even a few variations on the simple theme of relative distance in surface syntactic form can in fact provide—variations for which Cureton finds terms such as “contiguity, inclusion, fusion, interruption” and “intrusion.” To pile up further examples here would be to waste ink on a task already well done.

I shall conclude this chapter therefore with just a single example from my own work, an example which illustrates both the subtlety of the effects that can be achieved merely by manipulating syntactic proportion and the important role that those effects may play in establishing overall coherence in a passage. William Wordsworth’s narrator in The Ruined Cottage finds himself, as the poem opens, at the edge of “a bare wide Common,” and describes for the reader the landscape around him.³⁶ In the course of this description, he notes how

. . . all the northern downs,

In clearer air ascending, shewed far off

Their surfaces with shadows dappled o’er

Of deep embattled clouds.

(TMH: 3–6)

The derivational history of this sentence must, I take it, look something like this:

Underlying word order:

. . . showed far off [NP their surfaces [S[VP dappled o’er [PP1 with

shadows [PP2 of deep embattled clouds]]]]]

After PP-Fronting of PP1:

. . . showed far off [NP their surfaces [S[PP1 with shadows [PP2 of deep

embattled clouds]] [VP dappled o’er]]]

After Extraposition-of-PP applied to PP2:³⁷

. . . showed far off [NP their surfaces [S[PP1 with shadows] [VP dappled

o’er] [PP2 of deep embattled clouds]]]

While it is, of course, with the surface structure of this sentence that we are primarily concerned for the purposes of perceptual analysis, a review of the full derivation clarifies one important consequence of Wordsworth’s syntactic permutations for our perception of that surface form. Four major syntactic constituents have to be correctly located in an overall hierarchical structure if this sentence is to be adequately analyzed: the noun-phrase object (NP in the labeled bracketings of the derivation detailed above); its modifying verb-phrase (VP), which in this case represents the residue of a nonrestrictive relative clause (S); and two prepositional phrases, one (PP2) embedded within the other (PP1). The version of this sentence that Wordsworth actually included in The Ruined Cottage places those four crucial constituents not in the most derivationally transparent A-B-C-D order illustrated in the “Underlying word order” above, but in the disjoint sequence A-C-B-D. The result, it seems to me, is a charming syntactic metaphor for the dappled sunlight described by Wordsworth’s narrator; the now separated and alternating PPs, in effect, mirror syntactically what Wordsworth later calls the “interposed” effects of light and shade on the “northern downs.”³⁸

I shall return in the next chapter to a fuller discussion of the interpretive success that Wordsworth scores by his use of this particular perceptual strategy. For the present I content myself (one last time) with the purely perceptual observation that it is the precisely handled separation and distinctive placement of syntactic material in this particular passage that provides its stylistic “edge,” the respective distances between several syntactic constituents being carefully balanced to achieve a particularly pleasing alternation.

VII

This chapter has, I hope, started a number of hares. I have covered several examples under each of the three main subcategories of perceptual effects resulting from poets’ deployment of syntactic forms: those related to pattern, those related to scale, and those related to proportion. Along the way, various related issues may have demanded attention for a while: the nature of Augustan poetic syntax, for instance, or the ways in which perceptual patterning may affect readers’ parsings of complex structures. I close, however, with a simple reaffirmation of the claim that has lain, unspoken perhaps, behind all that has been said. However excellent, however linguistically advanced its technical methods, and however appealing its critical insights, stylistics will disregard at its peril the completely independent (and independently complex) role of aesthetics in the functioning of syntax as a medium for poetic expression. Stylists need desperately to develop both a vocabulary and a methodology adequate to the very considerable degree of sophistication of poets’ achievements in this important area.