In Jill Paton Walsh’s novel Goldengrove (1972), set shortly after the Second World War, the adolescent heroine, Madge, goes on holiday to Cornwall. She falls a little in love with a professor of English literature who has been blinded in action, and reads aloud to him. One of the passages he has her recite is an analysis of Donne’s ‘A Valediction: Of Weeping’:
Weep me not dead means: ‘do not make me cry myself to death; do not kill me with the sight of your tears; do not cry for me as for a man already dead, when, in fact, I am in your arms,’ and, with a different sort of feeling, ‘do not exert your power over the sea so as to make it drown me by sympathetic magic’; there is a conscious neatness in the ingenuity of the phrasing, perhaps because the same idea is being repeated, which brings out the change of tone in this verse.
Madge is reading from William Empson’s first book, Seven Types of Ambiguity (1930), which she refers to only as ‘the horrible ambiguity book’. Paton Walsh read English at Oxford in the 1950s, and Goldengrove is in many ways a response to the experience. Madge wants to be a character in To the Lighthouse, or a saintly Jane Eyre reading to a blind Mr Rochester. But instead of becoming part of a tradition of English writing by women she ends up uncomprehendingly reading an intricate analysis of a love poem to an emotionally predatory blind professor.
For Oxford English undergraduates of Paton Walsh’s generation, Empson’s close verbal analysis was the sexy but scarily cerebral alternative to the tweedy historical scholarship of the English Faculty. Both my parents also read English at Oxford in the 1950s, and for them Empson was a byword for brilliance – though their admiration was slightly dented when they witnessed him crawling around on the floor dead drunk during a poetry reading. They sang a song in an end-of-term revue that went: ‘There are seven types of masculinity/Whether you come from Keble or Trinity,’ which provided a perfect rhyme with ‘virginity’. Like a lot of in-jokes, this one looks claustrophobic in retrospect. Is it only types of masculinity that vary? And can every flavour of masculinity really be found in Oxford colleges? But it’s hard to think of any recent work of literary criticism that would be well enough known to be the subject of an undergraduate skit.
Empson was himself something of an entertainer. ‘One might be sentimentally playful like this before killing a young pig,’ he says when Satan gloats over Adam and Eve in Paradise Lost. When Satan is squat like a toad by the sleeping Eve he remarks: ‘Probably he made a very fine toad.’ This freedom of register was partly a result of his squirearchical background: he was a descendant of Richard Empson, a loathed minister of Henry VII, and grew up in a manor house in Yorkshire. He read maths and then English in the Cambridge of Wittgenstein and Bertrand Russell, but the biggest influence on him was his supervisor, I.A. Richards, who advocated what he called ‘practical criticism’, or the close analysis of passages of anonymised writing. An ideal reading of an ideal literary text would involve an ‘organised response’, in which different stimuli and reactions to them would be combined into a complex whole. Literature and literary criticism, he believed, could tense the mind into a counterpoise of sense against feeling, or sense against sense.
Empson was famously chucked out of Magdalene College, Cambridge, when condoms were found in his room. He spent the early part of his academic career teaching in Japan and China. He was a staggering drinker and a wild eccentric in his social manner, as well as in his disorderly mandarin-style beard (Geoffrey Hill was apparently reminded of Empson when he saw a prize-winning Yorkshire terrier). At Richards’s funeral he read a poem by Gerard Manley Hopkins. The organiser of the event, Richard Luckett, described him as speaking ‘inaudibly and inexactly against his own brilliantly simulated atmospherics of squeaks and high-pitched whistles modulated through a slightly damp moustache’. Empson later explained that in order to be sure he was audible he had removed his teeth before reading.
That rudimentary error about how best to get people to attend to what you are saying was not untypical of Empson. But he was – bear with me here – very good at metaphorically taking out his false teeth in order to make himself understood. His critical writing combines forensic analysis of alternative senses (in which his critical teeth grind a text into fine particles) with deliberately wide and vague gestures to the beyond (teeth out, evocative mumbling through a slightly damp moustache). Empson would quite often (and ‘quite’ is one of the function words about which he writes particularly well) grind through a list of alternative interpretations, set out with pseudo-mathematical precision, and then gesture off into the void of the unknowable with a ‘sort of’ to suggest that none of his carefully listed alternative interpretations could quite capture the overall effect of a given phrase. He does this in the passage which poor Madge had to read out to her blind professor when he interrupts his list of all the things Donne might mean in order to talk of ‘a different sort of feeling’. As he said in Some Versions of Pastoral, ‘probably a half-magical idea is the quickest way to the truth.’
Those flashes of strategic vagueness are vital elements in Empson’s style. They encourage his readers to believe that literary texts can take them beyond the limits of their own perceptions, and that, although generating lists of variant senses is one aspect of reading, jumping across a void is what it’s really all about. Empson described his own practice when he said Pope’s Essay in Criticism implied ‘that all a critic can do is to suggest a hierarchy with inadequate language; that to do it so well with such very inadequate language is to offer a kind of diagram of how it must always be done’. This can certainly generate frustrations, since he was quite capable of creating an interminable taxonomy of interpretative possibilities and then throwing it up in the air as inadequate in a way that would drive a philosopher nuts. He could even do that with entire books. The Structure of Complex Words (1951) concludes with the sentence: ‘All I should claim for this chapter is that it gives a sort of final canter round the field’ – as though he is no more than a stable lad giving the horses a spin. But he was among other things a master of the critical blur. As he put it in an essay on Paradise Lost, ‘it is a delicate piece of brushwork such as seems blurred until you step back.’
It’s hard to say that there is a typical Empson essay, since he wrote about so many things. His posthumous publications include books on images of the Buddha and on the role of the censor in shaping Marlowe’s Dr Faustus – a wildly entertaining account ruined by Empson’s conviction that the haphazard processes of Elizabethan censorship resembled those of 20th-century totalitarianism, which they did not. He also wrote about Alice in Wonderland, Marvell’s relationship with his housekeeper, Wordsworth, and Shakespeare, and made parenthetical references to more or less everything else. Reading an Empson essay is like being taken for a drive by an eccentric uncle in a terrifyingly powerful old banger. There are disturbing stains on the upholstery and an alarming whiff of whisky in the air, but when he takes another swig from his hipflask and guns the accelerator, your head gets thrown back so far that you just have to make yourself enjoy the ride – even if you’re not quite sure you’re going where you want to go.
This is particularly true of the essays gathered in Some Versions of Pastoral, Empson’s second book. When it appeared in 1935 many reviewers bleated that its sheep count was far too low for a book supposedly about pastoral. They had a point. Empson used ‘pastoral’ as a broad term for literary works which put ‘the complex into the simple’. He (mainly) meant by this works which explored socially sophisticated attitudes through apparently unsophisticated characters, but this conveniently loose definition of pastoral allowed him to make a book out of a series of essays about Donne, The Beggar’s Opera and the Alice books. Double plots, in which one group of people were thematically connected with another in a subplot, were also ‘pastoral’, because a plot that’s echoed in a subplot implicitly suggests that different social groups replicate or parody aspects of one another. The concern in metaphysical poetry with relationships between the ‘one and the many’ was ‘pastoral’ too, according to Empson, since here a single instance could stand for a range of examples and so bring the complexity of the whole into the single simple thing.
Through this unlikely patchwork of texts, Some Versions of Pastoral presented an alternative to Richards’s conception of an ‘organised response’. Rather than focusing on the equal and opposite impulses elicited from individual readers of a poem, Empson thought of literature as using the resources of language to operate across social boundaries: a socially inferior person or setting could encapsulate truths about social superiors, or about the structures of society more generally – and irony, or ideally double irony, would allow all those elements to coexist. Plurality was the key concept in his critical thinking, and it was a kind of plurality that allowed for a range of different voices and attitudes to exist within a single society, a single text, a single mind, or a single word. ‘Once you break into the godlike unity of the appreciator you find a microcosm of which the theatre is the macrocosm,’ he wrote. ‘The mind is complex and ill-connected like an audience, and it is surprising in the one case as the other that a sort of unity can be produced by a play.’
That is very like Richards in its emphasis on the convergence of different kinds of reaction in a single literary response, but it is crucially distinct from Richards both in its overt vagueness of phrasing (‘a sort of unity’) and in its willingness to accept that the mind can be as ‘ill-connected’ as members of an audience, rather than being an ‘organised’ regiment of counterpoised attitudes. In this way, as Seamus Perry says in his admirable introduction to the grand, fully annotated Oxford edition of this very miscellaneous book, ‘the poet’s inward conflicts, explored so dazzlingly in Ambiguity, are cast as versions of the conflicts that shape the culture of which the poet is a part (but not simply a product).’ That is, in Some Versions of Pastoral Empson managed to develop the linguistic concerns of Seven Types of Ambiguity into a social vision, in which a single text could register the shifting and multiple attitudes not just of one mind but of an entire age.
Empson’s own mind was complex and ill-connected, and contained many different voices: the poet, the patrician mathematician, the joker, the shocker, the drinker, the social critic, as well as the seraph of vagueness. At one point in his essay on Donne he offers a kind of parody mathematical definition of how Donne treats a single person or thing as an embodiment of a wider whole: ‘This member of the class is the whole class, or its defining property: this man has a magical importance to all men.’ He goes on to relate this use of the representative figure to his own concept of pastoral: ‘If you choose an important member the result is heroic; if you choose an unimportant one it is pastoral.’ That’s the Empson of Some Versions of Pastoral in a nutshell. You have the terrifying vrooom as his foot goes to the floor and your mind can’t quite keep up with where it’s being pulled, and then, perhaps, a slight sense that some kind of magic (or is it trickery?) has happened. And it probably has: the master of ambiguity uses ‘class’ here in a mathematical sense (of a particular category of entities) but with overtones of the social sense (of distinct social groups).
This simultaneous creation and blurring of conceptual boundaries more or less licenses his description of the phenomenon as ‘magical’: one person in a poem can stand for lots of people, and by virtue of the pun on ‘class’ all classes can be represented through a single figure. So if you use a high-class person to represent a wider ‘class’ of entities, it creates the heroic; doing so through a low-class person creates the pastoral. And then you think: cor, he’s right. Literary representation necessarily includes a range of entities beyond the particular, and top-down and bottom-up views of the world are structurally as well as generically distinct, and maybe pastoral actually doesn’t need to contain sheep or shepherds, but is indeed just a bottom-up view of human society depicted through a representative socially low figure. Then you may think: aw, come off it. Read what it says on the label and stop being so darned clever: you can’t take the pastor out of pastoral without collapsing it into all things.
Empson’s belief in the plurality of diverse elements within texts and within societies marked him as being on the left, but he was distinctly not revolutionary or overtly Marxist. So, à propos Don Quixote, he says: ‘Clearly it is important for a nation with a strong class system to have an art-form that not merely evades but breaks through it, that makes the classes feel part of a larger unity or simply at home with each other.’ This is not quite saying that pastoral is an ideological illusion designed to keep the working class in its place, because ‘a strong class system’ is not simply a bad thing or a good thing; it is rather saying that the social complexity of literary works can make it easier for different classes to coexist. There is a faint suspicion here that squire Empson regarded literature as a means of reaffirming social boundaries by their partial inversion, as though it were a kind of panto or village fête in which classes converge in order to see one another represented in and by the other, and in which irony and detachment from one’s own social group can serve to prevent one ever really trying to change things, except, perhaps, through some licensed banter over a drink afterwards.
The essays in Some Versions of Pastoral typically conclude by returning to the concept of pastoral after having veered and weaved around many themes and linguistic registers. As a result of Empson’s reluctance to display intellectual discipline of the commonplace sticking-completely-to-the-point kind, their final paragraphs can be a bit like those competitions that used to appear on the back of cornflake boxes, where you have to complete the following sentence in ten words or fewer: ‘I think Alice in Wonderland is pastoral because …’
Sometimes this can become a little desperate, as it is with the chapter about how the 18th-century classical scholar Richard Bentley emended Milton’s texts in a way which shows that somewhere in the pedantic depths of Bentley’s mind he understood what was good about the words Milton originally wrote. This chapter is really not about pastoral at all. Yet it winds up by saying that ‘so long as Milton is dealing with his first parents, it is fair to say that he has a pastoral sentiment about them. Adam’s life is so far from ours that it is necessary to imagine him with dignity from a distance, and Milton seems both to have the double feeling of pastoral, that he is both inferior and superior to Adam, and to be sufficiently at home with it to turn it into poetry.’
I’m not sure this persuades me that Paradise Lost is even in Empson’s sense ‘pastoral’, and probably he knew he hadn’t nailed the point: ‘it is fair to say’ smacks of desperation, and that double use of ‘both’ creates an obfuscating cloud of cleverness which is designed to involve readers in complexity so that they forget the simple truth: no, Paradise Lost is not really a pastoral though, yes, it could not have been as it was without Milton’s knowledge of pastoral and his scepticism about the ‘heroic’.
Being too clever was Empson’s main problem. It made him sometimes write sentences which coil the obvious around with pirouettes. It could also make him impatient of vulgar fact. His critical writings contain a high number of misquotations. These are usually corrected in the commentary of the Oxford editions, though when Empson cites Charmian as saying that Shakespeare’s Cleopatra was ‘Descended from so many mighty kings’ the notes only record the lesser error in the misquotation (‘from’ is wrong) and not the greater: the actual line includes the magnificent pleonasm of ‘Descended of so many royal kings’. Underlying Some Versions of Pastoral is also what might kindly be called an extremely vague social history, which sees ‘the Augustans’ as doing terrible things to society and the language, while the rise of Puritanism (ascribed in a throwaway aside in The Structure of Complex Words to the spread of syphilis) is supposed to have done whatever else is bad. The idea that children were innocent, Empson declares in the essay on Alice, ‘strengthened as the aristocracy became more puritan. It depends on a feeling, whatever may have caused that in its turn, that no way of building up character, no intellectual system, can bring out all that is inherent in the human spirit.’ Like an undergraduate who grins amiably while spouting what you both know to be crap, Empson could use a tone of aristocratic unconcern to get away with some terminally vague statements about ‘feelings’ and ‘whatevers’: ‘A feeling gradually got about that anyone below the upper middles was making himself ridiculous, being above himself, if he showed any signs of keeping a sense of beauty at all, and this feeling was common to all classes.’
The Structure of Complex Words was Empson’s next book. It was hard for him to write, partly because in the middle of its nearly twenty-year genesis the war happened, during which Empson worked in the BBC’s Monitoring Service, reporting on foreign propaganda. The book was an attempt to develop a systematic theory of language. The problem with that, as the Oxford editors note, is that ‘the peculiar receptiveness of his own linguistic antennae’ meant he was far more at home with instances than with generalities. Empson described The Structure of Complex Words as a ‘sandwich’, in which essays on particular words in particular texts (‘fool’ in Lear, ‘honest’ in Othello, ‘sense’ in The Prelude) are surrounded by chapters which set out a dense theoretical account of how words mean.
The Oxford editors explain that theory very clearly – indeed so clearly that readers could very easily skip the theoretical chapters. But the way Empson sets out his theory is a disastrous instance of keeping his false teeth in when he should have taken them out. He creates a monstrous regiment of over-elaborate symbols which are supposed to illustrate such finely differentiated kinds of sense that probably no one (including Empson) could quite grasp the criteria by which one is supposed to be distinguished from another. He gives charts, tables and formulae to show how words can carry what he terms a ‘compacted doctrine’, in which an implication of a word can function as a predicate of the same word, but then is prone to say that these formulae and symbols don’t really matter. His inability to grasp that not everyone was as clever as he was, combined with his reluctance ever to avoid a joke entirely, leads to sentences like: ‘A Mood is not a Sense but a sentence, and it tends to give the speaker’s personal judgement, so I shall use “£” for it as the only symbol on the typewriter which suggests valuation (American typewriters can use “$”).’
Individual words, he argues, can contain what he terms ‘equations’, which variously include or exclude aspects of their semantic or social implications. He lists five Types of ‘equation’ and then assumes for the rest of the book that a) the ‘equations’ are helpful and probably true (though he is willing to accept that his theory may be gappy or overturnable) and b) that readers will be able to remember which is which. I doubt any reader has ever believed a) or managed b) – and a second edition of the Oxford edition could benefit from a pull-out bookmark listing all the types of equation so that one could at least have a chance of seeing what Empson is on about when he assumes you can remember which was which.
The precise definitions of the ‘equations’ Empson found in complex words thankfully matter a lot less than his general purpose, which was twofold. His first aim was in the nicest possible way to set the dogs on Richards for insisting that language was either denotative or emotive and that poetry only made ‘pseudo-statements’. The second was to provide a more socially inflected view of how words function than was apparent from the lists of definitions in the OED, which was Empson’s vade-mecum but often also his adversary in understanding the inner intricacies of the language. ‘Language is essentially a social product, and much concerned with social relations, but we tend to hide this in our forms of speech so as to appear to utter impersonal truths.’ That social aspect of language is not something a non-native speaker could learn from the lists of senses in even the best of dictionaries.
If the theoretical bread around the sandwich that is The Structure of Complex Words is indigestible, the filling more than makes up for it. The chapters on ‘honest’ in Othello and ‘fool’ in King Lear are exemplary studies of how words can express and generate social turmoil, and remain among the best essays ever written on Shakespeare. The reason they’re so good is that they show how each word could be used not just with a range of senses, but with inflections of class and interpersonal attitudes which their users can only partially control. The depth and the complexity of Shakespeare’s plays are, for Empson, a consequence of Shakespeare’s ability to hear and to use these inner dramas within the ‘key words’ in particular plays. Othello repeatedly calls Iago ‘honest’ not just because he thinks Iago tells the truth, but because the word ‘honest’ carries with it a suggestion of social inferiority. A mere ‘honest’ fellow doesn’t have the gumption to set out to destroy his boss, and since you don’t respect an honest little fellow or bother to think too hard about his moral nature you don’t imagine him capable of plotting or lying. The tragedy, Empson suggests, grows from the cracks around and within ‘honest’. The word ‘fool’ in Lear operates in a similar way, both to trivialise the person thought of as a ‘fool’ and to enable him to advance under cover of folly into areas of social interaction that are potentially destructive, and which make fools of the lot of us. The exploration of these verbal dramas was an attempt to extend the social dialectic that ran through Some Versions of Pastoral into both the theory and practice of language: the ‘fool’ or ‘honest’ man can be the key to the whole show, the apparently simple which is actually complex, the lowly person in whom the world of the lordly is both included and inverted and potentially overturned.
The result is a kind of criticism that explores the social agency of words. Empson shows that the multiple senses and usages of individual words can create and imply social relationships between people, and that changes in their sense and usage can generate confusion or even tragedy for those who use them. The problem was, as always, the over-fertility of Empson’s mind, and the frantic pseudo-discipline he applied in the hope of controlling that fertility. Complex words are complex because of the range of social attitudes they can imply, and because of the range of senses they have, and because of their distinctive histories, and because of the particular circumstances in which they are used in any given text. All those statements are true, and to understand any complex piece of writing you have to some degree to engage with them all; but an attempt to provide a general theory of language will necessarily founder in the eddies of overcomplexity if it seeks to explain all of them at the same time. Our minds are always doing too much with words for us ever fully to understand them. That is actually a reassuring truth. It is why robots will never talk like humans (though humans of course may and do sometimes talk like robots), and is the main reason critical analysis of a particular piece of writing is so much more difficult than painstakingly setting out the inner logic of the word ‘ought’. But it can make literary criticism a great big mess of a muddle. Criticism is an attempt to do multifactorial analysis of things so complex that probably in the end all one can do is guide a reader’s intuitions in the right direction, and trying to do more is probably vain. But one can say of Empson that he understood more of the muddle and tried harder to unpick it than almost anyone else.
At the end of The Structure of Complex Words Empson argued that dictionaries should be reformed so as to present a kind of social anthropology of usage, which would (for instance) show how the use of words such as ‘vulgar’ or ‘common’ might ‘hint at the speaker’s aesthetic or even political opinions’. The OED has so far not reformed its definitions along the lines recommended by Empson, and that’s on the whole a good thing. It’s already a big book. But it’s surprising that the dictionary doesn’t yet include an entry for the word ‘Empsonian’. ‘Johnsonian’ is included as a head-word, as is ‘Derridean’, while ‘Empsonian’ moulders in an illustrative quotation for sense 1d of ‘ambiguity’: ‘A nuance which allows for an alternative reading of a piece of language’, the first usage of which is ascribed to Empson himself. When used today of literary critics, ‘Empsonian’ has come to carry the implication that a critic is ‘performatively intricate in expounding local verbal complexities and felicities’. It may be that the ‘Empsonian’ strain in literary criticism today is one of many reasons many people find criticism tiresome or easy to dismiss: do literary critics just display the refinement of their verbal sensoria by teasing out unimaginably complex implications that aren’t really there? Nice work if you can get it.
However, the equation Empsonian= ambiguity implies a shockingly reductive view of Empson’s critical practice. He believed that ‘literature is a social process, and also an attempt to reconcile the conflicts of an individual in whom those of society will be mirrored.’ The tricks of literature which he demonstrated by so magically pulling so many rabbits out of so many lexical hats were not inventions of the critic, but products of the complex interrelationship between individual linguistic performances and wider histories of linguistic and social change. Understanding that chaotic combination was really what Empson was about, at least in the earlier part of his career.
So why has the ‘Empsonian’ shrunk to mean, at its most negative, ‘showily preoccupied with ambiguity’? Everyone famous gets vulgarised, of course, and everything complex gets simplified over time, but one particular reason ‘Empsonian’ became so oversimplified is that Empson’s form of close verbal analysis readily fed into the methods of American New Critics, whose concerns were less sociopolitical than his. ‘Verbal icons’ of the kind analysed by W.K. Wimsatt tend to encourage worship rather than analysis of their sociopolitical origins. Another reason was that in Britain the potentially divergent elements in Empson’s criticism – the close analysis of multiple senses, and his belief that words spoke both for the divided thoughts of individuals and those of their societies – were in effect split in two.
In 1963 Christopher Ricks developed from Empson’s essay on Milton the dazzling explosion of verbal energy that is Milton’s Grand Style. Ricks, a grammar school boy, came from a background far removed from Empson’s family pile at Yokefleet Hall, and it might be said – though it may be a little unkind to do so – that middle-class aspirants to supremacy in literary criticism in the Britain of the 1950s and 1960s adopted Empsonian techniques of close verbal analysis as though they were tools and tokens of cultural mastery, so tended to downplay or neglect Empson’s interest in the social foundations of language use and linguistic change. It was a railway worker’s son, Raymond Williams, who ran with the idea that words had social histories embedded in them. His Keywords: A Vocabulary of Culture and Society (1976) was a book which Empson reviewed with an admiration qualified by the lordly but feline observation, like a lion casually cuffing a cub and knocking it out of the den, about Williams’s entry on ‘education’ that ‘it is totalitarian; quite unconsciously of course.’ Keywords is one manifestation of the ‘Empsonian’, since it responds to the historical and political aspects of The Structure of Complex Words. Its limitation is that Williams so wanted words to have impersonal cultural histories that he wasn’t very interested in usages that made difficulties for his larger historical narratives – and good poems typically do make such difficulties.
British literary critics who wore the label ‘Empsonian’ with pride tended to follow their master in disliking the overtly theoretical forms that criticism took in the later 1970s and 1980s. In the lectures I went to in Cambridge in the 1980s by Ricks and some of his most brilliant pupils, Empsonising (maybe another one for the OED) was the establishment alternative to what we were taught to think of as the French disease of structuralism. Empson himself was no fan of Derrida, whom he referred to as ‘Nerrida’ in a letter. The principled reason for his hostility to structuralism and post-structuralism was his conviction that the meaning of words is both social and personal: words mean what they mean because this person is using this word in this way to or about this other person, and because this word has this particular history which may or may not complicate how this particular person uses it. That root interest in how people speak to people prejudiced Empson against any depersonalised account of language as a system. It also led to such work as Using Biography (1984), which starts from the sensible belief that people write in the way they do because of the experiences they have had, before travelling from there far into the realms of biographical fantasy.
It is a shame that the ‘Empsonian’ fissiparated and dwindled in this way, since Empson’s earlier works display with exuberant abundance the key components of the strange amalgam that is literary criticism: they mingle sociolinguistics with historical understanding, and combine philosophical depth with acute sensitivity to particular verbal usages. Empson is always thinking about what literature does in and with the world, and his writing is always funny. Even his dashes of battiness and devil-may-care foolery are qualities from which the priestly dullness of much contemporary writing about literature could learn a thing or two. It would be good if the Oxford edition of his works were to encourage critics to rethink what the ‘Empsonian’ was, and to imagine what it could become. But scholarly editions, with all their commentaries and textual notes and introductions, are quite a weight on the soul. They can sit as heavy markers of the point at which an author has made a transition from live influence to historical monument: venerated, glossed, but no longer to be argued with constructively.