Perusing some of his essays, I’ve been lately thinking how unfortunate it is that Robert Louis Stevenson’s stock as a writer fell so dramatically sometime after World War I; once a literary sensation, he went from a world-renowned author to a rough and ready reputation as a scribbler of what were barely more than pulp adventure stories.
It’s nevertheless my view that he should be taken as seriously as Conrad (whom he greatly inspired) as a purveyor of English language prose. The reason I suspect he isn’t is because, whilst Conrad was admired by the Modernists, Stevenson wasn’t: Virginia Woolf wrote of Stevenson that he was “no critical artist” and guilty of “self-conscious […] sentimentality […and] quaintness.” Even prior to that, Leonard Bast in E.M. Forster’s Howards End (1910) is shown to have embarrassingly immature taste by enthusing (as I am about to do) over Stevenson’s essays. It seems strange now that there was once a time when the critics of Stevenson, like George Moore, were not only rare but based their attacks on Stevenson’s essentially being too great a stylist, and thus straying too far from the realist movement. It was in those days that Rudyard Kipling found in Stevenson’s work “the most delicate inlay-work in black and white”, and Oscar Wilde called him the “delightful master of delicate and fanciful prose.”
Fortunately, there has been some upturn in the man’s critical fortunes, thanks at least to three of my favourite writers in Graham Greene, Jorge Luis Borges and Italo Calvino, all of whom praised his fiction to varying degrees. (Greene seemed merely to like him, whereas Borges flat out worshipped the man.) He was, in addition, a fine critic and engaging essayist, and I have included below his essay on style because it elucidates much I’ve been trying unsuccessfully to articulate in the past.
It’s a rare thing to find literary criticism that engages with prose; far more popular subjects are the “themes” or perceived politics. Upon studying literature, an undergraduate is far more likely to encounter feminist or post-structuralist analytical models than commentary upon Milton’s economy of vowel/consonant interplay, say, or even the rhythm of Shakespeare’s blank verse between the metrics. It wasn’t until last year, when I read William H. Gass on Henry James, Charles Dickens and William Gaddis (stylistic essays collected in the rather excellent Temple of Texts, available from Dalkey Archive Press), that I had even thought about the use of, for example, alliteration by a novelist; and this is after years of both formal and casual immersion in literature and criticism. Actually hearing Gass reading his 1995 novel The Tunnel on audiobook was similarly formative in this regard, and knowing this is a writer who agonises over his sentences made it easier to understand why he took over 20 years to finish writing that particular book.
Let me be clear that I’m not of the Harold Bloom mindset, attempting to rescue literature from the “school of resentment”. Literature is like a diamond with many facets, that one needs a full drawer of looking glasses to properly appreciate. It’s no bad thing to excavate meaning with a Marxian trowel or to run a feminist needle pick along its contours; one can in this manner learn much about the concealed preconceptions of the society and artist, and reciprocally employ the text to illustrate Marxism and feminism to a student afterwards. Little, however, information about the aesthetics of the piece is gained with those particular tools – imposing a framework from without can only bring forth so much from within.
The formidable Northrop Frye, in his theory of modes (An Anatomy of Criticism, a book nobody serious about literary criticism can avoid) put these popular schools to one side to engage with the text on its own terms. What resulted was a pattern-spotting exercise hoping to unveil the formula great works follow. It is more this sort of from-the-inside-out approach rather than a top-down imposition I would advocate to be applied on the specific level of sentence construction and prose rhythm, though Frye’s subject was a more general overview of literary “genres” or “modes”, from the epic to the romance, on the basis of whether the protagonist is superior or inferior to us in “kind” or “degree”.
A brief diversion on this. In her essay Historical Genres/Theoretical Genres, which addresses Tzvetan Todorov’s commentary on Frye’s Anatomy and Stanislaw Lem’s critique of Todorov, Christine Brooke-Rose provides Frye’s classifications in the form of a table:
She notes the table is incomplete as more combinations are possible than are addressed by Frye, further evidence that Frye was a more excellent critic than he was a theorist. Perhaps this distinction is explanation why Marxism and feminism, say, still seem predominant: it is they who still reign supreme on the level of theory, and our non-ideological interests are less antagonistic reclamation than equally protagonistic operations on the level of criticism. The attempted apotheosis of Frye’s criticism to the Olympus of theory is unsuccessful because it has transferred from “mere” observation into hagiography of another framework. It’s all well and good to discover that Oedipus is high mimetic tragedy as opposed to the romantic tragedy of Beowulf, but a problem may arise when shoehorning Kubrick’s Barry Lyndon into a low mimetic comedy category in order to illustrate the theory: we have gone from descriptive to prescriptive without utilising any viable judgement of quality, a judgement that Frye is unwilling to and uninterested in making.
There are at least as many approaches to quality as there are to thematic recognition or genre qualification. The critic James Wood is said to frequently take a conservative, moralist approach, chastising Flaubert for his emotional distance from the cruelty he portrays. Book reviewers often laud or criticise a work on the basis of plot, which is roughly as useful as judging a coat by its coathanger. A judgment on “character development” in works that are not for instance bildungsroman or kunstlerroman, i.e. where that is not the focus, is likewise unappealing. The quality of style, whilst not the only consistently credible discernment of value one could make, is perhaps the most universal arbiter between good and great works. It’s prescription without imposition, leaving agenda at the door.
The first element of style Stevenson identifies is the choice of words. He compares words to building blocks, and notes that it is these that are the most immediately noticeable element, “this form of merit is without doubt the most sensible and seizing”. This is then analogous to noticing the architect has chosen to use marble and mosaic in the interior of Westminster Cathedral, and judging the quality of the marble and ordered composition of the mosaic, before one deduces the Neo-Byzantine design ethic.
A good choice and ordering of words is not in itself good writing – according to Stevenson, Cicero’s superiority to Tacitus and Voltaire’s to Montaigne cannot be thus explained – but it is certainly a component. Here Stevenson in part anticipates the famous dispute between Faulkner and Hemingway, a story that is more often than not presented as the retort of the wise and grounded Hemingway putting the “poor” highfalutin Faulkner, who thought “big emotions come from big words”, in his place. This anti-intellectual strain is also to be found in Orwell’s popular insistence that a writer “never use a long word where a short one will do”. Stevenson would recognise that this is missing the point – the length or size of the word is not as important as the word’s “harmonious” sound (Stevenson on a few occasions mentions an internal “ear” as being key) in the context of the words around it.
Though it’s nothing than the other side of the same coin as Orwell and Hemingway, I do have some sympathy with Will Self’s argument that using obscure words can rescue a text from banality and being sent to a dictionary, as Faulkner said Hemingway’s readers never were, is not necessarily a bad thing – surely a strong vocabulary is better than a weak one? However, it’s not hard to infer from that same article that Self apparently believes lexical decisions are all that separate Joyce, Nabokov and Woolf from the middlebrow, and a less kind reviewer might look to this lack of understanding for explanation as to why Self will never reach that level of the stylists he clearly aspires to.
Contra Orwell, I would also say that in philosophical writing an obfuscatory tone is not necessarily a bad thing. Perhaps this is because of my preference for continental and poststructural thinkers over the members of Anglo-Saxon, analytical schools who would recoil in horror at the Sokal-tempting language of a Derridean. But complex ideas deserve to be treated with complexity – that one cannot render deconstruction easily into immediately accessible English isn’t a weakness of the idea or its conveyor, and might actually be part of the point. And if one does misread, this in itself can be fertile soil for one’s own ideas.
The next section, which Stevenson entitles “the Web”, begins with an interesting commentary on the arts that may no longer ring exactly true. Essentially, Stevenson identifies a diegetic/mimetic divide (if I may be allowed to stretch the definition of diegesis a moment), with architecture, music and dance on one side and sculpture, painting and acting on the other. The diegetic set are the thing-in-itself, whereas the mimetic are instances of art imitating life. Since Stevenson was writing, the rise of abstraction has meant that sculpture and painting have been able to cross into the non-mimetic realm, joining literature, opera, ballet and, now, film as transaxial arts that can belong to either category. It could be said that some sonic experimentation by guitarists like Steve Vai, who can make their instruments sound like pigs or cats, has meant that music has shown the potential to be mimetic; though whether this is musicianship or onanism I will leave to the discerning judgement of the reader.
None of this is really relevant to Stevenson’s point – he simply includes it as a preface to what these disparate arts nevertheless have in common: a unifying pattern. The Web, if you like. Like Frye’s identification of modes, Stevenson’s identification of the Web is an exercise in pattern recognition – just not on a cross-textual basis. Literature is, with music, one of the two “temporal arts”, composed of sounds and pauses in time. Therefore, these sounds and pauses cannot, according to Stevenson, be fragmentary in the manner of verbal communication but must be woven into a coherent structure.
Of course, a lot has happened in literature and music since Stevenson died in 1894, before the effects of chromaticism had begun to be felt in the music of Schoenberg and the Viennese School. What followed was works characterised by atonality and dissonance, and later the avant garde, free jazz and electronic movements in which discord is more notable than harmony.
Does this mean we would have to reverse Stevenson’s criteria to judge Webern, Stockhausen, John Cage, Albert Ayler, Anthony Braxton, Captain Beefheart and Throbbing Gristle as the virtuosos they deserve to be known as? Whilst these artists lack an explicate pattern, all play to an implicate structure. Braxton’s non-compositional material is characterised as improvisational; however, if Braxton’s playing was, as it probably seems to the untrained ear, pure improvisation, it would indeed be analogous to everyday speech, Stevenson’s communication in the “business of life”, made in “broken words”, “carried on with substantives alone” and therefore “not what we call literature” (or, in this case, music). Of course, Braxton is not making it up as he goes along, but varying, deconstructing and playing against the themes of his predecessors, in the way that Schoenberg began by riffing (for want of a better word) on Wagner and the Romantics.
Even Braxton’s later music, which sounds like this:
cannot be dismissed. For whilst it lacks a pattern, it is not a random primordial soup in the way of everyday conversation: having rejected traditional Western forms, Braxton now builds towards new possible forms, saying of his Ghost Trance Music that it is part of “a process that is both composition and improvisation, a form of meditation that establishes ritual and symbolic connections (which) go beyond time parameters and become a state of being in the same way as the trance musics of ancient West Africa and Persia.” Convergent evolutionary parallels might be found in the Sufi devotional chanting of Nusrat Fateh Ali Khan or in the music of the Far East where, like Miles Davis or Morton Feldman, you play the notes that aren’t there.
The reason I include this is to remind us that just because something does not fit our tried and tested models and frameworks does not mean it lacks pattern. There is still a Web, just not a Web we would immediately recognise and easily contexualise within our cultural dialectic. Braxton in form might be seen as analogous to a science fiction writer in content, extrapolating beyond our temporal plane to see where our current technologies might take us. Within this analogy, Throbbing Gristle and other industrial musicians are alternate history scribes: like Philip K. Dick in The Man in the High Castle asked what would have happened had the Axis powers won the Second World War and Poul Anderson speculates on a contemporary reality where Rome lost the Punic Wars in Delenda Est, Throbbing Gristle attempted to project a world where blues was not at the root of popular music. Just because form and style is harder to recognise than content and theme does not make experimentation in one less valid than the other. Is it different, and difficult? Yes. Is it inferior, and wrong? No.
Our literary equivalents are likewise some of the greatest artists of the 20th century. Queneau, like Schoenberg always conscious of the rules even as he made up and rigidly adhered to his own, attempted to reconcile the improvisational manner of French speech with the stifling formality of it’s written grammar. Céline, like Queneau, used ellipses often, and thus his prose seems more characterised by dissonance than unity but nevertheless still clearly by rhythm. Burroughs, a fan of Céline, attempted to escape the temporal conventions of prose that Stevenson mentions by using the cut-up technique, evading what he saw as the “tyranny” of language and narrative to, like Braxton, allow “subliminal hints of the future to seep through” (according to Alan Moore, an idea upon which I have written before).
An aside: David Foster Wallace, a good writer who doesn’t quite merit inclusion alongside these titans, also experimented with breaking up prose, using footnotes rather than ellipses or cut-up.
Joyce went further than Braxton – rather than speculating or ruminating, he went ahead and crafted his own language. What’s extraordinary is there is still a great music to Finnegans Wake when read aloud – unlike Céline or Burroughs, it could not be said to be consciously atonal. Joyce was reputably a majestic singer as a young boy, which reminds us that language is born not in the mouth but in the lungs and throat, and that internal ear Stevenson speaks of must be employed to decipher its depths. Plato thought the written word an aberrant mutation of the spoken, whereas Derrida would have said the spoken into the written was natural state of change in the manner of a seed becoming a tree. It is in Joyce and other “musical” writers that we see there is no distinct and permanent divide between the two but a constant interplay. The phenomenon of reading literature in one’s head and never out loud is a relatively recent one, which may well have forced this artificial barrier that stops us from “hearing” the sonorous nature of great prose, or even from describing prose as “sonorous” at all.
Finally, there is Beckett; our Feldman, only better; the concave to Joyce’s convex; the returner of style to the corruption of the soil.
Would Stevenson have recognised these prose divergences from the Web as legitimate? Certainly he was able to sense the legitimacy of Whitman, who just as sure diverged from metre in his verse. In a different essay, Stevenson said that Leaves of Grass, “tumbled the world upside down for me, blew into space a thousand cobwebs of genteel and ethical illusion, and, having thus shaken my tabernacle of lies, set me back again upon a strong foundation of all the original and manly virtues […] New truth is only useful to supplement the old; rough truth is only wanted to expand, not to destroy, our civil and often elegant conventions.” Stevenson was, of course, addressing the morality of Leaves of Grass rather than its scansion, but he would be equally correct if speaking of prosody that seems to break the rules but actually pushes the boundaries beyond our immediate conceptions.
On sentence construction, Stevenson further defines the Web as a balance between pattern and logic, between fabric and argument. In order for the pattern to be strong, the argument must have some of the following criteria: brevity, charm, clearness. The weave of the pattern, though, should have a “knot”. It’s not entirely clear what Stevenson means by “knot”, though he uses it several times throughout the essay, and he gives no examples. I’m tempted to suspect it would be to the prose sentence what the caesura is to the verse line, but it’s more likely that Stevenson imagines the knot to be a word rather than a pause. Perhaps it is a hint of atonality within the sentence, a moment of dissonance – a clause that begins with “or” or “though”. Following the knot, Stevenson calls for suspension before solution, in the manner of a Petrarchan sonnet that poses a problem in the octave before resolving it in the sestet.
Let’s have a look at one of Stevenson’s own sentences, chosen at random (such is my confidence in Stevenson’s economy of language) from Kidnapped:
“She was still holding together; but whether or not they had yet launched the boat, I was too far off and too low down to see.”
The start of the second phrase here would surely be the knot, the rest of the phrase (up to “boat”) the suspension, before the sentence finally clears its meaning.
The sentence also holds up to the alliterative interplay Stevenson recommends later in the essay. There’s three pacing “s” sounds in the first three words, two “h” sounds in the next two, and “holding” is straddled by two equally spaced “t”s. There’s an almost discernibly prosodic structure to the middle of the sentence, too: the words that end in “t” – “but”, “not”, “yet” and “boat” – provide a baseline (please note this is asyllabic and anaccentual scansion):
but | whether or | not they had | yet launched the | boat
One two three, one two three, one two three, one.
This segues into the end of the sentence, where the “t”s are still important but the pattern is built around “too far off” and “too low down”:
I was | too far off | and | too low down | to see
One two, one two three, one, one two three, one two. The pattern now has a syllabic and perfect symmetry.
If this seems contrived and unlikely, it’s testament to how the sentence flows in perfectly natural progression. “There is nothing more disenchanting,” says Stevenson, “to man than to be shown the springs and mechanism of any art.” Let’s see if we can redraw the sentence in bad prose, or at least prose that is dull and flat, and still maintain meaning and grammar.
“I couldn’t really tell if the boat had been launched because I wasn’t close enough and not very high up, but at least it wasn’t in bits yet.”
This is, of course, artificially constructed for a purpose, but the point remains that so too is the original. Comparison between them should emphasise how the “ear” can immediately discern between inspired and uninspired writing, even in a relatively uninteresting sentence that only conveys a surface meaning. The breakdown provides us with why this is, and how it is accomplished.
It has long been my contention that an ear for prose is best prepared by an ear for verse, and Stevenson as a critic is unusually well equipped for this purpose, able to discern amphibrachys and amphimacers as easily as iambs and trochees. So, too, should be the stylist, and I would always recommend a training in verse for the writer who wishes to master good wordsmithing. I concur, however, with Stevenson when he recommends prose be rhythmical but not metrical. You will notice that the sentence quoted above can not be broken down into metre, and you’d be hard-pressed to find many sentences in Stevenson that can. A single instance of metrically scansioned English is acceptable, but another immediately succeeding promotes “an instant impression of poverty, flatness, and disenchantment”; further, it’s the responsibility of the prose writer to expand his scale – he is “condemned” to venture beyond the safe confines of verse! In a 21st century culture where even most “poets” couldn’t tell you the difference between a pentameter and a perambulator, it’s refreshing to read suggestion that the novelist might hide in prosody, fearful of the potential foisted upon him by prose. It points the way to a knowledge paradigm we should reclaim for the sake of taking pride in our art.
I don’t feel I’ve exhausted Stevenson’s essay, but I hope I’ve been able to use it to explicate some of the quintessences of craft. Whilst it’s true that style isn’t everything to literature, no more than brushstrokes are to painting or notes are to music, it is nevertheless as close to fundamental as either. Good prose is to a novel what good musicianship is to a song, but for whatever reason less and less focus seems to be placed upon it. Could this explain why none of our English playwrights seem as grand as Marlowe or Jonson? Why our poets cannot compete with Milton or Spencer? Why our essayists pale in comparison to Thomas Browne and our critics are shadows of Johnson and Hazlitt?
We complain regularly that contemporary art is all concept and no craft – could it be contended English literature suffers the same affliction? So dissatisfied are we with work being produced in our mother tongue that all major literary events these days seem to be translations, from Bolano last decade to Krasznahorkai this, and because we can’t read them in their original language we can only hope to divine scraps about their style through the translator’s glass.
This is an inevitable symptom of Weltliteratur without polyglotism; the solution, though, is not to be found in learning Spanish and Hungarian but to expand our style on our language’s own terms. Stevenson, who could speak French, notes that whilst more effort was being put into English prose in his time, French prose was still superior. Contemporaneous to Stevenson’s essay were mainly his fellow genre writers like H.G. Wells, H. Rider Haggard, Arthur Conan Doyle and Bram Stoker, few of whom possessed his talent. The acknowledged master Henry James, American-born as Mark Twain, would not consider himself British until long after Stevenson’s death, whilst the admired Anthony Trollope and George Eliot had recently passed away. This left the humourist Jerome K. Jerome and Thomas Hardy flying the Union flag, whilst under Le Tricolore worked Victor Hugo, Emile Zola and, of course, still the peerless figure of Flaubert. (Stevenson might also have had on his mind Dumas; though he died in 1870 and cannot exactly be called a living contemporary, Dumas and especially the character of D’Artagnan left a larger and more lasting impression on Stevenson than anyone outside of Shakespeare.)
It is unfair, then, to say of our island’s prose that it is any less a force on the world stage than it was in the 1880s and 90s. Nor am I advocating that we try to imitate the authors from time’s past – we already see that in popular middle class authors like Martin Amis and Ian McEwan. I concur with Gabriel Josipovici in Whatever Happened to Modernism? when he says:
“No composer would dream of writing like Tchaikovsky today, except in an ironic manner; no painter today would dream of painting like Sargent, except in an ironic manner; yet novelists writing in English seemed to want to write like the Victorians and the Edwardians.”
An awareness of style, I hope I have shown in this essay, should go hand in hand with experimentation and pushing the novel forwards.
There is, in fairness, a wealth of good literature in the language being produced worldwide that wasn’t there in the late 19th century, from the former Commonwealth and beyond. Should we, however, be complacent where the chairman of the Booker judges can confidently proclaim Hilary Mantel, a solid but unspectacular stylist, “the greatest modern English prose writer”? True or not, Peter Stothard’s comment should precipitate a call to arms among those of us feeling the draught of discontent.