Chapter VI, Part 1

 

VI. Style in the Study of Text and Discourse

 

VI.A  Style in theory and practice

1. In theoretical terms, the style of a speaker or author, or of a text or discourse, is the array of actual choices made from the vast repertory of potential choices. In practical terms, a ‘style’ gets attributed to a discourse [1622] or participant [1623].

[1622] a more colourful, gossipy style of writing took over from the self-consciously poetic late Victorian style. (Sport and the British)

[1623] Phrases such as ‘there is no alternative’ […] sum up the spirit of what she feels and argues. The warrior image is aided by her forceful style of speaking (Thatcherism)

Yet, as so often in our societies, the relation between theory and practice by no means constitutes a transparent dialectic (cf. I.2ff). Most of the many occurrences of ‘style’ as a Noun (e.g. 10,381 in the BNC alone) do not stipulate the relevant qualities or features, e.g., what renders a style ‘colourful’ as opposed to ‘late Victorian’. For most speakers and writers, theories of ‘style’ have remained implicit, even though projects to render them explicit have a venerable history.

2. In the study of text and discourse, our theory might define a style — along with a genre, register, or text type — as an intermediary system, more specific than the language and more general than the discourse (cf. II.153). Routinely interacting, these systems mediate between the entire potential available to text producers and the individual text to be produced. Such systems appear paradoxical: precise yet flexible, constrained yet adaptable. Copious stylistic criteria guide a search and selection process that rapidly and efficiently terminates in uttering or inscribing combinations of particular choices in Colligations and Collocations. Such a mode of operation could profit from massively parallel distributed processing whereby ongoing searches can interchange partial results.1 

3. Evidently, discourse sustains a multiple dialectic between common and uncommon, between general and specific, between habit and innovation — between what is typically said and what is actually said. If so, a language remains in the process of being both confirmed and constituted by discourse (I.36), which is neither totally fixed in advance nor totally free at the moment (cf. I.26; II.21, 157). This design is great to exploit in practice, but daunting to account for in theory.

VI.B Ancient studies of style

4. To a modern view, ancient studies of style, notably in the extant writings of Aristotle, suggest an intriguing of the three disciplines mentioned back in II.7. For poetics, he favoured combining ‘strange’ and ‘proper’ choices [1624].2

[1624] The perfection of style is to be clear without being mean. […] That diction […] is lofty and raised above the commonplace which employs […] strange (or rare) words, metaphorical, lengthened, [or] ornamental, […] while the use of proper words will make it perspicuous. (Poetics, 9.4)

For rhetoric to deliver a ‘speech’, he lauded ‘being clear’ over ‘poetic’ [1625], yet kept ‘metaphor’ as one ‘pre-eminent’ means for being ‘clear’ as well as ‘pleasant’ and ‘unfamiliar’, whilst enlisting ‘beautiful words’ [1626].

[1625] the virtue of style is to be clear, [whereas] poetic style [is] inappropriate to a speech; we should make little use of  exotic, compound, and artificial (Rhetoric, 3.2)3

[1626] Metaphor pre-eminently involves clarity, pleasantness, and unfamiliarity; [and] draws from words that are beautiful in sound or in effect or in image (same)

For grammar, he rated ‘extended style’ below ‘contracted style’, as if merging my concepts of appropriate and effective with efficient (cf. II.130).

[1627] extended style [is] unpleasant because of its unboundedness; [in] contracted style, […] a clause having an end in itself and an easily surveyed magnitude [is] both pleasant and easily learned, [and] can be delivered without drawing breath, as a whole (3.9).

And, as noted back in IV.4 and signalled again in [1627], he integrated prosody too.

5. An unexpected and jarring Platonic note intruded when Aristotle expressed reservations that, in respect to ‘the study of style’,

[1628] it makes a difference for explanation to speak in one way or another, but not that much; all these things are rather mere display directed at the listener. (3.1)

[1629] Justice requires contention from the facts themselves, so that all other aspects apart from demonstration are ancillary. Yet these have a great effect, […] because of the baseness of the audience. (3.1)

These reservations might cloud the prospects for a dialectic between theory and practice. Aristotle based the theory upon abstract values (lofty, clear, pleasant, beautiful, etc.) which in concrete practice would be applied by authors and experienced by ‘audiences’. Yet a disparate theory from ancient philosophy, preoccupied with epistemology and ethics, and thus assigning abstract values to truth and truthfulness, instilled some ambivalence about the authenticity of stylistic values versus ‘mere display’ and about the judgements of real audiences. Plato’s own reservations against the ‘imitative poet’ come to mind:

[1630] We shall be right in refusing to admit him into a well-ordered State, because he awakens and nourishes and strengthens the feelings and impairs the reason. […] He is a manufacturer of images and is very far removed from the truth. (Republic, 41)4

6. Even so, some dialectic was a professional requirement for an authority on style back then. The theoretician earned a livelihood imparting practical training in ‘style’ aimed at political or juridical benefits by winning the acclaim of ‘audiences’. And these might indeed not share the theories of authors (or orators) about what is ‘clear’ or ‘pleasant’ and might in practice value ‘display’ and ‘effect’ over ‘justice’ and ‘fact’. The theoretician was uneasily entrained in inculcating stylistic and rhetorical skills whilst surmising that they could be unethically exploited.5

7. The Aristotelian tradition was ably continued and elaborated in the treatise of Demetrius whose accredited English title is On Style.6 He recognised four styles and took more care than Aristotle to illustrate them abundantly with celebrated orators and poets.7 The ‘elevated’ or 'eloquent style' derives ‘impressiveness’ and  ‘stateliness’ from ‘repetition’, ‘long syllables’, ‘compound words’, and lengthy complex, and parallel sentences;8 and also from ‘metaphor’, ‘allegorical language’, and ‘onomatopoeic words’. The ‘elegant style’ relies on ‘geniality’, ‘grace’, ‘smoothness’, and a ‘pleasant cadence’. The ‘plain style’ seeks ‘clearness’ and ‘lucidity’ from ‘current words’, ‘familiar diction’, ‘vivid representation’, and ‘homely subject-matter’, ‘shunning ambiguities’ and long sentences, and adhering to ‘the natural order of the words’. And the ‘forcible style’ prefers ‘brevity and energy’, and ‘in many passages harshness’ and ‘vehemence’, plus ‘a discreet use of elaborate language’. Like Aristotle, Demetrius betrayed some inconsistency and ambivalence, namely about both the plain [1631-32] and the elaborate [1633-34].

[1631] Everything ordinary is trivial, and so fails to win admiration. (§ 60)

[1632] Diction [that is] grandiose, elaborate, and distinctly out of the ordinary […] pos-sesses the needed gravity, whereas usual and current words, though clear, are unimpressive and liable to be held cheap. (§ 77)

[1633] Overloading with figures […] betokens lack of taste and inequality of style (§ 67)

[1634] Exuberant and inflated language must not be sought after in a style meant to carry conviction. (§ 221)

Similarly, he asserted that ‘metaphors impart a special charm and grandeur to style’, but warned against their being ‘numerous’, ‘far-fetched’, or ‘daring’, or ‘conducing to triviality’ and ‘pettiness’ (§ 78, 83f).

8. Like certain other branches of knowledge and study, such as philosophy and geometry, ancient studies of style have hardly fallen into obsolescence. Indeed, they overshadow modern studies which are more restricted and self-consciously ‘methodical’ and ‘scientific’, and which fail to resolve the ambivalences that perplexed the ancients. Moreover, none has exerted a remotely comparable or enduring impact upon discursive practices in the cultivation of style as have Aristotle and Demetrius, and, through them, their unsurpassable models Homer and Demosthenes.

VI.C  Modern studies of style

9. Modern studies under the heading of ‘stylistics’ began to congeal in the early 20th century, when studies of language resonated more with a mentalist psychology than epistemological or ethical philosophy. That ambience featured the ‘affective’, ‘subjective’, and ‘expressive’ aspects of style,9 thus highlighting the audience over the author or orator. This work would hardly find favour in the ‘modern linguistics’, arising shortly after, which sought its identity first by fencing itself off from neighbouring disciplines and later by falling into line with the formalism, physicalism, and behaviourism of ‘unified science’, which colonised psychology and philosophy too (cf. I.`81). So stylistics did not regain full momentum until the 1960s and 1970s, when these ideologies were receding, and when text linguistics and discourse analysis were also gathering momentum (cf. II.120, 135).10

10. Without a secured academic home of its own, stylistics has understandably been a diffuse enterprise. Perhaps we could roughly map it out with a catalogue of prospective ‘theories’ about the nature of style.

10.1.1. ‘Each language as a whole has its own style.’ This theory would suggest that differences among languages have a stylistic value that should emerge when they are compared or contrasted. Yet practical comparisons of the ‘style’ of other languages with English, as in [1635-36], are not common in my data.  

[1635] The German version conforms to German style [so closely] that it is generally taken by German speakers to be a very well written ‘original’. (Coursebook on Translation)11

[1636] Had he clothed his thoughts in the half French style of Horace Walpole, or in the half Latin style of Dr. Johnson, or in the half German jargon of the present day, his genius would have triumphed over all faults of manner. (Life of Addison)

During my student days at the Sorbonne, one teacher decried my ‘German style’ of using French with ‘heavy syntax’, though the real cause was more likely my having read literary monuments in ‘classical French’ so resolutely that I took to speaking it in everyday affairs with the Parisians, who seemed peeved or bewildered.

10.1.2. Comparative studies of language might have prospered but for the compartmentalisations or rivalries among language and literature departments, or the notion in ‘general linguistics’ that ‘each idiom is a closed system’ and so ‘in practice forms a unit of study’ (Saussure).12 Still, ‘comparative stylistics’ has produced fundamental works in continental Europe,13 whereas ‘contrastive linguistics’ ironically emerged under the aegis of formalism even though formal contrasts are the least interesting, and afforded little support to ‘contrastive stylistics’.14

10.1.3. In practice, at all events, stylistic influences across languages are evident among non-native language learners, e.g., Brazilian Portuguese impacting the English of these São Paulo university students, shown with plausible equivalents:15

[1637] Art delays to be ready. [= art takes time to be achieved]

[1637a] Arte demora pra ser pronto.

[1638] Sometimes the marriage was just to the woman become a housewife.

[1638a] Às vezes o casimento era só para a mulher virar dona de casa.

[1639] What about start this modification of thoughts inside your home?

[1639a] Que tal começar essa modificação de pensamentos dentro de sua casa?

Somewhat depressingly, style has hardly figured in the teaching of English as a foreign language, where the preoccupation with ‘rules’ and ‘correctness’ is prone to foreclose aspects of choice and creativity.

10.2. ‘Each speaker or writer has a unique style.’ If the previous theory implied a small number of general styles, this theory implies a huge number of specific styles. Such a theory might underwrite social and discursive inclusion and actualisation, but risks unduly diluting the concept of style by equating it with raw uniqueness.16 Admittedly, in fine detail every speaker or writer makes a unique set of choices. Yet among those choices, only certain ones have significant discursive functions that lend them the status of stylistic indicators, which render a style not just different, but distinctive, as in [1640-41].

[1640] If you write primarily to see, then the style becomes an expression of your personality. […] One’s style must be an outcome of one’s vision of reality. (Hot Press

[1641] Whatever its place in the literary league tables, After Silence is undoubtedly entertaining: Carroll’s style has a high gloss, twinkling with near epigrams (lecture)BNC

And in practice, most recognition of ‘style’ has been given to literary works selected more for use in the acculturation projects of the schools (II.182, 188) than for any explicit consensus about the indicators of their styles. Otherwise, few of us get much recognition of our ‘styles’, least of all non-native learners of English.

10.3. ‘Each text or discourse has a unique style.’ Here, we raise the number of specific styles still higher, acknowledging that every text or discourse is unique in fine detail; even the same one read aloud twice by the same speaker evinces minor differences in articulation or prosody. But if  a language is itself a theory steadily in the dialectical process of being constituted by its practices (I.36; II.21, 157, 178; VI.3), then uniqueness is a routine by-product of text and discourse. Again, a style should be not merely different but distinctive, as invoked in [1642-43].

[1642] The style of the poem is in many ways seventeenth-century, though there are plenty of resemblances to later hymns (Wordsworth)

[1643] The style of the novel [A la recherche du temps perdu] [is] like the structure. […] His long sinuous phrases are designed to enclose the different levels and quirks of reality, just as liquid, spilt on a rough pavement, eventually seeps into every crack and cranny. (Ideas in Action)BNC

Yet here too, what qualities render the style distinctive (e.g. ‘seventeenth-century’ or ‘sinuous’) often remains unspecified in practice (cf. VI.1).

10.4.1. ‘Style is an issue of value or attitude.’ In my data samples from British and American Writers, Ameliorative Collocations like [1644-45] are far more common than for Pejorative ones like [1646-47]. 

[1644] Sechele conducted the prayer in his own simple and beautiful style (Dr Livingstone)

[1645] a more correct and elaborate style distinguished the discourse, or at least the compositions, of the church and palace (Decline)

[1646] Books are filled with trivial content and banal style, to make them ‘easier to read’. (New Scientist)

[1647] A flush suffused her face for such an awkward style of presentation (Wildfell

Attested Adjectives collocating with a good ‘style’ of language included ‘fine, high, glorious, eloquent, grandiloquent, imperious, sublime’; the Nouns collocating with ‘of style’ included ‘clearness, elegance, eloquence, purity, dignity, propriety, energy, tact, graces’. The only Adjectives I found collocating with a bad ‘style’ of language were ‘haughty, mechanical, conventional, vapid, pedantic’ (these last four all in the same sentence); and no Nouns at all.

10.4.2. These proportions suggest a generally Ameliorative view of ‘style’, witness expressions like ‘stylish’ [1648], ‘with style’ [1649], or a ‘style of your own’ [1650].

[1648] Thomas Herbert was at his best reading aloud his own work and delivering stylish penetrating eisteddfodic adjudications in his memorably sonorous voice. (National Bibliography)BNC

[1649] the novel speeds along — firing insults and insights with style. (Best)

[1650] Eduard Bagritsky’s first poems were in imitation of the Acmeists, a literary group of the early 1900s that advocated a concrete, individualistic realism, stressing visual vividness, emotional intensity, and verbal freshness. Before long, however, he began writing in a style of his own (Odessa Web)www

Some data suggest that style can be valued for helping to express ‘thought’ and ‘feeling’ [1651], especially ‘complicated’ ones in ‘simple words’ [1652]. But, like Aristotle’s distrust of ‘mere display’ [1628] (VI.5), some data suggest that a ‘brilliant style’ can also camouflage a lack of ‘knowledge’ or ‘caution’ [1653]. 

[1651] To use words so true and simple that they oppose no obstacle to the flow of thought and feeling from mind to mind […] is the essence of style (John Galsworthy)

[1652] in the lucid style which seemed able to put complicated thought into simple words, musical and measured, he read as he might have read a novel (Human Bondage)

[1653] The work, from its powerful and brilliant style, though displaying little accurate knowledge and a great want of scientific caution, immediately had a very wide circulation. (On the Origin of Species)

10.5. ‘Style is conveyed by linguistic units and patterns.’ This theory seems uncontroversial if not indeed self-evident. Yet only some units and patterns will qualify as stylistic indicators (VI.10.2), i.e.,  the relevant ones for determining the style, such as the Figures of Speech in [1654-65].

[1654] A train horned its way northward. […] The landscape trembled by. (Butcherbird)

[1655] All men were afraid in a world menaced by the invincible spiders of their own anxieties (Harmattan)

Describing the Figures is a challenge I shall take up further on (VI.39-67).

10.6.1. ‘Stylistics is a field inside linguistics.’ This theory should logically follow from the previous one if linguistics is the study of linguistic units and patterns. But the discipline has tended to formulate its methods too narrowly to underwrite a general study of style.

10.6.2. Comments on style in the discourse of the more formal linguists have thus been mainly sporadic and tentative. If ‘style is a technical matter of the building and placing of words’, ‘the major characteristics of style’ are ‘inescapably’ ‘given by the language itself’; ‘these necessary fundamentals’ might ‘point the way to those stylistic developments that most suit the natural bent of the language’, such as its ‘phonetic groundwork’ and its ‘morphological peculiarities’ (Sapir).17 In this theory, the potential style (of the language) strongly pre-conditions the actual style (of the text), which led the same linguist to disqualify stylistic influences between languages in the ‘semi-Latin’ of John Milton and the ‘Teutonic mannerism’ of Thomas Carlyle. Later, formal linguistics severed style from the core of language by declaring ‘the rules of stylistic reordering’ to be ‘peripheral’ matters of ‘performance’, not ‘bearing on the theory of grammatical structure’ (Chomsky).18

10.6.3. By contrast (as usual), functional linguists expansively saw ‘style’ ‘fusing’ the ‘elements of habit, custom, tradition,’ and ‘innovation’ within ‘verbal creation’ (Firth).19  Indeed, ‘there are no regions of language in which style does not reside’; the ‘central problem in the study of style’ is ‘determining whether any particular instance of linguistic prominence’ ‘is significant’ and ‘motivated’ (Halliday).20 A ‘prominent’ ‘feature’ is ‘foregrounded only if it relates to the meaning of the text as whole’; it may function either as ‘a departure from a norm’ or as ‘an attainment or establishment of a norm’.20 The ‘deterministic concept’ of ‘deviation’, which focussed ‘great deal of attention’ on ‘ungrammatical forms’ ‘prohibited by rules’ (VI10.11.1), is ‘of very limited interest in stylistics’.20

10.6.4. Thus, the prospects for integrating stylistics seem brightest for functional linguistics, defining style as a key factor in the dialectical relation between the actual, represented by authentic data, and the potential, represented by Lexicogrammar, Prosody, and Visuality. Aristotelian criteria like ‘familiar’ and ‘strange’ could be clarified by reference to representative corpora, if possible intuitively sorted for their style, such as ‘epic poetry’ (cf. VI.10.8.1ff).

10.7.1. ‘Stylistics is a field inside literary studies.’ This contrasting theory holds the advantage over linguistics of claiming a longer and more prolific tradition in studying the ‘style’ of literary works and authors, who naturally work to achieve an individual style. If the purview of linguistics has been distinctly narrow, that of literary studies has been distinctly broad, deploying numerous and diffuse terms for designating a particular ‘style’:

[1656] In the field of literary writing, […] the term [style] has been applied to the linguistic habits of a particular writer, [or] to the way language is used in a particular genre, period, school of writing, or some combination of these: ‘epistolary style’, ‘early eighteenth-century style’, ‘euphuistic style’, ‘the style of Victorian novels’. (Style in Fiction)21

[1657] it was written in Eliot’s early style: pleasing, pungent and persuasive. (T.S. Eliot)

The really tough challenge awaits in relating these terms to particular linguistic features, e.g., to those rendering a style ‘pungent’ (cf. VI.1).

10.7.2. The uncertain status of ‘style’ indicated by diffuse terms reflects the generally hybrid organisation of literary studies as a wide-ranging project for acculturation (II.197; VI.10.2). Close attention to language and text must share the agenda with the mediation of culture, history, and biography, three domains whose impact on literary works might foster innumerable forms. If literature is a discourse domain wherein the principle of ‘alternativity’ creates alternative worlds to enhance our understanding of our own world (cf. II.175, 182, 191), poetry skillfully applies the same principle to language, whereby the author is all the more motivated to master a distinctive style. In exchange, the relation remains problematic between the style of the literary text and the reality of the author, the more so for authors who recreate themselves as a literary character with a distinctive ‘life style’ too. 

10.7.3. As literature evolves through a vast plurality of styles, the potential for fresh innovation is steadily narrowed, at least insofar as imitation is not highly regarded. The historical irony of English literature lies in the early emergence of its unsurpassed (and unsurpassable) masters of style, casting their daunting shadows upon every later stylist who might try write ‘in the style of’ Shakespeare or Milton.

10.8.1. Style’ distinguishes a literary genre’. This theory seems reasonable for some cases, notably the ‘epic style’, as memorably developed by Milton:

[1658] Of Man’s First Disobedience […]

Sing, Heav’nly Muse, that on the secret top

Of Oreb, or of Sinai, didst inspire

That Shepherd, who first taught the chosen Seed,

In the Beginning how the Heav’ns and Earth

Rose out of Chaos: […] I thence

Invoke thy aid to my adventrous Song,

That with no middle flight intends to soar

Above th’ Aonian Mount, while it pursues

Things unattempted yet in Prose or Rhyme. (Paradise Lost)

In twin series of stylistic indicators, strategic cultural icons oscillate between pagan (‘Muse’, ‘Aonian Mount,’ i.e., Mount Helicon, mythical home of the Muses) and Christian (‘Oreb’ and  ‘Sinai’, mountains where Moses was inspired by God’s miracles), to augur a synthesis of ‘Classical’ and ‘Biblical’ cultural and religious traditions. Further indicators reside in the grammatical complexity with its multiple dependencies among the Clauses, which lends the flow a prosodic solemnity for a discourse proposing no less a mission than to ‘justify the ways of God to men’.

10.8.2. Comparative stylistics would be quick to point out that Milton’s epic style, ostensibly for ‘things unattempted yet’, was anticipated by the style of Virgil’s Aeneid. There, the ‘muse’ was famously invoked to aid ‘singing’ of quite different heroics but with similar stylistic indicators such as cultural icons of historical places and a grammatical complexity in Latin that Milton admired and partially emulated. (All ‘epic’ translations in this section are my own.)

[1659] Arma virumque cano, Troiae qui primus ab oris

Italiam, fato profugus, Laviniaque venit litora […]

Musa, mihi causas memora, quo numine laeso,

quidve dolens, regina deum tot volvere casus

insignem pietate virum, tot adire labores impulerit.

[Arms and the man I sing, who first from the coast of Troy

came to Italy, exiled by fate, and to the shores of Lavinia. […]

Muse, recount me the causes: how offended majesty,

or what angered, the queen of the gods impelled to undergo so many misfortunes

a man distinguished for piety, to endure so many labours.

From there we would acknowledge Homer (both of him), who tersely invoked the ‘muse’ or ‘goddess’ at the start of the Iliad [1661] and the Odyssey [1662].

[1660]  mênin aeide thea Pêlêïadeô Achilêos

oulomenên, hê muri Achaios alge ethêke

[Sing, O goddess, the anger of Achilles, son of Peleus,

that brought countless ills upon the Achaeans]

[1661]  andra moi ennepe, mousa, poluptropon, hos mala polla

planchthê, epei Troiês hieron ptoliethron epersen […]

tôn hamothen ge, thea, thugater Dios, eipe kai hêmin.

[Tell me, muse, of that much-travelled hero who exceedingly far

wandered after he had ravaged the famous city of Troy. […]

From whatsoever source, goddess, daughter of Zeus, tell those things.

Projecting forward again to the 16th century, the epic style of Luis Vaz de Camões betrays the same lineage, but with the pungent irony of bidding the song of the ‘ancient muse’ about ‘great voyages’ to cease and attend to ‘greater merit’.

[1662] As armas e os barões assinalados,

Que da ocidental praia Lusitana,

Por mares nunca de antes navegados

Passaram ainda além da Taprobana […]

E também as memórias gloriosas

Daqueles Reis; […]

Cantando espalharei por toda parte,                                       

Se a tanto me ajudar o engenho e arte.  […]

Cessem do sábio Grego e do Troiano

As navegações grandes que fizeram; […]

Cesse tudo o que a Musa antígua canta,

Que outro valor mais alto se alevanta. (Os Lusíadas)

[The arms and the distinguished barons

That from the western shore of Lusitania [ancient name for Portugal]

Through seas never before navigated

Passed even beyond Taprobana [ancient name for Sri Lanka] […]

And also the glorious memories of those kings […]

Singing I shall spread everywhere

If so much will aid me genius and art. […]

Cease of the wise Greek and the Trojan

The great navigations they made;

Cease all that the ancient muse sings,

For other, higher merit has arisen.]

An ‘interlingual style’ like the ‘epic’ indicates how a comparative stylistics might offset the usual compartmentalisation of ‘literature programmes’ (VI.10.1.2).

10.9.1. ‘Stylistics should be objective, not subjective.’ This theory reflects the general ambitions of ‘classical science’ (in the sense of I.81) reapplied to language studies and linguistics, and periodically to literary studies. Applying it to stylistics as well has encouraged a recourse to statistics;22 yet by substituting quantity for quality, statistics is not well equipped to distinguish which of the counted items are stylistic indicators. When Aristotle stressed the infrequent (‘rare’) choices ‘raised above’ the frequent (‘commonplace’) (VI.2), he was probably thinking of the proportions in general usage. A specific author or text might associate frequency with consistency, and infrequency with creativity, but the style is still not described.

10.9.2. A tougher problem is that stylistic indicators are often not single items but combinations and patterns whose frequency is uniformly low and thus look much the same in statistical terms, such as Shakespeare’s ‘to be imprisoned in the viewless winds’ (Measure for Measure) or ‘whose high upreared and abutting fronts the perilous narrow ocean parts asunder’ (Henry V).  The apparent simplicity and objectivity of numbers hinder them from capturing the complex and subjective phenomenon of style — they’re the smoke, not the fire.

10.9.3. Moreover, statistics discounts the crucial factor of audience response to style.23 In The Two Noble Kinsmen, a play of mixed authorship, my intuitive response is that [1663] (pleading for the Duke to give the ‘slain kings’ a proper burial, a ‘grace’ even granted to suicides) is by Shakespeare, whilst [1664] (pleading for the audience not to ‘hiss’ the play, whose ‘story’ came from Chaucer) is definitely not.

[1663] Remember that your fame

Knolls in the ear o’ the world: […] your actions,

Soon as they move, as ospreys do the fish,

Subdue before they touch: think, dear Duke, think

What beds our slain Kings have. […]

Those that with cords’, knives’, drams’ precipitance

Weary of this world’s light, have to themselves

Been death’s most horrid agents, humane grace

Affords them dust and shadow. (I, 1, 133-34, 137-38, 141-44)

[1664] Chaucer (of all admir’d) the Story gives,

There constant to Eternity it lives.

If we let fall the Nobleness of this,

And the first sound this child hear, be a hiss,

How will it shake the bones of that good man,

And make him cry from under ground, ‘O fan

From me the witless chaff of such a writer…’ (Prologue, 13-19)

My response to both samples is based on prosodic flow, and on lexicogrammatical combinations which are statistically insignificant in my complete corpus of Shakespeare plays, but which are creative in the one sample (e.g. ‘knolls in the ear’) and trite in the other (e.g. ‘hear a hiss’).

10.10.1. ‘Stylistics should be descriptive rather than evaluative’. This theory is a less ambitious correlate of the preceding call for objectivity, and reflects the aspiration of ‘general linguistics’ to establish a science by opposing its descriptive stance to the more traditional evaluative stances (cf. II.38). Yet the close correlate of favouring the description of the most general or even universal aspects of language is hardly conducive to studies of style.

10.10.2. Within the ambience of formal linguistics, descriptive stylistics has predictably been concerned with measuring the proportions of word-classes or clause types, the length or syntactic complexity of clauses and sentences, and, as already noted, the statistical frequencies of individual words. Such research could (and did) confirm — small surprise — that Edward Gibbon’s style is far more complex than Ernest Hemingway’s,24 but could not explain why I find Gibbon’s prose stimulating and Hemingway’s tiresome (perhaps because as a teenager I was reading the polished histories of Tacitus, Suetonius, and James Henry Breasted). Nor could it explain how I single out ‘authentic’ Shakespeare inside a play of mixed authorship [1663-64]. Such explanations presuppose being able to show how descriptive categories like ‘frequent’ and ‘infrequent’, or ‘simple’ and ‘complex’ influence how audiences actually evaluate styles. Evidently, I value the infrequent and complex aspects of style more than do the devotees and epigones of Mr Hemingway, who might find them insufficiently macho.

10.11.1. ‘Style is the interplay between norm and deviation’.25 This theory was also inspired by formal linguistics, though it may sound reminiscent of Aristotle’s counsel that ‘by deviating in exceptional cases from the normal idiom, the language will gain distinction’ (Poetics, 9,4). The analogy from linguistics seems to have been the interplay between ‘grammatical’ versus ‘ungrammatical’ (cf. VI.10.6.4); yet the latter term designates utterances which simply don’t occur, such as ‘sincerity may virtue the boy’ (II.78), and which are accordingly uninformative about issues of style.

10.11.2. Again, the reticence of linguistics about evaluation incurs problems. Obviously, various modes of deviance may have quite disparate stylistic value (or none), nor can value simply increase along with degrees of deviance. Compare:26

[1665] It dropped so low — in my Regard —

I heard it hit the Ground —

And go to pieces on the Stones

At bottom of my Mind —

Yet blamed the fate that flung it ­less

Than I denounced myself

For entertaining Plated Wares

Upon my Silver Shelf. (Emily Dickinson)

[1666] All the sun long it was running, it was lovely the hay

Fields high as the house, the tunes from the chimneys, it was air

And playing, lovely and watery

And the fire green as grass. (Dylan Thomas, Fern Hill)

[1667] my father moved through dooms of love

through sames of am through haves of give,

singing each morning out of each night

my father moved through depths of height (e.e. cummings)

[1665] is normal in Grammar and mildly deviant in the Lexicon, making abstractions concrete in a self-reproach about an undeservedly noble (‘silver’) esteem for something superficial (‘plated’) — what it was, we are free to imagine. [1666] deviates mildly in Grammar:  perhaps ‘sun’ for ‘sunny day’, ‘it’ for ‘childhood activities’; ‘green fire’ for the glow of ‘watery’ dew on the ‘grass’. [1667] deviates most strenuously, but rather gratuitously, like a riddle to be unscrambled to reveal humdrum content, e.g.: ‘my father’s fate was to be loved throughout a stable and generous lifetime; greeting each morning after each night, my father lived to make depths seem like heights’. But what makes me personally hold Dylan Thomas to be the foremost English poet of the 20th century is not any list of deviations but the total effect of the Lexicogrammar, the sonorous Prosody of his recorded recitations, and the measured Visuality of the lines and stanzas.

10.11.3. By contrast, a passage like [1668] doesn’t seem deviant, but would in respect to the norms of poetry as invoked by other lines in the same text like [1669].

[1668] My nerves are bad to-night. Yes, bad. Stay with me. Speak to me.

(T.S. Eliot, The Wasteland)

[1669] The river’s tent is broken: the last fingers of leaf

Clutch and sink into the wet bank. The wind

Crosses the brown land, unheard. (same)

Far more deviant, in my view, is the text’s literal and peremptory expropriation from other texts, some in foreign languages; a more apt title for The Wasteland might have been The Jumble Sale or Hypocrite Auteur.

10.11.4. Extreme deviance may produce a style too strenuous to bear, viz.:

[1670] Many mellow Cydonian suckets

Sweet apples, anthosmial, divine

From ruby-rimmed beryline buckets

Star-gemmed, lily-shaped, hyaline

Like the sweet golden goblet found growing

On the wild emerald cucumber-tree

Rich, brilliant, like chrysoprase glowing,

Was my beautiful Rosalie Lee (Thomas Chivers)

This deservedly obscure 19th-century US poet might be suspected of audacious parody, if his (unpublished) volumes did not bear such languishing titles as Phials of Amber Full of the Tears of Love.27  Rosalie Lee — fittingly rhymes with ‘fiddle-dee-dee’ — or some parts of her, reminded the impassioned speaker of Lebanese pastries (‘Cydonian suckets’ =>  Sidonian sweetmeats, i.e., from Sidon, a city in ancient Phoenicia, now Lebanon) and ‘sweet apples’ by the bucketful — and what opulent ‘buckets’! Made of emeralds (‘beryline’), ‘rimmed’ with ‘rubies’, ‘gemmed’ with ‘stars’, ‘shaped’ like ‘lilies’ (sloppy to pour, though), ‘glowing’ like ‘chrysoprase’… You see my point.  She was a regular pail-faced Tiffany’s.

10.12. ‘Style results from the interplay of unmarked and marked’. This theory is a functionalist correlate of the theory of deviance, but with an incisive difference. Marked choices need not be deviant nor infrequent; they are chosen for strategic motives, whereas unmarked ones are chosen when no such motive applies (VI.14). In [1671], both the Direct and Indirect Objects of the Imperative ‘bring’ are Fronted in marked positions; and ‘shadows brown’ might be something more to ‘bring’, but is more coherent as another melancholy destination for the sun’s ‘beams’.

[1671] His flaring beams, me, Goddess, bring

To arched walks of twilight groves,

And shadows brown, that Sylvan loves

Of pine, or monumental oak (Il Penseroso)

Whereas ‘deviance’ suggests an event violating or falling outside the system, ‘markedness’ suggests an event refreshing or flexing the system.

10.13.1. ‘Style should be explicitly developed’. Here at last, theory is resolutely allied with practice; but so far, the practices are running ahead of a creditable theory. For example, some advice on ‘style’ tells you to avoid repeating the same words, whereas other advice warns against what H.W. Fowler, beefeater guardian of The King’s English, famously spurned as ‘elegant variation’, which ‘includes all substitutions of one word for another for the sake of variety’. Neither advice will do for all cases: for me [1672] is better style with repetitions, and [1673] without them.

[1672] It seems that the woman was a young woman, and a jealous woman, and a revengeful woman; revengeful to the last degree. (Great Expectations)

[1672a] It seems that the woman was a young maiden, and a jealous lady, and a revengeful female; retaliatory to the last degree.