CHAPTER SEVEN

Some Values of Complex Processing

IN this chapter we address a second major topic in literary criticism that has remained on the periphery of earlier chapters, namely, the aesthetic effect, function, or value of ambiguity and difficulty. There are, very broadly, two basic ways that expressive value has been ascribed to difficult writing. The first is the method of Empson, which has been used primarily for poetry and which works at a level of close or fine detail. The second method applies at a higher level of generality and to notably mannered writers (such as James and Faulkner). We will begin with the first ’school' and with an examination of Empson’s practice in Seven Types of Ambiguity since it has been a model for 'close reading' for several decades. Empson discusses a number of passages involving parsing difficulties, pronominal reference, and ellipsis, including the following lines from “The Waste Land”, some of which were cited in (III.35):

(1)In vials of ivory and coloured glass
Unstoppered, lurked her strange synthetic perfumes,
Unguent, powdered, or liquid—troubled, confused
And drowned the sense in odours; stirred by the air
That freshened from the window, these ascended
In fattening the prolonged candleflames.

In regard to the third line he says, “after powdered and the two similar words have acted as adjectives, it gives a sense of swooning or squinting, or the stirring of things seen through heat convection currents, to think of troubled and confused as verbs” (p. 78). The experience of the reader, that is, directly enacts the thing being described. Empson further notes that these could have perfumes, sense, and/or odours as antecedent, observing as the effect, “there is a curious heightening of the sense of texture from all this dalliance; a suspension of all need for active decision” (p. 78). The language here is uncannily reminiscent of Chapter Five: Empson appears to be saying that the reader begins to give up active structuring of the sentences and to surrender to a flow of impressions, thus experiencing a kind of passivity and lassitude in processing which here reflects the theme of debilitating sensuality.

Perceptual uncertainty is again the source of the effect described in the next passage, which is also from T. S. Eliot:

(2) Webster was much possessed by death
And saw the skull beneath the skin;
And breastless creatures underground
Leaned backward with a lipless grin.

Empson points out that the last two lines could be taken as an independent clause ("meanwhile they leaned...") or as coordinate to skull ("saw the skull beneath the skin and saw creatures that were leaned backward...."), concluding that “The verse, whose point is the knowledge of what is beyond knowledge, is made much more eerie by this slight doubt” (p. 79).

Since Empson’s method is essentially that of pointing out a correlation between processing and the theme or point of the passage, it will fail to convince if either term is inaccurately described. Some of his comments on a passage from Samuel Johnson’s “The Vanity of Human Wishes” are based on difficulties that seem to me to be unreal:

(3) What murdered Wentworth, and what exiled Hyde,
By kings protected, and to kings allied?
What but their wish indulged in courts to shine,
And power too great to keep, or to resign?

Empson first considers whether power is coordinate to wish or to shine (this second possibility seems forced to me, as it would require assumption of an ellipsed for “wish to shine and [for] power.. .”). He then asks what the last line might mean in terms of unexpressed causality (why could it not be kept or resigned?). He concludes, “the line, I think, conveys by its knotted complexity, by the sense that there are grammatical depths the casual reader has not plumbed, some such ideas of fatal involution as these I have been elaborating” (pp. 69-70). Here one must register a demurral: the last line is not grammatically knotted, and in fact I experience little difficulty with respect to the “unexpressed causality,” which I trace to a general maxim,

One may of course ask why this maxim is true, but one does not have to ask, or run over traditional beliefs about power, any more than one must think up a large portion of elementary economics to comprehend the maxim,

When supply exceeds demand, prices fall.

This example illustrates the usefulness of a model of processing which is able to predict difficulties: it provides a kind of independent check on whether a putative difficulty is a real one. This is a point we have touched on before in regard to ellipsis of logical subjects with participials (Chapter III), the 'two spots or one' argument between Fish and Rader (Chapters IV and V), and ellipsis of various things in James (Chapter V). In all of these cases, an adequate model of processing can set certain limits to the subjectivity of the critic.

There is perhaps no parallel check on 'the point' of a passage, though error and fudging are possible here also. One might, for example, quibble slightly about the notion that “knowledge of what is beyond knowledge” is the point of (2). In general, Empson’s method is most convincing when the point is pretty directly the propositional content of the passage, as it was in the first part of example (1) and as it is, for that matter, in the passage from Paradise Lost describing the entry of the devils into Pandemonium (Chapter II [41]) or (19) of Chapter V from The Prelude on the inability to distinguish. Stephen Booth uses Empson’s method heavily: the following comment on the slippery syntax of “Sonnet 33” is representative: “Each violation of the reader’s confidence in his expectations about a syntactical pattern evokes a miniature experience for the reader that mirrors the experience of betrayed expectations which is the subject of the poem....”¹ The method is less convincing when the point is more remote from propositional content and approaches vacuity when the point is something imagined but not said. Paul Alpers is following Empson’s procedure when he cites the following lines from Marlowe:

(4) Home when he came, he seem'd not to be there,
But like exiled aire thrust from his sphere,
Set in a forren place, and straight from thence,
Alcides like, by mightie violence,
He would have chac'd away the swelling maine,
That him from her unjustly did detaine.

and comments, “The lines are rather hard to get into focus.... But the very active sentence structuring here demanded of us is turned to poetic use: the concentrated effort we give to these lines supports the sense of heroic expenditure of strength in those that follow.”² I agree that the lines are hard to get into focus: the main question is whether set in a forren place is another complement of seem'd ("He seem'd not there to be but... set in a forren place") or attaches to aire parallel to thrust ("aire thrust from his sphere, set in a forren place"). Set in aforren place is not too good semantically as a modifier of aire because it calls for a physical object more strongly than thrust from his sphere does. The extra energy and resolution come when we reject this latter possibility, resisting the temptation to attach set... to the immediately preceding phrase, and skip back over to he instead. Alper’s statement of 'the point', however, seems wrong to me: there is no “heroic expenditure of strength” here, only a fairly grandiose (and unacted) desire to sweep the ocean aside.

There is, by the way, another possibility here based on an emendation of aire to heir suggested by L. C. Martin.³ This textual crux is worth a bit of our attention even though it requires an excursus because it raises the question of why one would choose an apparently ungrammatical reading over a grammatical one. Aire has the textual authority but was a common Elizabethan spelling for heir. Reading “heir,” however, poses the difficulty that one would expect an in front of it—i.e., heir, unlike air, is a count noun requiring an article in the singular. Actually, the reader familiar with Elizabethan poetry might know that a(n) was frequently omitted following like, as, and then,⁴ but the absence of an is a reason not to read heir for the modern reader. Millar McClure rejects the emendation (or modernization) to heir because he says the line makes sense with air.⁵ True enough, but the whole construction makes more sense with heir, which accounts for exiled (one exiles heirs but expels air), set in aforren place, which can now attach to heir (linking up to exiled), and the water which separates the exile from his proper sphere (the English characteristically think of exile in terms of leaving the island). The general point is that the sort of thing textual critics and editors do is simply a specialized version of what the reader must do in deciding between two readings: the decision is an interpretive act.⁶

There are other variations in the way the point is defined. In a recent “Note on Ruskin’s Mythography” Frederick Kirchhoff argues that Ruskin’s style in The Queen of the Air touched on in Chapter IV is designed to evoke in the reader the “mythic consciousness” which Ruskin desires to recover. In Surprised by Sin, for another example, Stanley Fish argues that the general point is the fallibility of human understanding, and that this is enacted by the reader as he misperceives. Fish employs similar arguments in Self-Consuming Artifacts. It might be argued that we have shifted from aesthetic value to didactic value, but the common ground is that one is arguing that the language “does what it says,” or leads the reader through a process that replicates or enacts the meaning of the passage.

Difficulties in Faulkner and James do not lend themselves to this sort of treatment, and the expressive function of complexity has been judged instead according to an odd canon of verisimilitude: the language of a passage is effective if it conveys “consciousness” of one sort or another. The following statements are fairly typical:

I have tried to trick the reader by quoting out of context and replacing the authors' names by he as indicated: the first two citations are describing Faulkner, the second two James. The similarity of the passages is a bit unsettling, since the experiences provided by these authors are so diverse, suggesting that some further analysis is in order. One can with effort discriminate between the passages on James and those on Faulkner: those on Faulkner describe experience only partly ordered or conceptualized, while those on James describe the process of conceptualization. Such differentiation seems to go in the right direction, and our question becomes, “What is the basis for it in the language, or, how can it be clarified by analysis of the difficulties peculiar to each style?”

In an important sense, the passages cited are describing illusions, in that the reader’s direct experience is entirely verbal, yet the critics describe senses of pre-verbal or coming-to-be-verbal experience. Certain passages in James are so overt in this regard as almost to call attention to their artifice, as for example the following:

(5) ... many marks of a taste whose discriminations might perhaps have been called eccentric. He guessed at intense little preferences and sharp little exclusions, a deep suspicion of the vulgar and a personal view of the right. The general result of this was something for which he had no name on the spot quite ready, but something he would have come nearest to naming in speaking of it as the air of supreme respectability, the consciousness, small, still, reserved, but none the less distinct and diffused, of private honour. The air of supreme respectability—that was a strange blank wall for his adventure to have brought him to break his nose against. It had in fact, as he was now aware, filled all the approaches....

[James, AMB, 146]

The Jamesian consciousness does not simply and quickly name things and impressions—it struggles through the process of defining and articulating its initial impressions. Once the impression has been named properly, however, the name becomes a reality one can break his nose against and a point of departure for further meditation. The pronouns in this passage are typical: they are at first vague in reference ("the general result of this/' “it”) and then highly specific ("that was a strange blank wall," "it had filled ..."). Not all impressions receive this lavish articulation: the hedged might perhaps have been called eccentric is an 'easy' naming that is not the stimulus to thinking the next item is. It is perhaps worth stressing that this use of appositives to define and specify is not inherent in the construction. Stevens, for example, uses them in his early manner to underscore the arbitrariness of the 'names' for reality, though, as Helen Vendler argues, he uses them in his later manner more in the fashion of James to sharpen a formulation.¹¹

In James, this process of groping toward articulation does not (usually) create the impression of a fumbling or inarticulate mind but rather of an acute and exacting one. What then is the source of the difficulty? As Tompkins' remarks suggest, the difficulties arise (for the reader) from the parenthetical and qualifying material which might be said to be the product of a hyper-acute consciousness—one which deflects and obstructs the development of its own thinking toward its chosen goal, or, if you prefer, enriches and qualifies it even before it is fully there to be qualified.

There is an important difference between the remarks of Menikoff and those of Tompkins, however, that returns us to the central point of artifice. Menikoff attributes the impression created by the style to the character, but Tompkins attributes it to the narrative voice—it is a mode of speaking, not thinking. Tompkins at the end of the passage says the interruptions “do not seem the result of any planned dislocation of syntax,” by which she seems to mean they are not patently 'rhetorical' in the tradition of the high style (not, for example, produced by inversions of normal order). The interruptions do nonetheless regularly fall in certain unusual places in James' late style. In the late style, and only there, the 'qualifications' occur very heavily between auxiliary and main verbs, and between nouns and prepositional complements (there are illustrations of both in [5]). Now, interruptions at these points produce special perceptual problems for the reader (as I have shown in Chapter Two), but these points are not what one might call natural joints in thoughts. That is an admittedly unsupportable statement, but it does seem to me that one does not hesitate or ponder over what verb to use after a modal or what complement to add to a noun as one thinks—the concept is already there. Tompkins is right, I believe, to emphasize the voice speaking: the effect is of a voice that knows what it is going to say. The qualifying thought is an 'extra' qualifying what in many cases has not yet been said (or processed by the reader) but is already in the narrator’s mind calling forth the qualification. Similarly, the speaker presumably knows the referents of nouns and pronouns, however much effort the reader must expend to figure them out. This is why readers sometimes feel teased or pestered by James, wanting him to get to the point which he knows and is withholding from us. Likewise, James frequently splits an idiomatic string which is only clearly recognizable as an idiom when the last word is read:

(6) He pulled himself then at last together....

[James, AMB, 67]

Again, the speaker cannot be uncertain of what is to follow. Thus the difficulty of perception experienced by the reader is not precisely that experienced by the speaking consciousness: the effort we experience does, however, tend to be projected onto the character as his effort to bring a complex impression to full articulation. Similarly, Ian Watt gives as a function of the abundance of negatives in the opening of The Ambassadors that “it enacts Strether’s tendency to hesitation and qualification.”¹² The abundance does not enact anything—the reader does, and the hesitation, in particular, is surely more his than Strether’s: Strether is never more sure of his plans and intentions than at this point in the novel. It is not so much that James “designs sentences whose very structure simulates the process of the mind, the manner in which we apprehend or perceive an idea,” as that he designs sentences whose structures trigger processes which simulate (but do not mirror) this “process of the mind.”

The difficulties with Faulkner are not the result of parenthesis and qualification interrupting the sentences at strange points. R. W. Short established that the source of difficulty in James is not the length of sentence,¹³ but length is certainly a factor in Faulkner, and the chief sources of length are relative, participial, and adverbial clauses: phrase and clauses sprout modifiers, and these modifiers have modifiers, at every turn. When a Subject noun phrase becomes weighted with appositives, participials, and relative clauses, it is simply hard to hold in the mind as a unit while processing the other parts. In general, the primary source of confusion is the profusion of things and properties which may overload the reader’s capacity for sorting out—Beck’s “saturated solution” metaphor is not bad and is certainly better than Zoellner’s metaphor of the “mass in a lump.” The appositives, for example, add to and extend the initial phrase rather than narrowing or specifying it more exactly as they do in James. The effect in Faulkner is a sense of great richness, each thing a plenum bearing relations to other things even more diverse and numerous than the teller can pack in. The relative and adverbial clauses tie each thing and event into so many other events and relations that the current function in the sentence recedes in importance and is lost. Conrad Aiken described this quality of fullness in relation to the speaking voice:

Warren Beck and more recent critics tend to accept the illusion and to speak of Faulkner’s sentences as presenting “consciousness” directly. Thus Aiken’s “terrifically concentrated effort” of telling becomes Beck’s heightened, “crowded, composite consciousness” and Zoellner’s “total consciousness.”

As noted in the previous chapter, Aiken sketched a defense of Faulkner’s style in terms of the hypnotic immersion that it promotes in the reader. This theme has been expanded by other critics, notably Walter Slatoff. Slatoff is not quite ready to say flatly that Faulkner induces a hypnotic trance or dream state, but he does argue that Faulkner’s “presentation,” which thwarts the usual categories and processes of rational thought, has positive value insofar as it frees “the emotional life from the trammels of critical thinking.”14 Indeed, he argues that Faulkner’s style is a means of containing and articulating, but not logically resolving, a great deal of ambivalence and uncertainty about the human condition: there is resolution and release of an emotional kind, but the reader is often unable to say exactly what has been resolved. I believe this is a just assessment: a minor example is the lovely passage cited at the end of Chapter Four of the vision of the eternally renewing hunt, and a larger example would be Chapter Twenty of Light in August, which contains numerous sentences still largely uninterpretable for me, but does afford this sort of emotional release in the final image.

A curious result of this line of thinking is that the claim of verisimilitude falls out: if the value of the language is the experience it gives the reader, then it really does not matter whether it represents consciousness “like it really is.” The reader may never experience a moment of consciousness like it except when reading more Faulkner—indeed, in a strict sense, he won't—but the value of the experience of reading Faulkner can simply be taken to be the value one places on that experience. And the same can be said of all the authors we read.