The Surrealist Fantasy of Herbert Read, Part 1: Theorizing Fantasy

Herbert Read

Perhaps the most celebrated theoretical account of fantasy was given by J R R Tolkien many decades before the genre became an established presence on the shelves of bookshops. The first version of Tolkien’s essay ‘On Fairy Stories’ was delivered as the Andrew Lang Lecture at the University of Saint Andrews in 1938; it was later expanded and published in 1947, and published again with minor changes in 1964. A decade before Tolkien gave his lecture, however, another essay on fantasy was published by an academic with very different convictions and interests. Herbert Read was an art critic, literary commentator, socialist and thinker who (among many other things) provided a critical framework for the importation of surrealism from France to England in the early ’30s. Read’s essay on fantasy makes up one of the chapters in his book English Prose Style (1928), which was highly regarded by (among many others) Graham Greene – the editor who helped Mervyn Peake publish his first novel, Titus Groan, at the end of the Second World War. English Prose Style was reprinted many times; the edition I’ve read is the seventh impression, dated 1942, and has been changed quite a bit from the 1928 version. I’d like to suggest here that Herbert Read’s essay, together with Read’s only novel, The Green Child (1935) – which is based on a ‘fairy story’ that takes a central place in his chapter on fantasy – gives us a context in which to understand Mervyn Peake’s place in the development of the genre.

Ithell Colquhoun, Gouffres Amers (1937)

I shall suggest, too, that Read’s essay gestures towards a thread or current of fantasy that runs somewhat counter to Tolkien’s version: an experimental, materialist fantasy which has less to do with tradition, historical scholarship and religious faith than with finding a means of articulating the sheer strangeness of the twentieth century. This post represents a first attempt to sketch out what such a fantasy might have consisted of if writers had chosen to follow Read’s version of the genre rather than Tolkien’s. And it will end by considering whether it might be helpful to think of Peake as in some sense a follower of Read’s. I haven’t any evidence that he was one, or that he even knew Read’s work – though it seems very unlikely he did not. Peake did know the surrealist painter Leslie Hurry, after all, and in 1939 drew a sketch of the surrealist painter Ithell Colquhoun, who was admired by Walter de la Mare, another friend of Peake’s who wrote quasi-surrealist prose and whose verse was published alongside the poetry of Herbert Read, as well as the poetry of a third friend of Peake’s, Dylan Thomas – also connected with surrealism. Read’s novel The Green Child, meanwhile, was reprinted in 1945 by Grey Walls Press, which later published The Drawings of Mervyn Peake, and reprinted again in 1947 by Eyre and Spottiswood, the publishers of the Gormenghast sequence, with an introduction by Peake’s friend and editor Graham Greene. It is tempting, then, to see in Read’s essay on fantasy, and in The Green Child, forerunners of Peake’s Titus novels, at least on certain levels. And that’s how, by way of thought experiment, I propose to think of them here.

Roland Penrose, Night and Day (1937)

Read’s book on prose style is concerned less with what he calls the ‘interest’ of literature – its contents, that is – than the formal techniques by which it achieves its effects. It is divided into two parts: ‘composition’ and ‘rhetoric’. Composition is concerned with the ‘objective use of language’: the building blocks of prose, so to speak, including words, sentences, metaphors and paragraphs as well as its overall arrangement (‘disposition’, in the terms of early modern rhetorical theorists). Rhetoric is concerned with persuasive techniques, of which fantasy is one. The part of the book that deals with rhetoric begins with chapters on ‘exposition’, which might be glossed as explaining or expressing oneself in an apparently logical manner, and ‘narrative’, which describes rather than explains, and deals with either events or objects, making it ‘either active or passive in character’ (p. 104). Fantasy is assigned to the third chapter of the second part of English Prose Style, the part of the book that deals with rhetoric. For Read, it is a persuasive technique that has not yet been given much attention, and is more closely allied with exposition than with narrative. It is, in other words, a way of writing that gives the appearance of being logical and detached, not emotionally charged as narrative is. This is unexpected, to say the least, because of the definition of fantasy that opens the chapter, which suggests that it is very far from logical.

Marion Adnams, The Distraught Infanta (1944)

The opening paragraph separates fantasy from the mental quality of phantasy, which means the imagination – the faculty of ‘forming mental images of things not actually present’, as Tolkien calls it (p. 46), following the Oxford English Dictionary. Fantasy, by contrast, is ‘caprice, whim, fanciful invention’ – the process of making things up. It is not, however, a random or passing whim or caprice; it involves sustained invention, Read insists; and this, being the place in which he diverges from dictionary definitions, would seem to lie at the heart of his conception of fantasy. He thinks of the imagination or phantasy as being driven by ‘sensibility’ – emotion or affect – whereas fantasy is more closely akin to rational thought; it is ‘cold and logical’ in the way it develops its initial whims or caprices, whereas the imagination is ‘sensuous and instinctive’. In saying this, Read claims to be building on the famous distinction between imagination and fancy in Coleridge’s Biographia Literaria. This means that Coleridge’s book is mentioned both in Tolkien’s essay on Fairy Stories and in Read’s chapter – though Tolkien is more concerned with the ‘willing suspension of disbelief’, which he wishes to replace with a different concept, ‘secondary belief’, involving a more complete mental commitment to an invented world than Coleridge’s phrase implies. For Read, by contrast, Coleridge’s definition of fancy exactly describes what Read thinks of as fantasy.

The main difference between Read and Coleridge is that Read is far more interested in fancy than imagination, whereas imagination is the faculty Coleridge favours, as Read himself points out (pp. 150-1). Fancy, Coleridge says, is concerned with ‘fixities and definites’, which Read takes to mean it is in some sense ‘objective’, dealing not with ‘vague entities’ but with ‘things which are concrete, clearly perceptible, visibly defined’. For me this implies that works of fancy or fantasy are less concerned with the creepy feelings aroused by half-seen ghosts, gods or monsters than with unexpected objects: the tea set on the table of the Mad Hatter, Mr Tumnus’s umbrella and parcels, Bilbo’s Ring. What makes these objects fantastic or fanciful is that they are evoked, in Coleridge’s terms, through an act of memory – everyone remembers having seen a tea set, an umbrella, a plain gold ring – but memory ‘emancipated from the order of time and space’. Carroll’s tea set is fanciful because the Mad Hatter’s tea party is continuous, not governed by the conventional schedule, and because the tea never seems to run out. Mr Tumnus’s umbrella is a thing of fantasy because it’s being used to ward off the everlasting snow of Narnia and is owned by a classical faun, half man half goat. Bilbo’s Ring removes him from sight and therefore to some degree from space, extends his lifetime artificially, and shortens the distance between himself and Sauron’s terrible Eye. Fantasy, then, Read tells us, is unlike exposition or narrative in that it ‘deliberately avoids the logic and consistency of these types of rhetoric and creates a new and arbitrary order of events’ (p. 138). This statement seems directly to contradict his earlier statement that fantasy is ‘cold and logical’ (p. 137); but it’s worth noting that in his earlier account of exposition Read explains that he does not use the term ‘logical’ ‘in any precise scientific sense’ (p. 92). Instead he affirms that logical exposition is ‘the art of expressing oneself clearly, logic being implied in the structure of the sentences employed’ (p. 92, Read’s emphasis). The logic of fantasy, then, is ‘implied’ rather than actual, a function of grammar rather than of rigorous syllogisms. Meanwhile its emancipation from the order of time and space – in other words, from those particular ordering principles that underpin the world we live in – frees it from the values we have been conditioned to accept. And this emancipation is an act of will rather than the involuntary detachment from coherence that takes place in a dream or hallucination. Admittedly, the will too is conditioned or given direction, Read accepts, by ‘our mental and physical environment’ (p. 138). In other words, it’s not entirely under our control. Even apparently arbitrary sentences will be driven by what Coleridge calls the ‘law of association’, that is, by the way our culture and our individual experiences have conditioned us to position things in relation to one another. But the fact remains, Read insists, that sentences in a work of fantasy or fancy ‘do sometimes present [an] arbitrary appearance’, and that this apparent arbitrariness is brought about through the ‘conscious choice’ or will of the writer or speaker.

Ithell Colquhoun, Scylla (1939)

For Read, the emancipation of a narrative from the order of time and space is often achieved, ironically enough, through the operations of time and space. The only form of literature he sees as perfectly exemplifying fantasy or fancy is the fairy tale, a form of collective verbal property that has gradually lost its links with any particular time and place by being handed down from one generation to the next, and by being transferred from one location to another, changing as it goes as if in a game of Chinese whispers. The theme of each tale remains constant, he affirms, ‘but there is a gradual accretion of subsidiary details’, and a gradual loss of emotional investment – of sensibility, that is, or affect – so that the ballad or folk tale eventually becomes ‘a clear objective narrative’ which is ‘encumbered with odd inconsequential but startlingly vivid and concrete details’ (p. 139). The claim that fairy tales accrue concrete details as they get passed down seems to me a little odd; could one not just as easily argue that certain details get lost over time, and that this gradual loss of details is what makes any given tale seem arbitrary? Tolkien too claims that new ingredients are always being added to what he calls the ‘soup’ of story – the communal source of imaginative nourishment for succeeding generations – but he doesn’t suggest that the new ingredients add up to a steady accretion. The best-known fairy tales, after all, aren’t overburdened with details, as they surely would be if accretion were continually in process. The oddness of Read’s claim is compounded when he offers as an exemplary fairy tale a narrative that has little in common with the fairy tales of Andrew Lang, or Joseph Jacobs, or any of the major collectors who helped to naturalize the term in the English language. It’s the story of the two Green Children, and it provided Read with part of the plot of his only novel.

The story concerns two young children who were found in a specific, extant place – St Mary’s of the Wolf-pits in Suffolk, now known as Woolpit – near one of the pits that gave the place its name. The children had green skin, and were taken for questioning to a local dignitary, Sir Richard de Caine, who lived at nearby Wickes. The children did not speak English, so at the time of their discovery nothing much could be learned about them except from their actions. At first they would eat only beans, a detail described in Read’s account with great specificity: when the beans were placed before them the children ‘opened only the stalks instead of the pods, thinking the beans were in the hollow of them; but not finding them there, they began to weep anew’ (p. 139). The little boy died soon afterwards, but the girl lived on as one of the knight’s servants, gradually becoming used to ordinary food and losing her green appearance. She was ‘rather loose and wanton in her conduct’, the narrator tells us. Interrogated about her birthplace, the girl insisted that everyone there was green and that the sun did not shine there; instead the land existed in a state of perpetual twilight. While minding sheep, she said, she and the boy had wandered into a cavern filled with the ‘delightful sound of bells’, and got lost in an underground warren of caves and passages. Emerging at last into the open air, they found the sunlight so dazzling and warm that they lost their senses. They were woken by the noise of their approaching captors and tried to flee, but were caught before they could find the entrance to the cavern.

Dorothea Tanning, Eine Kleine Nachtmusik (1943)

So the story ends, without a trace of the ‘eucatastrophe’ or uplifting ending (conventionally signalled by the formula ‘and they lived happily ever after’) which was for Tolkien one of the defining elements of the fairy tale tradition. The story has no perceptible moral purpose, and much of its narrative seems to build on the one seemingly impossible detail it contains: the fact that the children had green skin. Their diet of green vegetables ‘explains’ the greenness of their appearance, and this greenness in turn ‘explains’ the girl’s wanton behaviour (the colour green being often associated with sexual promiscuity in medieval and early modern culture; leeks, for instance, were emblematic of randy old people, because they had white heads and green tails). Greenness also suggests pastoralism, as represented in the children’s work as shepherds, and a plant-like link with the earth, which explains the children’s discovery in a pit and their connection with underground caves. The story, then, gives every impression of having a logical structure, even if the logic is founded on something that seems unreasonable or at least unprecedented: the existence of green children. It perfectly exemplifies, then, Read’s description of fantasy as coldly objective (think of the way the boy child dies without being mourned, even by his sister) and seemingly logical (though of course there is no real logic in any of the spurious ‘explanations’ I’ve just suggested). It is some way, though, from exemplifying a conventional fairy tale.

One reason for this is the fact that it does not seem to be wholly detached from the order of space. It’s carefully located in an actual parish in Suffolk, and has the name of a knight – perhaps a real one – attached to it, a man who lives in a neighbouring parish and whose identity could therefore be traced, in theory at least, with the help of old records. The possibility for such research promises to pin the story down in time as well as space. At the same time, there are unexpected gaps or absences in the narrative. The story gestures towards the presence in the world of the people known as fairies, without going so far as to name them. The green children could easily have been explicitly identified with those people if the narrator had wished; after all, fairies famously live under the hills of Ireland and Scotland, they are associated with the colour green, and their well-known predilection for strange food – such as green beans contained in the stalk of the plant instead of the pod – makes it dangerous for mortals to eat from their tables. The term ‘fairy’, however, is never used in Read’s version of the story, and this renders the experience of reading it much stranger than it would have been otherwise. The narrative, then, is both attached to a specific place and time and detached from a traditional folkloric tradition. And its structure is not quite story-like, either. To say that it has ‘no moral’ is to say that it doesn’t really have an ending; one of the children dies, the other becomes ordinary, and we never hear what became of her after her naturalization as a human being. In choosing this not very fairy-tale like story as an example of fairy tale, then, Read seems deliberately to be detaching it from any familiar frame of reference. It hovers between fiction and history, between common ingredients of folklore and a set of circumstances that are wholly unaccountable by any known frame of reference – circumstances rendered stranger still by arbitrary details, like the implied existence of unusual beans or the ‘delightful sound of bells’ in the cavern.

Lee Miller, Portrait of Space Near Siwa, Egypt (1937)

Read goes on to assert that ‘a story such as this is the norm to which all types of Fantasy should conform’, including fairy tales which are ‘literary inventions’ where the ‘will or intention of the writer has to take the part of the age long and impersonal forces of folk tradition’. His example of such a literary invention is Robert Southey’s story of the Three Bears, whose heroine is now a little girl called Goldilocks, though in the original version she was a nameless wicked old woman. The story doesn’t have much in common with the story of the Green Children, however; it involves intelligent bears who live in a house and eat porridge, it does not locate itself anywhere specific, and it has a conventional story structure, with the heavy stress on patterns of three (three bears, three bowls of porridge, three chairs, three beds, three trials for the wicked old woman) which is omnipresent in the oral tradition. What attracts Read to Southey’s tale is its lack of moral intention or indeed any identifiable purpose, which sets it in contrast to Charles Kingsley’s The Water Babies, which has a clear didactic purpose, or rather several. Read also admires its ‘objectivity’ or lack of sentimental bias – we’re not expected to sympathize too deeply with either the woman or the bears – and its arbitrariness: the central premise concerning bears that live in houses, eat porridge and own household furniture has been randomly plucked, as it were, from thin air. The instructive purpose of The Water Babies, by contrast, attaches it too securely to the culture of its author to make it a perfect fantasy, and the same is true, as far as Read is concerned, of Alice in Wonderland, which is not so much moralistic as cultured; ‘the intelligence active therein is too sophisticated, too “clever”’ (p. 144). Read doesn’t explain what he means by this, but he may be referring to the knowing parodies it contains of familiar poems – ‘How doth the little crocodile’, ‘You are old, father William’ (the latter based on a poem by Southey) – or to the conventional Victorian manners which are constantly being violated by the people Alice meets, or their use of sophistry or chop logic, a form of discourse that depends for its comic effect on a recognition of its playful violation of conventional reasoning processes. Alice is anchored to its time and place by its sophistication and allusiveness, just as Kingsley is anchored to his time and place by his didacticism, which puts him into direct dialogue with the educators, legislators and churchmen of his period.

What becomes clear about Read’s examples of fantasy in this chapter is that they are so diverse as not to have anything much in common at all, apart from the capriciousness with which they introduce manifest impossibilities into their narratives (although one example, a passage from James Joyce’s Ulysses, doesn’t even contain these). Utopias like W. H. Hudson’s The Crystal Age are for Read too satirical and moral in their aims – too specifically directed at targets in their own time, as Hudson himself acknowledged in his preface to a late edition of his first book – to be ‘pure’ fantasies. H. G. Wells’s The Time Machine is very nearly fantastic, but too enamoured of the conventional rationalizing discourse of science to be fully so (and Tolkien thinks the same thing; what stops it being fantasy for Tolkien is the presence in it of the pseudo-scientific device of the time machine itself). A text Read does identify as a perfect literary fairy tale – an extract from a story by the Russian writer Aleksei Mikhailovich Remizov, translated by the great fantasy novelist Hope Mirrlees and her partner Jane Harrison – is to my eyes decidedly sentimental, and hence hardly ‘objective’, which is one of the features of fantasy Read insisted on earlier. Its sentiment is extended to a fallen star, however, so perhaps he sees it as an exemplary fantasy because of its emancipation from familiar assumptions about the difference between people and objects: ‘The poor little star was dozing by the hare’s form, and the thaw of a little tear rolled down her star cheek and then froze again’ (p. 145). The Thousand and One Nights is the ideal fantasy epic, the text Read would most like to have seen emulated in English – though not in the manner of William Beckford’s Vathek, which for Read adheres too closely to the original to be easily distinguished from it, and was in any case written in French. Read’s list of examples ends with a passage from one of Philippe Soupault’s surrealist short stories, ‘The Death of Nick Carter’, which is ‘still too allusive’ to be a pure fantasy, too enmeshed in details that root it in a particular time and place. By the end of the list of fantasy passages – a list that was greatly extended between the first edition of English Prose Style and the one I’m citing – one could be forgiven for having lost all sense of there being a ‘norm to which all types of Fantasy should conform’, which Read tells us is exemplified in the story of the Green Children.

Dorothea Tanning, Children’s Games (1942)

It seems clear that this is entirely deliberate on Read’s part. Read’s conception of fantasy as having the appearance of arbitrariness and being emancipated from the orders of time and space would surely be undermined by a list of examples that were similar in subject or technique, or that fell into a consistent narrative form, involving for instance multiple repetition of the pattern of threes that structures Southey’s story of the three bears. His examples, whether of ‘perfect’ fantasy or of imperfect near-fantastic passages, are constantly flinging the unwary reader in new directions, which emphasizes the seeming arbitrariness of each new set of inventions. Their variety tends even to obscure the set of criteria by which he sets out to limit the fantastic; almost the only thing they have in common is their arbitrariness. Read himself admits that very few of the passages can be described as wholly emancipated from time and place, while it’s hard to see some of them as displaying anything like cold logic or objectivity. Fantasy emerges as a rhetorical strategy that refuses stability and conformity and embraces innovation as vigorously as it can without rendering itself incomprehensible.

There’s a gap, then, between the theory of fantasy Read advances in the first half of the chapter and his examples of it in the second. The gap is anticipated in the introduction to English Prose Style when Read makes a distinction between the titular terms prose and style ‘in the abstract’ and any examples one might offer of each (p. ix). ‘In the abstract’, Read tells us, means ‘a priori, without the prejudice of particular examples, and as a preliminary to a more minute analysis’. This makes ‘in the abstract’ sound like a reference to one of Plato’s Ideals, the original things or concepts of which everything in the world is merely an imperfect shadow or copy. Fantasy, too, as it is explained by Read with the help of Coleridge’s fancy, is given to us ‘in the abstract’ at the start of the chapter he dedicates to it; it’s an a priori ideal rather than a concept that can be arrived at by considering examples of it, which will only serve to ‘prejudice’ those who examine them. Where Tolkien’s fairy story is a kind of narrative that emerges out of the past, in other words, Read’s fantasy is a kind that may not yet exist; already-extant instances of it will invariably fall short; perfect examples of fantasy as Read conceives it are yet to come, and need to be traced from imperfect past examples into the literature of the future rather than from the present into the annals of history and the shadows of prehistory, as is the case with Tolkien’s fairy stories. Read was already beginning to be famous, at the time when he first wrote the introduction, for his facility in tracing future directions of art and literature in the work being produced at the time of writing; and in the 1930s he wrote some of the most celebrated essays on abstract art of what we now call the Modernist period. The examples he supplies of fantasy deliberately move through time from the deep past of the Thousand and One Nights to the fourteenth century, when the story of the Green Children was first recorded, to the recent experimental prose of Joyce, Remizov and Soupault (published in English in 1922, 1926 and 1927 respectively). For Read the most perfect examples of fantasy, as of modern art, came from overseas – the Thousand and One Nights is the standard to which he urges English-language writers to aspire – and the best English passages are provided only as evidence that such an achievement might be possible in his native language, not as rivals for the Persian or Arabic epic. His chapter, then, is a call to artistic action as much as an analysis; a work of rhetorical exhortation as much as of scholarship.

Lee Miller, Bathing Feature (1941)

The sense one gets, in fact, from Read’s discussion of fantasy is that he’s less concerned with establishing its properties and formal techniques than with the political possibilities it embodies. If rhetoric is about persuasion, fantasy for him is specifically about persuading the reader to imagine liberty, and hence must give the appearance or air of being liberated in terms of its form and content. The key to this concern with politics is his stress on the highly political term ‘emancipated’ in Coleridge’s definition of fancy. Emancipation from the order of time and place is hardly a lucid statement of stylistic technique, but it can certainly be read as a statement of a political position; or rather, not a position so much as a strategy. Read’s own politics, while remaining strongly attached to the Left, were changing constantly in terms of his affiliation with different movements. In his youth during the Great War, for instance, he was attracted to the anarchism of Kropotkin, but he later flirted with Marxist Communism, Trotskyism and Guild Socialism, and even spoke in 1934 of welcoming the notion of a ‘totalitarian state, whether in its Fascist or Communist form’ (though he was thinking of totalitarianism here as an ‘economic machine to facilitate the complex business of living in a community’). Read eventually returned to anarchism in response to the Spanish Civil War of 1937. Common to all these shifts of ground, however, was a refusal to be pinned down to a singular position, a formulaic narrative – and in particular the refusal to submit himself to authority, whether of an individual or a party (apart from his half-flippant comment about totalitarianism). Read was always in quest of the ideal society, the ideal way to live in a community as an enfranchised or liberated subject, and embraced anarchism in the end as a means of continuing that quest indefinitely instead of being bound to a party line by the dictates of some unaccountable central government. The same quest or impulse propels the narrative in his only novel, The Green Child, which can be read as the ultimate example of the ideas on fantasy expressed in the continually changing pages of English Prose Style.

The Green Child shall therefore be the subject of my next blog post.

Ithell Colquhoun, Gorgon (1946)

Book List

W. H. Hudson, A Crystal Age, Fourth Impression (London: Duckworth, 1919), Preface (from 1906)

Herbert Read, English Prose Style, 7th Impression (London: G. Bell and Sons, 1942)

Herbert Read, The Green Child, introd. Graham Greene (Harmondsworth: Penguin 1969)

Herbert Read, The Green Child, illus. Felix Kelly (London: The Grey Walls Press, 1945)

J R R Tolkien, ‘On Fairy-Stories’, Tree and Leaf (London: HarperCollins, 2001), pp. 3-81.