It is hard to imagine a world without maps.

Now stop—and diagram that sentence. Break its syntax apart. You can parse it in at least two valid and meaningful ways:

• It is hard | to imagine | a world without maps. The use of maps is so embedded in our daily lives, so essential to our normal functioning, that the idea of a pre-cartographic society is as alien as the thought of a pre-literate one. On top of this, our idea of what it means to be a mapped society is itself confined to our familiarized expectations of what maps are like. How did people get by without maps—or rather, without the sorts of maps we know and understand?

• It is hard | to imagine a world | without maps. Maps govern the way we think about space, and that extends to imaginary or hypothetical spaces. Without a graphic representation on paper or in our heads, our plans for things not yet built—homes, roads, electric circuits—may be cloudy and ambiguous. They may lack precision in the same way we have trouble with describing things that are outside our linguistic abilities. This is a negative definition of maps as a form of language: to be without a map is to be without language, and it impedes us from communicating ideas in the mind—to others, yes, but also to ourselves.

In both of these senses, maps of fictional places are remarkably challenging texts.

One of my chief interests in fiction, along with art in general, is how it presents itself as evidence of the way people receive the existing cultural data around them before they process it and spit it back out. (In literary criticism you will encounter words like allusion and intertextuality, but I think of them as subtypes of a broader cognitive activity.) When an author plans out a story’s setting in place, or when a reader attempts to reconstruct it from the words alone, the maps they produce tell us not only how they imagine the depicted geography, but also how they imagine the idea of maps. Furthermore, the author/audience distinction isn’t always sharp: some privileged readers, such as the illustrators at a publishing house or manuscript historians like Christopher Tolkien, participate in the interpretive stage as well as the official construction of the space for everyone else.

So when we open up a novel to find a map, we can think of the map as an act of narration. But what kind of narration? Is it reliable narration or a deliberate misdirection? Is it omniscient knowledge, a complete (or strategically obscured) presentation of the world as the author knows it? Or is the map available to the characters in the text? If it is, then who drew up the map, and how did they have access to the information used to compose it? If it isn’t, then through what resources do the characters orient themselves in their own world? And finally, does anyone even bother to think about these questions before they sit down to place their woodlands and forts?

In the post that follows, I am going to informally sketch out a theory of fictional maps, which is to say that I will put up a lot of pretty pictures from novels and talk about why they are neat. There is likely some academic work on this somewhere—I would be astonished if there weren’t—but I’m not aware of any, and certainly nothing that has accounted for modern critical approaches to the history of cartography. Map history and the comparative study of commercial genre literature are niches within niches as it stands, and my aim is to entwine them together.

Perspective

Let’s begin with something familiar.

Depending on how you look at it, this map is one of the following:

• The map that J.R.R. Tolkien drew up for The Hobbit, which appears in the endleaf of the original 1937 edition as well as most (if not all) of the English editions still in print today.
• A map drawn by the dwarvish king Thrór depicting the environs of Erebor, the Lonely Mountain. Elrond deciphers the runes in Chapter III (“A Short Rest”).
• A reproduction of Thrór’s map, copied and translated by the hobbit Bilbo Baggins.
• A reproduction of Bilbo’s copy of Thrór’s map, received and delivered by one J.R.R. Tolkien from There and Back Again, the first part of the discovered manuscript known as the Red Book of Westmarch.

The complexity of the document is that it serves as all of these things at once. As Tolkien’s map, which we recognize to be a fictitious construction along with the rest of the text, the map is a device to orient the reader in an imagined world. But if we dive inside the fiction, the map is also Tolkien’s way of reporting to his readers what Bilbo and Thorin were looking at—no different than if your copy of the book came bundled with a replica of Bilbo’s sword, Sting.

What’s more, in style and technique the map is fully believable as something put together by the dwarves (apart from the lettering in English, which we can think of as Bilbo’s translation if we wish to suspend disbelief). Notice that rather than being a high-fidelity communiqué of how Tolkien imagined Middle-Earth, the map is a minimalist sketch of the world according to the dwarves. The sparsely chosen landmarks appear in relative (not absolute) position, the illustrations are abstract, and the inscriptions allude to people and events that would have been known to Thrór. Scale doesn’t even enter into the equation. (Tolkien’s original draft, which I’ll say more about later on, was even sparser: aside from the runes and text, its only graphic elements were the Running River and a top-down outline of the Lonely Mountain.)

Not to be neglected, of course, is that the map also functions as a two-layered riddle. In The Hobbit, we learn that while the runes on the left (in red above) are directly visible—”five feet high the door and three may walk abreast,” they read—the runes in the centre only reveal themselves when Elrond holds the map up to the light of the moon. (This custom-made replica demonstrates the effect.) The moon-runes provide a further clue: “Stand by the grey stone when the thrush knocks, and the setting sun with the last light of Durin’s Day will shine upon the key-hole.”

To complicate things further, when the dwarves first lay out the map in Chapter I (“An Unexpected Party”), Tolkien makes an authorial interjection in the text:

“There is one point that you haven’t noticed,” said the wizard, “and that is the secret entrance. You see that rune on the West side, and the hand pointing to it from the other runes? That marks a hidden passage to the Lower Halls.” (Look at the map at the beginning of this book, and you will see there the runes in red.)

Not only that, but it will say this in the text whether your edition has the map printed in red and black or not! (Now that we’re in the age of paperback dominance, it’s unlikely that this is the case for you.)

So the multifaceted nature of the map isn’t limited to its plurality of authors, each at a different level of absorption into the fictional world; also in play is a plurality of potential readers. Tolkien’s real-world readers aren’t expected to go in knowing how to decipher the runes (though nowadays, you’d be surprised at how many of them do). But the further concealment of the moon-runes tells us that within the narrative, the mapmaker had a restricted audience in mind.

If the map is an act of narration, what kind of narration is it? We have a good system for answering this type of question with respect to prose. First, there is the distinction of person—first, second, or third—which is largely a question of using pronouns to position the narrator and reader in relation to the action. The concept of person doesn’t map neatly onto cartographic works, however, unless there are pronouns involved. A better apparatus for distinguishing between the possible authors of a fictional map is what literary scholars call levels of diegesis—a technical way of delineating whether something is outside the text, inside the text, or inside a text within the text.

At minimum we are always dealing with three layers of reality, though they are not always separate: the author, the narrator, and the characters. In non-fiction, for instance, we observe no distance between the author and the narrator, and we assume that the inhabit the same plane. In the most basic form of first-person narration, we assume that the narrator is among the reality of the characters, even if he or she is far removed from the action. If we think about maps in diegetic terms—if we ask whether the documents and their authors belong to the world in the book, or if they come from outside—we unlock two of the most powerful concepts for thinking about perspective: omniscience and reliability.

Diegesis

What’s the appeal of a fantasy map, anyway? I doubt too many would disagree when I assert that maps add to the sense of immersion. If they come directly from the author—keeping in mind the intermediaries of the publisher and illustrator, too—a map tells the reader that the creator of the fictional space has really thought this place out. Storytelling always happens in façades, but evidence of the author’s forethought fills out the setting’s illusion of depth.

If we see a map of an imaginary land, we feel like we know more about the place. Like any supplementary material—timelines, family trees—it satisfies our latent curiosities. Also present, I think, is an element of bowing to the world-builder’s authority: by looking at a map we don’t simply know more about the world—we know more about how the author imagined the world to be.

But here’s the trap: the attitude of wanting to know more about a world—and moreover, believing that a map can draw us closer to it—leads the audience to default to a certain passivity. To a certain extent this is the criticism that has always been made of illustration, cinema, or any kind of embellishment beyond mere words alone: that when something is imagined on our behalf, we are robbed of our duty to reconstruct the textual reality for ourselves.

It’s strange, for instance, that the Yoknapatawpha novels of William Faulkner are notorious for compelling readers to cobble the logic and action together from jumbled scraps of unreliable narration, but with the right edition in hand we get the geography delivered to us on a plate:

From the perspective of literary history, I very much admire Faulkner’s maps: as you can see from the University of Virginia’s collection of Faulkner manuscripts, his sketches were instrumental in developing the sense of overlaying his novels on top of each other like an eternal palimpsest. And I don’t have a problem with sitting back and letting a provided map do some of the work for me, myself. The real danger lies in the assumption that authors’ maps can do this for us at all.

Regardless of whether a map depicts a fictional or real-world space, we always have to ask ourselves whose knowledge the map represents. This notion is intuitive when we think of real-world maps, where we really never have access to an omniscient perspective that lies outside our own reality. All that we have to work with are the source materials and surveying techniques that are known to us, and the rest is speculation. Increasingly, scholars of cartography now recognize that maps embed our political dispositions and socio-economic practices like any other texts: even in the satellite-enabled age of Google Earth and GPS, we start out from inside the world and set specific agendas for what we wish to represent, and how. Only superficially are maps ever “objective” views of the land.

Fiction throws a serious wrench into the way we think about maps because of our familiarity with omniscient points of view. In narration, we typically think of omniscience in terms of the narrator’s godlike reach into the interior experience of the characters, but the notion also applies to whether the narrator has access to information that the people within the fictional world do not.

In the case of an extradiegetic map, a map explicitly outside the text that serves as a direct conduit from author to audience, we tend to assume that the information it communicates is a sort of objective truth. We don’t mind so much if the liberties taken with the style of illustration—the medium, the lettering, the colour—would not have been available to the people who inhabit the imagined setting; if anything, such extravagances add to the romance and mystique.

But with maps that purport to speak from within the tale—maps with here-be-dragons blanks that delimit what the people in the fiction perceive or care about—we run into the same paradox that authors who deal with far-flung places frequently encounter with language. The author feels a certain duty to communicate to the reader with maximal precision, but the higher the fidelity of the message, the more it draws on our present-day idioms and conventions, and the more it strains plausibility.

A truly plausible map is one that we could imagine being created with the techniques and materials available to the people within the world; one where we could see the world’s cartographers as having the inclination, purpose, and skill to create something that looks like the product in front of us. Thrór’s map in The Hobbit is a classic example. Jim Hawkins’ map in Treasure Island is another:

The depth soundings, the rendering of the coastline, the dated handwritten addenda, the note that this copy is a facsimile (thereby accounting for its tidy, finished look)—everything about this illustration plays the part of an authentic nautical chart.

Then again, Treasure Island has the benefit of being set in a recognizable culture and time period, and we have a historical point of reference by which to gauge its illusion of authenticity. This can’t be said of secondary-world fantasy, the stories that take place in lands that have little to no geographic relation to the planet Earth we know.

The fictional cartographer’s paradox of perspective is more apparent in a map like this:

This illustration comes from Star Wars spin-off novel Vector Prime. Published by Del Rey when they first acquired the license to the tie-in books, it was the first officially sanctioned map of the galaxy depicted in the merchandising empire that has grown out of the George Lucas films. It’s an attractive design: the simple spiral arms and the dashing lettering recall the retro-futuristic pulp-serial aesthetic that inspired the Star Wars series. (Far better, at any rate, than some of the soulless pseudo-3D charts that followed it.)

Yet you can see how its heart is in two places at once. It’s a two-dimensional printed work depicting a universe where there isn’t a shred of print anywhere to be found: Star Wars, you will remember, is famous for projecting any and all visual information into a hologram or wireframe schematic. There are no illusions here that this map originates from within the galaxy, as it clearly doesn’t. But the map also situates itself in a limited heroic point of view. It’s the world according to the protagonists, with its hyperspace trade routes and dotted-line expansion frontiers, its Unknown Regions and Wild Space—all of it condensed and flattened into the pancaked realities of the twentieth-century printed page.

In a way, the map’s abstraction of the galaxy reminds us that we are dealing with a comic-book reality, albeit a thoroughly developed one. It has the omniscient privilege of sitting outside the text, but instead of filling in the blanks it elides them with the flair of escape.

Materiality

As it turns out, the pragmatics of publication have a noticeable effect on the kinds of maps we see in fictional works.

If we return to our first example, Thrór’s map of the Lonely Mountain, notice the orientation. The Iron Hills to the east are somewhere up top; the spiders of Mirkwood out west are down at the bottom. Now, as readers receiving The Hobbit as a freestanding text divorced from history, it’s easy to come up with all sorts of justifications for why the map is east-side-up. We could say that it’s a cultural peculiarity of the dwarves. We could even say that it befits the direction of Bilbo’s quest, upwards being the way forwards. But if we delve into how the book ran its course from manuscript to first edition, a different story emerges.

Tolkien’s original concept for Thrór’s map was a page in portrait orientation, with north on top and south on the bottom as we would typically expect. As we know from Tolkien’s correspondence of January 1937, the plan was to have the map inserted into Chapter I at the point where Gandalf first shows it to Bilbo and Thorin. In the original unpublished map, the moon-runes were inscribed in reverse on the back of the page so the reader could reveal them by holding the map to the light, just as Elrond does in Chapter III of the novel. (I unfortunately don’t have a picture of Tolkien’s original manuscript map at hand, but it’s in the Bodleian Library at Oxford (MS. Tolkien drawings 33); I believe it also appears in The Annotated Hobbit.)

The orientation of the map was rotated into its present form when Tolkien’s publisher, George Allen & Unwin, refused to produce it in the form that Tolkien intended, primarily for reasons of cost. Thus the map was redrawn as an endleaf in landscape orientation, with the dimensions of two pages side by side.

Where the plot thickens is when you look at The Hobbit in translation. Here is Mikhail Belomlinsky’s map from the Russian edition of 1976:

The illustrations in the Soviet Hobbit have a very distinctive look; Gollum, in particular, is fantastic. As for the map, Frank Jacobs of Strange Maps has already written an excellent analysis of its Slavic character, which I will not repeat here.

One thing to notice, however, is the map’s use of space—specifically, its complete spread over the page. Rather than adhering strictly to the canonical geography of Middle-Earth, which was by then widely known (outside of Russia, anyhow) thanks to The Lord of the Rings, Belomlinsky tucked something interesting in every nook and cranny. Look at how Mirkwood and the Misty Mountains snake along with the rivers. Spanning the whole adventure from Hobbiton to Smaug, west to east, bottom to top, one almost forgets to catch that the directions on the compass rose are wrong.

Observe: с, в, ю, з—северо, восток, юго, запад—north, east, south, west. By preserving the orientation of Thrór’s map in the English edition but not adjusting the compass directions to match, we end up with an erroneous map where Hobbiton is to the south and the Lonely Mountain is to the north. Preposterous! Yet we arrived here at the first place because of a series of publishing decisions across multiple editions that determined if the map would be printed on one page or two.

Flipped cardinal directions are hardly unique to The Hobbit, mind you. For a very long time, maps of L. Frank Baum’s Oz had the Munchkins to the left and the Winkies to the right, despite how clear it was to every Oz reader that Dorothy landed on the Wicked Witch of the East and made her way westward to Winkie country:

If we think of fictional maps as a study in the medium constraining the message, where it gets interesting is when you consider the maps that exist as documents within the story. Publishing considerations have the power to shape the reality of the imaginary place. Thrór’s map in the narrative plane of The Hobbit is oriented east-side-up because Tolkien’s map had to be so.

Stepping inside the fiction, we may ponder if the material constraints on mapmaking play a part in the story. In the context of these imaginary worlds, is there any suggestion of how maps are produced, transported, and preserved? Does their content spring forth from an overriding purpose or utility? These are considerations that buttress the depth and plausibility of a map, much as how Tolkien’s meticulously crafted languages and folkloric songs carry the impression of a bottomless history—the sense that this world and its people weren’t born yesterday.

Indeed, one of the finest touches in the Peter Jackson films of The Lord of the Rings is the scene at the exact midpoint of The Two Towers: Faramír, having captured Frodo and Sam and returned to his refuge at the Window to the West, goes over a map with his lieutenants. It’s a minor but effective embellishment on the book that sums up the various troop movements on the board while providing a brief visual treat for the Tolkien enthusiasts—and rather than displaying the map as an animated overlay with voiceover narration, it’s presented as a thoroughly creased parchment over which Faramír runs his finger. The map isn’t any old god’s-eye-view: like a real-world map, it’s a strategic instrument on the field. As a study in condensing an immensely challenging work for the screen, this subtle directorial decision stands out as a masterstroke.

Verisimilitude

One crucial point to take away from what I’ve said thus far is this: the coherence of a map with respect to the fiction it represents is not necessarily the same as its realism or level of detail. As much as we may desire to imprint an author’s vision onto our own with the utmost fidelity, I do not believe this ought to be the aim of fantastic cartography. As anyone who has been following the brief history of computer animation would know, verisimilitude isn’t at all the same thing as immersive reality. It is not the case that maps draw us closer to a believable fictional space with greater topographical accuracy or ever more lifelike terrain.

Take, for example, the six-book series by William Horwood that begins with Duncton Wood. The Duncton books revolve around communities of moles that worship the ancient stone circles of Great Britain, and some of the moles are capable of inscribing things in writing with their claws. One of the characters, Mayweed, is a talented navigator who sketches a map that appears (in translation, of course) in the third volume, Duncton Found:

Now contrast the deliberate, playful simplicity of this map with the one below, which appears in the second Duncton trilogy:

The second map looks substantially more attractive at first, thanks to the naturalistic depiction of terrain features. Unlike the original sketch, it looks like the sort of work that only a professional illustrator could pull off, and so it has the air of being the more polished, artistic piece. But it is also indisputably drawn from a human’s overground point of view—odd for a series where the moles largely keep to their burrows and humans are barely present apart from the odd occasion where they zoom along in their “roaring owls” (automobiles). And we might be fine with taking it as an omniscient artistic rendering from the publishing house, but the claim in the caption—”Based on Mayweed’s map found in Seven Barrows”—so brazenly contradicts the perspective of the map that it only serves to throw us off.

So as readers, we may appreciate the second map of Moledom as the superior illustration purely in terms of artistic merit—but it’s not clear at all that this map is better suited for the books. To the contrary, it presents a greater impediment to our ability to suspend disbelief.

The gravitation towards realistic detail in maps is especially remarkable if you remember that maps are inherently abstractions. The whole point of a map—of any variety, not solely the geographical kind—is to pack the chaos of information into a selectively delimited and instrumentally efficient container. Short of the 1:1 scale maps of Lewis Carroll and Jorge Luis Borges, cartography is a form of compression.

In terms of utility, the value of going from abstractions back to high-fidelity detail lies in how distinguishing visible landmarks—coastlines, mountains, fortresses—is useful for navigation. Before the proliferation of contour lines and coloured heat maps as methods for representing elevation, solutions for shading peaks and valleys led to the maps we now look back to fondly as exquisite in their artistic finesse. Satellite photography, which offers the highest fidelity of realistic representation we can achieve, is an excellent general-purpose tool because for a clientèle as diffuse as Google Earth’s, one never knows which topographic peculiarity might be useful at any given time.

Contrast that with maps of imaginary places, which are like their real-world kin in that all of them are works of art, but differ in that not all of them have a targeted function beyond serving, somewhat vaguely, as guidance for the reader (and often the author, as architect and city planner). Beyond looking pretty, framing the action, and setting the scene, the stylistic decisions often seem to lack any functional rationale.

Have a gander at this:

This is Ellisa Mitchell’s much-admired painting of the lands of Robert Jordan’s Wheel of Time, which for some reason is held up by some as one of the gold standards of fantastic cartography. When the up-and-coming fantasy novelist Saladin Ahmed called on Internet artists to produce a map for his forthcoming book, one that would outstrip the “very serviceable, basic, black-and-white line map” that his publisher could provide and “move beyond utility”, Mitchell’s map was one of the exemplars he had in mind.

Now, it’s obvious why Mitchell’s map is well received. It doesn’t take Sherlock Holmes to deduce that inveterate fantasy readers like Saladin Ahmed admire its colour and natural gradients, things you would never find in a mass-market paperback for reasons of cost alone. And embellishment for purely aesthetic reasons is certainly nothing new: back when printed atlases were engraved, colourists would fill in rivers and trace political borders on individual copies by hand. But look at how the version above compares to the black-and-white paperback map:

At first glance it seems like Mitchell’s painting is simply a more expensive and lovingly crafted reimagining of its monochrome counterpart. What it actually does, however, is emphasize topography at the expense of features that are more narratively functional, like political borders and roads. By foregrounding the terrain, the linear elements of the map recede.

(Doubly fascinating is that for all her good intentions, Mitchell unwittingly captures the trappings of the Robert Jordan series—indulgent top to bottom, muddled to the point of being unreadable, and plastered with a faux-medieval ethos that screams inauthenticity. I know, I know; I’m being much too harsh. Comparing the map to the books is incredibly unfair to the map.)

So here’s the question we should ask: why is it so attractive to conclude that the painted poster is the superior piece of work?

In part, this has to do with the realities of the mass-market fiction industry. Considering the practical limitations in the age of desktop publishing and (in many cases) direct-to-paperback, it’s uncommon that books will come with endpaper maps printed in multiple colours like the first edition of the The Hobbit. You’ll certainly never see historically accurate engravings in the mode of Treasure Island. In this environment, it isn’t altogether surprising that readers would treasure maps in other media for their relative rarity—even though in truth, maps for the flagship fantasy brands (particularly those adapted to film) are issued and sold en masse.

What this explanation doesn’t address, though, is why topography is the element that takes precedence to everything else in receiving what you might call the artistic treatment. In all probability, this is grounded in the desire for immersion; read any encomium for fantasy mapmaking and the first thing you’ll hear is that maps make the world more real. But realism is really just a code for saying that something is more in line with our embedded assumptions about what it means to perceive the world, and in the cultural value system we live in today—empirical, literal, photographic—it refers to the imagined experience of seeing the physical terrain with your own eyes.

Weltanschauung

The most striking thing about fantasy maps as a whole, especially the sort that dominates the industry of doorstopper mythopoeia that claims to be descended from J.R.R. Tolkien, is how rigidly they stick to convention. No matter how nice they look, structurally they reduce to orthogonal landmass drawings. Rarely, if ever, will you find visualizations of an imagined cosmology like the commonplace depictions of Midgard and Asgard wrapped around Yggdrasil in Norse mythology—and you would think high fantasy in secondary worlds is the genre that could use them most.

This is a problem that plagues much of the fantasy genre in its modern form, and why it has yet to fully escape the unfortunate stigma of juvenilia. The more you learn about history, and the more you are able to see through the anachronistic façades, the less imaginative conventional fantasy seems. We see this in action with language all the time: it’s almost expected these days to see gratuitous apostrophes as a desperate grab for an illusion of foreignness, not to mention unquestioning adherence to the “glottals ugly, labials pretty” phonetic valuation that Tolkien laboured to design. The same applies to maps.

(Ironically, children’s books are rather good at averting the implausibility trap, thanks to their embrace of figurative thinking. Witness the chart from The Phantom Tollbooth, featured at the top of this essay: like the novel it accompanies, it takes the abstractions of interior experience and projects them into real space.)

Indeed, as fantasy has entrenched itself in a self-propagating commercial set of norms, it has developed a reputation for being extremely conservative in form and politics. Jonathan McCalmont explained it well in his breakdown of the norms and values of “fat fantasy”: accessibility, immersion, and (less obviously) the safe escape of a reactionary aesthetics. All of this put together, along with the commercial considerations of how the publishing industry works, accounts for why a genre that benefits so much from cartography yields maps that are so extraordinarily literal.

For the commercial market, officially published maps are passive documents that serve as the easily readable evidence that the author thought things through. The maps embed our familiar expectations and unquestioned ideologies because the novels they accompany do the same. Perfectly content to assume that a map is the world, these fictions ignore the map as an instrument for grasping the world. Even if we step back from lofty cosmology and into functional lay-of-the-land geography, it is shocking how rarely fantasy maps explore the notion of the map as a visualization of a subjective and contingent worldview, a picture of an imaginary people’s collective weltanschauung.

Just to pluck one example out of thin air, in all of modern fantasy fiction, you will perhaps never find a map as fantastic as this:

The map contained in this 18th-century print by the Dutch publisher Pieter van der Aa is Giovanni Domenico Cassini’s Planisphere terrestre of 1696, which was meticulously assembled on the floor of the Paris Observatory from the most ambitious global survey of the time. (Working for Louis XIV assuredly had its benefits where funding was concerned.) The azimuthal projection centred around the North Pole looks warped to us today, accustomed as we are to the distortions we get from Mercator. But get past the initial unfamiliarity, and we can unpack all sorts of information from the structure of the map. We can see how it reveals its own construction in standard longitudinal slices, made possible by advances in astronomy and observational logistics; how the South Pole encircles the rim as the outer limit of human exploration.

Given the quiet proliferation of challenging fantastic fiction that is conscious of literary nuances in prose, one would hope that the cartography eventually catches up. Maps could do a lot more to dive into the perspective of an imagined land’s inhabitants, revealing how the people see their own world as well as their techniques and motivations for piecing that picture together. There is much room for subtlety in made-up maps, just as there is in lexicons, timelines, and family trees, and authors who do not feel bound to mainstream fat-fantasy conventions are in a unique position to explore the possibilities.

That said, it’s understandable why this hasn’t happened yet, and it has much to do with the specialization of labour along the pipeline of publication. With the rise of the e-book, prose content has become increasingly divorced from the totality of the book as an object, and authors are prose-module specialists more than ever.

One of the stories about Tolkien that has gotten a bit lost these days, I think, is the considerable control he exercised over his own work as a book designer who drew up his own cover art, runes and all. This isn’t to say that he had the visual talents of, say, William Blake, who remains the supreme English example of the all-in-wonder author/artist, but Tolkien’s relationship to his publisher is continuous with that tradition. This is the precondition that allowed a quirky thing like Thrór’s map of the Lonely Mountain to make it into the public sphere.

Outside of the vanity presses, this level of control is practically non-existent these days. Probably the best solution towards opening up the variety of maps we see is to develop the kind of author/artist collaboration for cartography that you see for graphic novels, where the writing doesn’t exercise any peculiar authority that confines the illustration to a secondary, supplemental role.

Authority

The question of authorship is a fascinating issue all to itself. Where exactly does a map (or indeed, an illustration of any sort) cross over from narration to interpretation? Is this solely a question of who is officially licensed to participate in the world-creation of an intellectual property? We can tear down the intentional fallacy all day from a theoretical point of view, and insist (quite reasonably) that the author has no better say than anyone else in the interpretation of the text; but in practical terms, reader communities who crave immersion derive their sense of a “canon” from officially sanctioned materials.

Consider this quandary: how would we mediate a discrepancy between Christopher Tolkien (who has privileged access to his father’s manuscripts and the legal authority of the Tolkien estate, and who delivered most of what we “know” about Middle-Earth in posthumous publications) and Barbara Strachey’s essential interpretive atlas, Journeys of Frodo? Do we rule in favour of the archival drafts, or do we side with the charts that are directly inferred from the final published text of The Lord of the Rings?

To return to what I said about omniscience early on, perhaps the demarcating factor that defines fictional cartography—what sets it apart from pictorial representations of real spaces—is the presence, however provisional, of the author as God. When we read non-fictional maps, we can perform a kind of analysis similar to what I’ve walked through here, looking at means of production, political agendas, and underlying worldviews; but it’s fairly unambiguous which plane of existence the cartographers reside upon.

Fictional maps introduce the complication of having, at minimum, two layers of authorship: the layer outside the text that has the power to dictate and reshape the world, and the layer that belongs to the reality of the world. It’s clear that the author is in the first and the characters are in the second, and that having the first speak for the second passes for a kind of ventriloquism or free indirect discourse. But these are not the only stakeholders in play. The “narrator” of the map, if it’s discernible as a separate voice, can belong to either layer or both. And once we introduce the other living participants—the readership and the publishing apparatus—determining who influences our perception of the fictional space becomes considerably trickier.

For one thing, it isn’t safe to take it for granted that immersion in a world means the same thing as immersion in the author’s mind, as if the goal of literature were some sort of telepathic telos of lossless communication. Among other problems, this attitude towards literary immersion as a matter of filling in the blanks has no way of dealing with deliberate ambiguity.

To see what I’m getting at, take a look at this:

This is the fold-out from The Discworld Mapp, a companion book to the Terry Pratchett series designed to the instructions of Pratchett and Stephen Briggs. It’s a fascinating map in view of what I’ve said thus far, not because of its superficial resemblance to the Cassini planisphere if you zoom out far enough, but because it achieves the look of an azimuthal projection via being a fiercely literal map in a two-dimensional rectangular coordinate system. Much of this, of course, is due to how the Discworld was conceived to be exactly that—a world where an azimuthal projection is a one-to-one circle-to-circle mapping, and therefore no transformation at all.

As far as companion publications go, The Discworld Mapp sounds par for the course—that is, until you look into how the map came into being. Stephen Briggs was a playwright who made a name for himself as an amateur Pratchett loremaster who adapted several books for the stage (and who, unlike Steve Vander Ark, knew better than to publish derivative works without legal sanction). When Briggs first decided to map the Discworld novels, beginning with The Streets of Ankh-Morpork, he did so with Pratchett’s support; but the project went against the author’s original insistence that Discworld was a fluid, aleatoric wander of the imagination that could not and should not be mapped.

Terry Pratchett’s opinion on cartography still appears on his official website:

There are no maps. You can’t map a sense of humor. Anyway, what is a fantasy map but a space beyond which There Be Dragons? On the Discworld we know There Be Dragons Everywhere. They might not all have scales and forked tongues, but they Be Here all right, grinning and jostling and trying to sell you souvenirs.

This is a very sensible position, and as much as I love cartography I am somewhat inclined to agree. There is something completely bounded about maps. We all know that the data they present is highly selective, as is also true for prose, with strategic omissions and gaps left for future exploration. Yet the boundedness of maps commits to an illusion of having reached a roughly finished state, as if everything inside has already been fixed and everything outside is still untouched and malleable.

From the perspective of someone involved in creating a world, particularly in a series that continues to emerge over time, a map intended to serve as an aid may also be a suffocating constraint. Mapmaking does not seem to permit carefully targeted ambiguity with the same flexibility as prose alone. With other forms of book illustration, one always gets the sense that the visual depictions could always be replaced or re-envisioned some other way. Maps exert a stronger form of authority: any improvements or revisions by readers or in future editions take place within the author’s borders as if they were immutable, objective truths.

It is a strange twist indeed that we are less liable to accept in fiction than in reality that cartography is a form of language: a medium for our perception of place, not to be confused with place itself. If there is a remedy for this, it may resemble the solution we developed for language, and take the form of self-conscious experimentation with maps as narrative voices—subjective, perspectival, and often unreliable. Literary writing deserves a literary map.

As for the alternative, let’s defer to what Lewis Carroll said in The Hunting of the Snark:

He had bought a large map representing the sea,
Without the least vestige of land:
And the crew were much pleased when they found it to be
A map they could all understand.

“What’s the good of Mercator’s North Poles and Equators,
Tropics, Zones, and Meridian Lines?”
So the Bellman would cry: and the crew would reply
“They are merely conventional signs!

“Other maps are such shapes, with their islands and capes!
But we’ve got our brave Bellman to thank”
(So the crew would protest) “that he’s bought us the best—
A perfect and absolute blank!”

Further reading

I am certainly not the first to gush over fantasy maps in the blog format. The most thorough look at fantastic cartography that I found while researching my post is “Beyond the Aryth Ocean”, a four-part series (1, 2, 3, 4) by Jason Denzel, who manages the biggest Robert Jordan community online and who therefore holds opinions that differ substantially from my own. He knows a lot of genre fiction I don’t, however, and his series on cartography is definitely worth a look.

Brian Sibley, who co-authored several map books with the eminent Tolkien painter John Howe, wrote a post about fictional maps with an inside perspective on the community of artists who work on these kinds of projects.

Of the dedicated cartography blogs on the Internet that cover fictional maps from time to time, two in particular stand out. The first is Frank Jacobs’ Strange Maps, which you probably know about already if you’ve followed some of the links in this essay or are interested enough in cartography to have made it this far down at all. (I also recommend the book.) The second is Jonathan Crowe’s The Map Room, which devotes equal attention to historical geography and modern surveying, and features an extensive subsection dedicated to imaginary places.

The Tumblr site entitled Fuck Yeah Fictional Maps really speaks for itself.

The Map History site maintained by Tony Campbell is an extremely comprehensive resource, mainly directed at serious scholars but certainly useful for enthusiasts of early maps.

For those of you with academic journal access through JSTOR, Imago Mundi is the history of cartography’s inexhaustible resource par excellence. The journal dates back to 1935, and skimming the tables of contents alone will give you a picture of how cartographic scholarship has developed over the past century.

Speaking of academic map history, one name to look out for when it comes to the literary turn in cartographic studies is J.B. Harley, best known for his collaboration with David Woodward on a compendious multivolume history of cartography that remains to be finished. (Harley and Woodward are both now deceased.)

Late in his career, Brian Harley became the radical subversive postcolonial deconstructionist of map historians, but he isn’t nearly as scary or loopy as that iconoclastic description makes him sound. In fact, the sort of critical theory that literary scholars overworked to the point of Sokalian fashionable nonsense seem positively fresh and sensible when applied to maps (see for yourself), and it’s rather puzzling that in the twenty-year history of the literary study of maps, fictional cartography has remained largely untouched.

There is a lot of rubbish being written about Avatar, and I freely admit to letting my own contribution stew in my draft box while I correct its pungent odour with the appropriate spice. But for the time being, I want to draw attention to one particular response to the film. Jonah Lehrer writes about the neuroscientific basis for cinematic immersion, and concludes:

What these experiments reveal is the essential mental process of movie-watching. (Other research has also emphasized the ability of stories to blur the difference between fiction and reality.) This doesn’t mean that every movie needs to be an action packed spectacle, just as Greenberg was wrong to suggest that every painting should imitate Pollock. But I think it helps reveal why Avatar is such a success. At its core, movies are about dissolution: we forget about ourselves and become one with the giant projected characters on the screen. In other words, they become our temporary avatars, so that we’re inseparable from their story. (This is one of the reasons why the Avatar plot is so effective: it’s really a metaphor for the act of movie-watching.)

When I think of films that act as “a metaphor for the act of movie-watching”, the director that instantly comes to mind is Alfred Hitchcock. And it so happens that the Hitchcock film most commonly read in this way also has a protagonist laid up in a wheelchair.

Psychoanalytic criticism has long thrived as a route into Hitchcock’s oeuvre, not least because he was familiar with psychoanalysis and popularized it in his 1945 film Spellbound, but also because his characters were marked with disorders, obsessions, and pathological instabilities of personal identity. You can see it in his choice of literary adaptations, chiefly Rebecca, where the second Mrs de Winter (Joan Fontaine) is consumed by the lingering household presence of the first; and in later films like Vertigo, where Madeleine (Kim Novak) “becomes” her suicidal great-grandmother through gazing at a painting in the museum (or so it would seem). It is Rear Window, however, that openly sets up its hero, L.B. Jeffries (James Stewart) as a passive voyeur behind a fourth wall that encloses an exterior apartment complex, where he sees fragments of his own life and relationship reflected back at him.

None of these interpretations are terribly hard to arrive at by yourself, but if you really want to get fancy, step back one level further and look for films where people watch Hitchcock. (We’re all familiar with the typical shot of a character sitting in a cinema, backlit by the beams of the projector, but pay attention to their faces and how they react to the film embedded en abyme.) The most recent example off the top of my head is Ang Lee’s Lust, Caution, where Wong Chia Chi (Tang Wei) gazes at Joan Fontaine in Suspicion as if looking into a mirror.

Of more interest from a sci-fi perspective—which will hopefully lead us back to Avatar—is how Terry Gilliam cues the final act of Twelve Monkeys with a scene from Vertigo, right when Madeleine Stowe takes after Kim Novak in turning her character blonde. A decade and a half after its release, Twelve Monkeys holds up today as one of the finest original pieces of sci-fi cinema (with all respect to its inspiration, La jetée), and it seems oddly prescient today in the face of James Cameron’s more conventional showpiece about a guy dumped into a tank to infiltrate and warn a society in which he is ultimately subsumed.

How, then, does Avatar differ from all these films? If the dissolution of identity is so key to its appeal, as Lehrer suggests, then why is it such an anomalous mainstream success?

The easy answer is that the kind of cortical stimulation Lehrer talks about comes equally from the overwhelming visuals of Cameron’s film, especially if you experience it in 3D. But that dodges the very questions of story and theme that Lehrer wants to raise. The thematic answer, as I see it, is that Avatar plays it safe: completely unlike the films of Hitchcock, Lee, and Gilliam, it never dares to convey the madness of a dissolved identity or bother its audience to consider the schizophrenia of immersing itself in film. On Pandora, a world where USB ponytails plug into any living thing, bodily escape is free of risk. The film doesn’t spit us out and force us to look at ourselves; it does the opposite instead, encouraging us to enjoy what Lehrer calls “a pretty nice cognitive vacation.”

What interests me about Lehrer’s article, though, is not his assessment of Avatar as a film but his neurological approach. I have long wondered about the attraction, almost the fashionability if you will, of cognitive science in the criticism of the arts. In literature, materialist explanations based on quantifiable observations about how audiences react to what they read were the natural-born children of evolutionary psychology and reader-response theory. (For a better look at this, read the OnFiction blog.) But I expect that cinema is where cognitive science will pervade scholarship in the twenty-first century as psychoanalysis did in the twentieth.

In the history of thought, there is a substantial continuity between the two. Maps of brain activity now occupy the place of psychoanalytic concepts like transference, repression, or the gaze, but the premise of psychological criticism remains unchanged. And while neuroscientists will be quick to remind me that their quantifiable, testable claims are a far cry from all that Oedipal fluff, it is not at all clear to me that the epistemological status of criticism—the task of applying the theory to our understanding of how film works—is made to be any different. The question of whether cinema produces certain audience reactions would appear to have more validity, but how it does so is the same framing question as before, and for reader-response theory that may always be out of reach. This does not mean cognitivist claims are invalid, only too reductionistic to be complete.

Last June, the film scholar David Bordwell wrote a stupendous review of the cognitive turn in film studies: what it is, where it comes from, and where people are taking it. I would not hesitate to call it essential reading for anyone interested in science and arts criticism. On psychoanalysis, Bordwell writes:

[Cognitive film studies] is naturalistic. The explanations it mounts try to fit in with current understanding of human capacities as analyzed by the social sciences. That entails that psychoanalysis, another mentalistic theory of human action, has not on the whole proven a source of reliable explanations. Some cognitively inclined researchers would add that psychoanalytic inquiry has been fruitful for pointing to areas of behavior that answer to naturalistic investigation.

Lehrer’s piece on Avatar fits squarely in the cognitivist mould. But the tradition of psychoanalytic criticism is not to be ignored; to do so would be to throw the baby out with the bathwater. In order to appraise Lehrer’s informal hypothesis—that Avatar draws its power from being a film about the experience of watching film—any experimental study of whether this is the case must account for the other films that are said to function in the same way. An explanatory account of cinema should ideally tell us about what led us to arrive at interpretive accounts.

Lastly, should cognitive studies produce a limited theory of film, as I suspect it must, we should remain open to the possibility that material disciplines apart from studies of the mind will clue us in to the structure and interpretation of fiction. Neuropsychology’s centrality as the site of consilience is something of a fashion, and one that we owe to the place of evolutionary ideas in the popular consciousness, over and above the public awareness of other sciences. I look forward to seeing whatever comes next.

In brief: This collection of a dozen variegated essays—some broadly accessible, others strictly for the interest of literary scholars—is a grab bag of erudite criticism that serves as thorough sampling of Frye’s one-man theory show. The academic pieces, which attempt to deduce overarching mythic cosmologies from the poetic output of writers such as Milton and Blake, are an ample demonstration of Frye’s method. Far more compelling, however, are the pieces that argue for the continued relevance of the imagination following its dislodgment from the objective world of science and history.

(The Wednesday Book Club is an ongoing initiative of mine to write a book review every week. I invite you to peruse the index. For more on Spiritus Mundi, keep reading below.)

Northrop Frye is possibly literature’s most readable theorist, and definitely so, should we confine ourselves to the critical morass that is the latter half of the twentieth century. I attribute this to two factors.

First, there’s his predilection to file the entirety of Western literature into neat analogical boxes like a Victorian lepidopterist, where everything fits into a grand ur-structure of isomorphisms between the four seasons, the genres of drama, the phases of man, the Biblical mythos, and everything else. There’s a clarity to Frye’s writing that doubtless originates in the orderliness of his thought, in stark contrast to the pluralistic mess of postmodernism at its most extreme, where everything means anything and therefore nothing.

Second, there’s his singular insistence on departing from the critical orthodoxy of “schools of thought” derived from preexisting -isms, and developing an independent theory of literature that grows out of the empirical data before asserting its own autonomy—a theory that is to literature as mathematics is to physics, or as chord-scale theory is to music. Reading Frye leaves an encouraging impression that one is, in fact, learning something; again, not something that can be said for your average theorist.

Most of this was explained better in Anatomy of Criticism, a theoretical masterwork that qualifies as desert-island reading for the literary scholar. But there is something to be said for the versatility of a thinker who just as happily boils down his discipline into the basic introductory questions of why we bother with it at all, as Frye does with the utmost commitment to accessibility in the Massey Lectures published as The Educated Imagination.

The twelve essays in Spiritus Mundi cover the spectrum between the two extremes, the Northrop Frye who muses about grand mythological archetypes in case anyone is incidentally listening, and the Northrop Frye who stands as a plainspoken advocate for literary criticism as a dignified, valuable discipline. They are cordoned off into three parts:

1. “Contexts of Literature”—essays concerning the place of literate culture in the intellectual and political environment of academia, the enduring relevance of books, and the coexistence of mythological and scientific ways of thinking.
2. “The Mythological Universe”—studies in genres as diverse as literary criticism, charms and riddles, masques, and grand unified theories of history.
3. “Four Poets”—readings of Milton, Blake, Yeats, and Stevens that attempt to deduce a cosmology, or symbolic universe, from the works of each poet.

It’s hard to summarize any of the essays individually, as Frye has a tendency to meander from one thought to another instead of pursuing a single line of argument at a time. He makes it clear that he doesn’t believe in the “position paper” as a format, as he finds that it enslaves the study of a work of literature to demonstrate an ideology that has already been assumed. Indeed, it is quite easy to get very lost in the essays that rely on close poetic analysis, not because of any obscurity in Frye’s writing, but because he presumes a thorough holistic knowledge of the works he is analyzing in isolated parcels. You might think you know Milton, but then you see Frye go on for pages about a lesser-known work like Samson Agonistes and all you end up absorbing are the occasional conclusive paragraphs that jump out with an “aha”.

So there’s not much I can say about the arguments themselves: I don’t have the literary (or, in many cases, theological) background to follow every logical step and refute them if necessary. What’s interesting to see here is how a writer can dedicate himself to clarity in a syntactic, lexical sense, and still produce essays that are distinctly academic. There remains an irreducible difference between academic and popular writing: the prerequisite of pure knowledge. And as much as Frye would like criticism to be a study that has something to show for itself as a closed system, the evidence underlying his grand mythological constructions—the close analysis of the literature itself—often only makes sense to the degree you are aware of the literature.

But the essays that take a more holistic approach, predominantly those in the first part, are sublime. Some of them are a product of their time, but are food for thought nonetheless. In “The Renaissance of Books”, Frye shows off his healthy conservative scepticism towards the oft-announced death of books at the hands of emergent media, contending that the permanence of written documents (contrary to the “linearity” suggested by McLuhan) makes them the enabling artifacts of democracy. Well, sure, but what do we make of that argument today, in the age of mass digitization of texts—where we have impermanent digital media that are simultaneously more accessible to the public? We’re still at least a generation away from assessing what effect digitization will actually have on our cultural memory, but it currently looks as though Frye’s defence of the book could apply as well to the e-book.

Equally fun is “The University and Personal Life”, Frye’s thorough dissection of late-sixties hippie leftism that exposes it as a far and purposeless cry from the old guard of intellectual social movements, which were fundamentally driven by a teleological vision of history. Frye has a taste for grand theories of history, even if they have been thoroughly discredited in practice: he repeatedly insists that mythologies never become wholly obsolete, since they remain interesting as mythologies—a view he takes of Oswald Spengler’s four-seasons model of the rise and fall of civilizations in “Spengler Revisited”.

One of Frye’s contentions in Spengler’s defence is that his critics, who showed with the obvious means that Spengler ended up being flat-out wrong, never proposed any elegant models of civilization as a replacement. This is understandable in light of Frye’s viewpoint that criticism should learn a few lessons from science, where the refutation of a theory leaves a vacuum that can and must be filled by a new theory that offers a better explanation accounting for the evidence of its predecessor, and then some. However, since the publication of Spiritus Mundi in the 1970s, postmodernism has driven much of literary “theory” off into a different corner entirely—one that rejects grand narratives outright, and aims precisely to take them apart without constructing anything in their place. I do not imagine Frye would be very pleased with the state of affairs today.

The most compelling essay of the lot is “The Times of the Signs”, a mind-blowing address delivered to the Royal Society of Canada in 1973 to mark the five hundredth anniversary of the birth of Copernicus. If one were to compile an introductory package of Northrop Frye essays that offer a grand tour of the full spectrum of his thought, this would surely be among them.

Here, Frye illustrates, through a miniature history of Western poetry, the effect that developments in the natural sciences had on the imagination—primarily, but not exclusively, the realignment of our place in the world from a Ptolemaic centre to a heliocentric orbit. What did it do to a Christian consciousness? Or a Renaissance consciousness? And what the heck is Galileo doing in Paradise Lost, anyway?

Frye’s romantic schema, where our natural knowledge of the world that is coexists with an imaginative culture expressing the world as we would like it to be, is as mature an assessment of the interaction of science and myth as one is likely to find. We may be living in times when the common poetic impulse is to retreat into a vehemently unscientific symbol-set—observe the resurgence of astrology—but that wasn’t always so, and it hopefully won’t always be that way. In an apathetic environment, our imaginative response, which takes the form of our cultural production, is the source of human ideals and desires.

As a complete volume, I do not suggest Spiritus Mundi as an introduction to what Frye is all about: a student of literature is better served with Anatomy of Criticism, and a layperson with The Educated Imagination. But it is in Frye’s short works that the extremes meet in the middle, and one gets to witness a fine demonstration of his method, which takes collected morsels of literature as evidence for a grand pattern to the universe of human dreams.

That said, I have to hold him to account for his recent post on Futurismic, “How to Dismantle the Wall Between an Author and Their Work”, for its egregious misrepresentation of the intentional fallacy. I recommend that you read the full text; despite its flaws, it is an ample demonstration of when it is valuable to take the author into context when reading a work of fiction. Permit me to quote some of the boldfaced declarations:

1. “[The intentional fallacy] is a denouncement of the idea that one can ever infer what an author was intending when he wrote something and [the affective fallacy] is a rejection of the idea that the effect that a text has on a reader is any kind of basis for thinking about it.”
2. “There is no ‘logical’ reason why one should not interpret a book in terms of what one knows about the life of the author.”
3. “[...] if one cannot infer from a book what an author means then one cannot infer from anyone’s words or actions what they are thinking.”
4. “The inner life of an author before they write a book is as much a part of that book’s meaning as the words on the page.”

Unfortunately, statements (1) and (2) rely on a disingenuous summation of what the intentional fallacy is and how it is put into practice. Furthermore, an argument for the validity of biographical criticism is not equivalent to an argument for its primacy or necessity. The latter opposes the thrust of the intentional fallacy, but the former does not.

1.

First of all, the concept of the intentional fallacy, be it in the original Wimsatt/Beardsley essay or any of its other permutations, says nothing about the inference of the author’s intent. In fact, the formalist analysis advocated by the New Critics has no interest in the author’s intent at all, insofar as that intent produces a specific direction for how to interpret the text. McCalmont cites the Wikipedia article on the intentional fallacy, but I’m not sure he read it; the article clearly outlines Wimsatt and Beardsley’s stratification of empirical literary evidence into three categories—one of which is contextual and possibly biographical.

From the Wimsatt/Beardsley essay (emphasis added):

There is a difference between internal and external evidence for the meaning of a poem. And the paradox is only verbal and superficial that what is (1) internal is also public: it is discovered through the semantics and syntax of a poem, through our habitual knowledge of the language, through grammars, dictionaries, and all the literature which is the source of dictionaries, in general through all that makes a language and culture; while what is (2) external is private or idiosyncratic; not a part of the work as a linguistic fact: it consists of revelations (in journals, for example, or letters or reported conversations) about how or why the poet wrote the poem‑to what lady, while sitting on what lawn, or at the death of what friend or brother. There is (3) an intermediate kind of evidence about the character of the author or about private or semiprivate meanings attached to words or topics by an author or by a coterie of which he is a member. The meaning of words is the history of words, and the biography of an author, his use of a word, and the associations which the word had for him, are part of the words’ history and meaning. But the three types of evidence, especially (2) and (3), shade into one another so subtly that it is not always easy to draw a line between examples, and hence arises the difficulty for criticism. The use of biographical evidence need not involve intentionalism, because while it may be evidence of what the author intended, it may also be evidence of the meaning of his words and the dramatic character of his utterance.

Part of the problem is that Wimsatt and Beardsley do not provide the most lucid explication of what it is they oppose, even though they coined the term: they redefine “literary criticism” within certain boundaries, displace the external and contextual evidence into “literary biography”, and regard the two as valid but separable fields of study. Keep in mind, also, that they openly admit the instability of the distinction. The label of the “intentional fallacy” applies not to the existence of biographical or contextual criticism, but to their use as an overriding interpretive authority. The New Critics do not argue that one must not pull evidence from outside the text. They claim, instead, that one need not pull evidence from outside the text, and still produce a valid piece of criticism. Unfortunately, the semantic fogginess of the term “criticism” as distinguished from “biography” clouds the issue.

A much better explanation can be found in Northrop Frye’s “Polemical Introduction” to Anatomy of Criticism: Four Essays:

The poet may of course have some critical ability of his own, and so be able to talk about his own work. But the Dante who writes a commentary on the first canto of the Paradiso is merely one more of Dante’s critics. What he says has a peculiar interest, but not a peculiar authority. It is generally accepted that a critic is a better judge of the value of a poem than its creator, but there is still a lingering notion that it is somehow ridiculous to regard the critic as the final judge of its meaning, even though in practice it is clear that he must be.

[...]

The poet speaking as critic produces, not criticism, but documents to be examined by critics. They may well be valuable documents: it is only when they are accepted as directives for criticism that they are in any danger of becoming misleading.

2.

McCalmont claims that the intentional fallacy is not a “fallacy” at all, and that the use of the word is “a rhetorical tool, a means of imbuing what are ultimately arbitrary fashion-statements with the authority that comes from logic.” He later states that “the taboo against delving into an author’s brainspace is not a matter of logic but of fashion” and that (correctly) “flouting that very taboo has created many great works of criticism.” But to dismiss the isolated reading of intrinsic textual evidence as ungrounded fashion is a case of assertive hand-waving.

Once again, the problem here is the false equivalence of biography and intention (a line that Wimsatt/Beardsley didn’t tread very carefully themselves). No, an appeal to authorial intention is not anything like a formal propositional calculus, but it is a specific case of a well-recognized fallacy in rhetoric: the argumentum ad verecundium, or the appeal to authority. I would provide an original example, but McCalmont has handed me one on a silver platter.

In a stunning contradiction, McCalmont cites Elaine Radford’s evisceration of Orson Scott Card’s Ender’s Game as a fascist apologetic as an ideal example of the value of biographical context, especially when juxtaposed with Card’s ideological sympathies—most recently, his demonstrated homophobia. (I should note that the Radford piece is actually diligent about drawing on textual evidence and levelling an indictment of the novels, and not so much Card himself; while for all I know, Card believes in some form of inheritable genetic determinism, I’m not about to draw any conclusions about him as a person based on the genetic determinism that pervades his work. However, McCalmont is quite right in pointing out that Card’s “love the sinner, hate the sin” endorsement of Proposition 8 does not speak in his favour.)

Radford does not publish the text of Card’s rebuttal for copyright reasons, but the very fact that a rebuttal exists tells me that Card denied, to some measure, the claim that Ender Wiggin is a stand-in for Adolf Hitler. To me, this is an excellent demonstration of textually derived criticism in action. That Radford’s piece presents a coherent and defensible thesis against the stated will of the author is a triumph for the critical method that Wimsatt and Beardsley advocated from the beginning. Nobody could reasonably accuse Radford of committing the intentional fallacy; to the contrary, the commission of the intentional fallacy would involve discarding the Radford piece on account of Card’s denial (again, assuming it was a denial). As for Jonathan’s observation, I find that it sheds more light on the link between Ender’s Game and Orson Scott Card than the one between Ender Wiggin and Adolf Hitler; if we were to pigeonhole it, it would belong to the class of literary biography that Wimsatt and Beardsley permit.

Once more: the concept of the “intentional fallacy” is not here to exhort that one must ignore contextual knowledge, but that an interpretation that divorces a text from its author is a valid subclass of literary discussion. The validity of biographical criticism, which McCalmont amply demonstrates, does not preclude intrinsic, text-driven readings from being an equally valid scope of analysis. What Orson Scott Card says about Orson Scott Card may well be a defensible interpretation in its own right, but it is never the “true” interpretation by virtue of who said it—and it never, ever imposes a restriction on critical activity.

3.

I now return to statement (3) above, and quote it in full:

One could argue that if one cannot infer from a book what an author means then one cannot infer from anyone’s words or actions what they are thinking. Indeed, it is no accident that the intellectual tradition founded by thinkers such as Derrida and Barthes (who famously proclaimed the “Death of the Author“) questions not only the idea of an author’s interpretation being the ‘correct’ one, but also the idea of ‘truth’ itself as anything more than a social construct.

Admittedly, there does exist a sound basis for hysteria over the apparent moral poverty of reading texts out of context, in the Derridean mode. The tragedy of postmodern theory is that it often combined textual readings with the belief that history is all a matter of representation, with no accessible basis in empirical truth; and in doing so, relinquished any moral high ground for the refutation of, say, holocaust denial. (I’ve written on this before.) Famously, when Paul de Man’s former association with a collaborationist newspaper came to light in the 1980s, Derrida came to de Man’s defence with an ill-conceived deconstruction of his writing, attempting to show that the writings published in an anti-Semitic rag actually advanced the Jewish cause if you read it in a certain, contorted way.

(Honestly: all Derrida needed to do to fend off those who wished to discredit de Man’s valuable contributions to literary theory was say, “That’s ad hominem.” But deconstruction doesn’t play that game, you see.)

We should realize, however, that the moral poverty of ignoring empirical evidence when it comes to the construction of historical or scientific narratives—which must, as a matter of methodology, have a basis in the externally observable—has no analogue in the reading of fiction. When we read something that is declared as a work of fiction, we accept that it is by its very nature a constructed text that has every right to defy or ignore reality.

Thus, the self-contained reading of the text has no ill implications for how we read people or interact with society at large, as McCalmont would like us to believe. A work of fiction has no responsibility whatsoever to replicate the views of its author. Robert Heinlein was free to advance a libertarian ideal in The Moon Is a Harsh Mistress and defend a militaristic system of civic participation in Starship Troopers without coming off as a discombobulated contradiction of an author, precisely because we need not equate one book or another with his real-world political views.

4.

Now that I have shown how Jonathan McCalmont’s defence of biographical criticism refutes not the intentional fallacy, but a straw man of the same, I must turn my attention to what he does advocate—namely, the essential importance of “the inner life of an author before they write a book” to the book’s meaning.

While biographical evidence can certainly make for a good interpretation of a work of fiction, I take issue with the claim that the author’s life becomes anything intrinsic to the book’s meaning—especially where speculative fiction is concerned. The assumption that novels are necessarily representative of their authors’ beliefs is a troubling one; often, it is simply not true.

One case study we can look at is the composition of fantastic fiction by atheist authors: two that come to mind are Philip Pullman and Salman Rushdie. (Yes, I’m aware that Rushdie is an outsider to the sci-fi/fantasy tradition, whatever that is, but he writes his own brand of magical literature, and I feel perfectly justified in including him here.) Both of them are on the record as ascribing to a godless, material universe.

In Pullman’s case, the author’s declared atheism has coloured many a reading of His Dark Materials—particularly the last volume, The Amber Spyglass—as a fundamentally atheistic text. Even as perspicacious a reader as Michael Chabon was guilty of this, as I discussed in a previous post. True, the His Dark Materials trilogy is unambiguously opposed to organized religion, denouncing the project to sever children from sin and celebrating the deposition of God, the Authority. But to claim that Pullman’s books are intrinsically atheistic as an extension of the author’s beliefs is to wilfully ignore its Gnostic cosmology and its frequent recourse to magic, as in its reimagining of dark matter as an element of fantasy and its depiction of a scientist, ostensibly the serpent of Eden, who finds truth in the I Ching. The case for a naïve biographical reading of the author’s theological agenda pales in comparison to the rigour of reading the books for what they actually depict.

Similarly, Rushdie’s “apostasy” from his Islamic heritage and his, ahem, controversial satire of the birth of Islam in The Satanic Verses are connected, but not congruent. There is no reason to believe that his renunciation of Islam on atheistic grounds is equivalent to the anti-authoritarian case in The Satanic Verses—a book that brims with the supernatural, with its magic lamps, men surviving falls from exploding aircraft, and heads swapped with those of donkeys. As with Pullman, the biographical knowledge that the author is an atheist allows us to draw a connection between the author and the book, but it is only responsible for a reader to acknowledge where that connection fails.

As another example, I point to Ursula K. Le Guin’s The Dispossessed, which I recently reviewed. The elegance of the book lies in its undecidable debate between the anarcho-communist utopia of Anarres and the capitalistic progress of Urras. A perusal of online reader reviews—particularly the negative ones—reveals an equal measure of mudslinging by those who think it’s anti-communist tripe, and those who think it’s pro-communist tripe. The magic of the book is that it reveals the flaws of either system, as well as their necessary and inescapable shackles. Certainly, it would be limiting to argue that one side “wins” over the other on the basis of what Le Guin actually believes. Novels that explore a complex range of ideas and give them a fair hearing make it difficult, as well as unsatisfying, for one to suppose that one particular side has the last word.

5.

McCalmont claims that biographical criticism is on the decline, a claim that I find curious. The trend in literary studies after deconstruction, and especially after postcolonialism and recent developments in gender theory, has strongly tended towards cultural studies and theories of identity—which, at its worst, have descended into a despondent swamp of narcissism.

However, I am unsurprised that SF criticism has moved towards the formalist, author-free end of the spectrum: after all, it is in SF that fiction authors have made the deepest inquiries into the information age, and whether we can dispense with the whole notion of intellectual property. We see this displayed most prominently in cyberpunk and its derivatives, be it William Gibson or Cory Doctorow, but also in the increasing sense of entitlement and ownership prevalent in fan-fiction communities. There is no question that this is evidence of a rising belief in the death of the author, on top of McCalmont’s keen observation that reviews in SF magazines are predominantly author-shy.

While I agree with McCalmont’s concern that the utter lack of biographical criticism is unhealthy for what should be a diverse intellectual climate, I think it is important to remember that the speculative genres, by the very fact of how they exercise the imagination in a way that defies the mimicry of the real, rely on the ability of authors to get beyond themselves and write about subjects of more universal scope. There is no denying that their personal histories inform their imaginations, as how an awareness of drug addiction informs the novels of Philip K. Dick; nevertheless, science fiction and fantasy are essentially incapable of acting as thinly veiled autobiography. So far as I am aware, there is no Jack Kerouac of sci-fi. (Let’s ignore William S. Burroughs; I would hardly call Nova Express thinly veiled.)

Genre writers imagine not only characters, but whole societies and sets of counterfactual laws. If we can read their books apart from their personal histories, we should see that as a virtue—as evidence of complexity and universal scope. The staying power of mythology lies in its capacity for escaping its socio-cultural origins. Mythmakers and myth-readers alike should keep that in mind.

I jumped into the comment-box fray myself, but—in an ample demonstration of exactly what I was saying—others in the discussion phrased the same ideas in more succinct and elegant terms.

Rich Puchalsky:

But basically I agree that if knowledge is to be knowledge, it has to get past its original writer. If you can only understand the concepts in Derrida by reading Derrida, then you’re not reading him for knowledge, you’re reading him as a literary text.

Luther Blissett:

At their best, [good summaries and guides] provide us with the foundation to read better when we turn to the original text. Even when the roadmap is over-simplified or not quite right, I find that students can question the map more effectively having used it than they could if they only had a first reading to go on.

Ironically, that always seemed to me to be the point of Derrida’s work: to provide a reading—not a reduction but a distillation—of a certain aspect of a philosophical text, so that when we return to the foundational texts—Plato, Rousseau, Hegel, Heidegger, Descartes—we do so with fresh eyes, standing on the shoulder of a giant, so to speak.

It’s odd that Joseph is defending the host/parasite binary in a defense of Derrida.

I also recommend Andrew Seal’s excellent response at Blographia Literaria. Seal makes a crucial distinction that I have been advancing for ages: that a call for more transparent philosophical writing is not populist pandering, but an urgently needed reform for the sake of maintaining a healthy intellectual culture. An excerpt:

Derrida, then, becomes nothing more than a genial literary critic of his own corpus, writing to and for Derrida enthusiasts. This kind of flight into the personal is precisely the move conservative critics take as a sign of the weakness of post-structural thought. Whether or not this is fair, it is highly important to question the value of such a move if it ends up inevitably sticking us with charges of “meaninglessness,” “relativism,” and “charlatanry.” This is the bedrock problem of the mischaracterization of post-structuralism, gender/queer theory, critical race theory, post-colonialism, etc.—the reactionaries listen to us denounce repeatedly the notion of an integrated, coherent, autonomous subject, and then we say something like “well, Derrida didn’t mean for everyone to understand his work—his books are intimate and personal writings for people who take the time to really get to know him.” I’d throw my hands up too, if I weren’t typing.

When some defender of theory does make one of these appeals to the “personal,” what they’re really doing is making an appeal to the hieratic: if you’re not an initiate, you shouldn’t be paying attention. If you haven’t taken the time to make Derrida “personal,” you don’t have standing in this field.

Let me be clear: I’m not attacking this move on populist grounds. I’m attacking it on elitist grounds: this is an incoherent and unstable elitism, one more dangerous to the elites than to the masses.

Naturally, this won’t amount to a hill of beans to the apologists who deny the existence of an empirical reality outside discourse and have no use for the positivist prejudices of the hegemons who have the nerve, the nerve to quantify things. Well, it’s their loss.

As someone who studied both the hard sciences and literary theory in considerable measure, I am compelled to make a few quick remarks of my own. If theory in its present incarnation is indeed collapsing, I lay the blame on what I like to call the Two O’s (ooh!): overextension and obscurantism.

Overextension

Literary studies have a reputation as a rotting graveyard where discredited ideas go to die. Cognitive psychologists chased Freud out of town, and economists did the same to Marx. Lo and behold, pick up a stack of undergraduate English papers and you can wallpaper your study with Freudian-this and Marxist-that.

I don’t see this as a problem. In fact, I find it perfectly understandable. The magic of literature is that you can remove or replace the external, referential reality signified by the words on the page, and you still have something worth studying. So it’s no surprise that homeless theories set up camp under literature’s big tent as relevant case studies of how we construct representations of the world and understand it through metaphor—never mind if those representations are “true”. (This is also my position on religion.)

On top of that, it is imperative for students to develop some fluency in the jargon of the history of ideas, whether those ideas remain current or not, because formerly empowered discourses served as the basis for existing work in the discipline. If you know your Freud, you will never run out of things to say about Alfred Hitchcock.

Why, then, are some of the most reliable workhorses falling out of fashion? I pin it on inflated delusions of political relevance—a bubble now ready to burst with disillusionment.

There is no doubt that what the study of literary artifacts tells us about the construction of signs has an important role to play in our understanding of how language shapes the documents—scientific, historical, or otherwise—that we ordinarily do not regard as “literary”, viz., the reader has no contractually agreed-upon disposition to suspend belief. The trouble occurs when theorists, intoxicated by the Whorfian extremities of linguistic relativity, mistake it for the irrelevance of the observable world. At that point, the applicability of the philosophical tradition cowering under the literary umbrella becomes suspect.

As a consequence, we are now stuck with a tenured radical left that no longer challenges hegemonic power structures with empirical inquiry and cool logic, but regards those trusty instruments of free thought as part and parcel of a subversive imperial project to eradicate other cultures under the socially constructed banner of global development. However noble the intentions of these theorists may be, this is a position of grave moral poverty, and it will not stand.

Admittedly, “scientific approaches to humanity” also have a record of overextension. There is a lot of bad science out there, most of which involves the advancement of spurious social claims unsupported by the evidence that actual scientific rigour would demand. (This is equally the fault of irresponsible journalism and a persistent incentive to secure grant funding by exaggerating the applications of trivial results.) But scientifically motivated excesses like the Nazi regime’s specious taxonomy of race have a built-in fail-safe mechanism: they leave themselves open to summary rejection by scientific means. The misapplications may be perverse, but the methodology is sound.

Now consider this: as soon as you ascribe to the total relativism of much postmodern theory and its insistence on reading texts against themselves, and are so bold as to apply those lessons to every text in sight, you surrender any and all intellectual grounding for discrediting Holocaust denial. How d’you like them Godwins?

Obscurantism (I)

Theory is hurt, more than anything, by the stubborn refusal of many theorists to express their ideas clearly.

One of the things you learn when you read about the construction of freestanding towers like Burj Dubai is that the higher you build a structure, the stronger the lower levels must be. The first story must support more weight than the second, and so on. While this is completely intuitive, it’s not a consideration you easily infer from your everyday experience with plastic LEGO towers or wooden Jenga towers, where the blocks are so light that compressive force will simply never be an issue. Many theorists in the humanities seem not to have learned it.

As it happens, the principles of civil engineering apply to ivory towers as well as they do to concrete ones. The measurable decline of theoretical buzzwords in the JSTOR corpus is, I suspect, the direct repercussion of these disciplines’ architectural habit of building upwards, not outwards. I suppose you could say they start teetering.

You can see this reflected in the syntax and morphology of the obscurantists, which thrive on concatenation. Notice how readily these writers stretch every clause into unparsable verbal spaghetti and enrich their lexicon by appending suffixes upon suffixes (popular culprits include -ity, -ism, and -ize). Why say “problem” when “problematic” [n.] will do? Corollary: why say “problematic” [adj.] when “problematical” will do? And if the difference is important, why explain it?

James Miller provides a comprehensive overview of the problematic problem in “Is Bad Writing Necessary?”, where he depicts the debate over clarity as an ongoing battle between George Orwell and Theodor Adorno. I will not discuss all of the key passages here, but I do wish to focus on the argument from “technical expressions”, a common argument from Adorno and his obscurantist acolytes (among them, Judith Butler). In Miller’s paraphrase:

“It avails nothing ascetically to avoid all technical expressions.” Serious writing sometimes requires jargon: the sorts of terms that circulate in any highly evolved science. It would be absurd to demand of a physicist like Einstein or Bohr that he write in prose intelligible to the layman. In Adorno’s eyes, German philosophy has some claim to the title of science, and it certainly has evolved its own glossary of technical terms. Just because a bunch of American yahoos have never read Kant and Hegel is no reason to abandon an exacting vocabulary.

[...]

Like Adorno, today’s critical theorists have steeped themselves in the vocabulary of German philosophy, from Kant and Hegel to Husserl and Heidegger, augmented by an infusion of terms from more recent French philosophers, especially Foucault. Jargon that is intolerable to a general reader is not only a source of power; it is also a convenient shorthand for conveying the results of inquiry in most academic disciplines, from physics to sociology, and not excepting feminist theory, literary criticism, and cultural studies.

The flaws in the “technical expressions” argument are manifold.

First of all, it only pertains to vocabulary, and says nothing of tangled syntax.

Furthermore, in science we study not Einstein, but relativity; not Bohr, but the atom. It doesn’t matter if we write E = mc² or $ = &@². You could rewrite every convention of arithmetic to read like the recursive nests of Polish notation, where Einstein’s equation can be expressed as the truth of the statement (= E (^ (* m c) 2)), and maybe replace the numeral “2″ with a bug-eyed squiggle while you’re at it. It makes no difference, because you can follow your own course of independent observation and derivation and arrive at the same proportionality of mass to energy. If we were not so keen on giving credit where credit is due, we could tear up Einstein’s original papers and rebuild them from scratch. His prose is irrelevant. The derivation exists outside of its textual expression: it is logical, not rhetorical.

In philosophical writing, a Husserlian or Heideggerian specimen of rhetoric itself becomes an object of study to be decoded; we flatteringly call these “primary sources” and insist upon consulting them in case any of the meaning has been misplaced or lost in paraphrase and translation, which happens all too often. Why? Because the meanings of overdetermined words are unclear and cannot be independently traced. We should be concerned that these writings require decoding at all. The study of philosophy, in the continental tradition especially, is thus inherently person-centric and inherently susceptible to fallacious arguments from authority.

The specialist terminology in scientific disciplines is consensual and well defined. Some people are grumpy about the criteria for distinguishing planets from dwarf planets, but they do not defy the standard; they acknowledge the presence of authority where it exists. I see no evidence of such a consensus in philosophical jargon, where you can look at this winning paragraph by Judith Butler…

The move from a structuralist account in which capital is understood to structure social relations in relatively homologous ways to a view of hegemony in which power relations are subject to repetition, convergence, and rearticulation brought the question of temporality into the thinking of structure, and marked a shift from a form of Althusserian theory that takes structural totalities as theoretical objects to one in which the insights into the contingent possibility of structure inaugurate a renewed conception of hegemony as bound up with the contingent sites and strategies of the rearticulation of power.

… and, all things considered, have no idea whether the “power” she speaks of is Nietzsche’s, Foucault’s, or somebody else’s entirely. Philosophers have a nasty habit of appropriating words that are already loaded with conventional meaning. Physicists make up words like “quasars” or at least string them together in unconventional compounds (“strange attractors”, “dark matter”); when they use common words like “wave” and “particle”—and yes, “power”—the contextual definition is unambiguous.

Of course, I would not expect philosophers who chain themselves to radical ideology (exercise: define “radical”) to ascribe to any consensus, as they regard all reified, conventional meanings to be unjust instruments of power that mask their origins. In truth, it is the pollution of meaning that hides the origins of signification.

Obscurantism (II)

What is clarity in writing? It is when the medium of language recedes as far as it can to ground the figure of the message. (B-b-but sir, McLuhan said the medium is the message! Yes, he did. I’ll get to that in a second.) This doctrine only makes sense when the ideas in the message’s “content” have an independent referential existence. This isn’t to say that the referred-to object, or signified, must “actually” exist in an empirically observable sense: what I mean is that the signified concept should be independent of its notation.

In short, clear writing is translatable. As Douglas Hofstadter explains in Le Ton beau de Marot (which I reviewed last week), language is at its least translatable when the medium is so tightly wedded to the message that the message is a self-referential statement about its containing form. The medium is the message, but different messages “are” their media to different extents. The tightest binding occurs when the medium is the object to which the message refers, and the basis on which it depends.

Obscurantist theorists write in a manner that is deliberately constructed to defy reduction, paraphrase, and translation. This is because many proponents of the postmodern school discard the idea that there is anything outside language to which we can refer. To define their terms and condense their ideas, they say, is to “do violence” to the text and play into the hands of the hegemonic structures they wish to subvert.

Consider Judith Butler’s hysterical self-victimization in “A ‘Bad Writer’ Bites Back”, a piece in The New York Times to which I half-jokingly refer as the only thing Butler has written in the English language:

[...] why are some of the most trenchant social criticisms often expressed through difficult and demanding language?

No doubt, scholars in the humanities should be able to clarify how their work informs and illuminates everyday life. Equally, however, such scholars are obliged to question common sense, interrogate its tacit presumptions and provoke new ways of looking at a familiar world.

[...]

If common sense sometimes preserves the social status quo, and that status quo sometimes treats unjust social hierarchies as natural, it makes good sense on such occasions to find ways of challenging common sense. Language that takes up this challenge can help point the way to a more socially just world. The contemporary tradition of critical theory in the academy, derived in part from the Frankfurt School of German anti-fascist philosophers and social critics, has shown how language plays an important role in shaping and altering our common or ”natural” understanding of social and political realities.

But what you cannot derive from foundations, you cannot teach. This is why the jargon of the postmodern theories that are now falling into disfavour are doomed to have about as much traction as the gender-neutral pronouns that people periodically invent to supplant “him” and “her”. One simply cannot wage a singlehanded war against linguistic convention. And if you try, you must at least dip into those loathed conventions to explain what you are doing.

There is much in Derrida that is worth a look; I have an especial fondness for his critique of the subservience of writing to speech, the logical manoeuvre at deconstruction’s heart. And there are a great many things we can learn from deconstruction, should we be so prudent as to restrict its problem domain to the study of representation, or semiotics, where the signified referents of language are welcome to sit on the sidelines. Unfortunately, most of the scholars who immerse themselves in Derrida (Gayatri Spivak, anyone?) are so eager to practice what he preaches that they wind up disseminating neither Derrida’s ideas nor their own. Eventually, those ideas will die.

Freud and Marx at least had the courage to express themselves not in ouroboric platitudes, but in language that left their respective theories and methodologies open to scrutiny. Their ideas cannot hide behind the shield of “You don’t understand”, “You must not be reading it correctly”, or “Terry Eagleton only damages himself by refusing to read and engage Gayatri Spivak’s important contribution to the theory of cultural studies with the seriousness that it deserves.” (So says Judith Butler, whose defence of Spivak amounts to, “A lot of people read her and nod their heads; ergo, she must be ‘provocative and indisputably important.’”)

Like the worst elements of religious orthodoxy, spurious theories abuse language to protect themselves from challenge by erecting a tautological wall. In the end, we are left with a localized pool of incestuous memes that know how to survive, but not how to expand their reach. Meanwhile, the rest of intellectual history evolves some legs and marches on past.

In brief: What begins as a comprehensive study of poetic translation evolves into a treatment of human empathy and intercultural understanding, a refutation of John Searle’s Chinese Room argument against artificial intelligence, and a solemn remembrance of the author’s deceased wife. With its exclusive focus on language, Le Ton beau is a substantially less technical and more streamlined tome beau than Gödel, Escher, Bach; the mathematically averse may find it a more accessible point of entry to Hofstadter’s thought, as there is no talk of recursion or formal incompleteness in sight. Those who prefer their poetry devoid of metre and rhyme will take issue with Hofstadter’s conservative aesthetics; those who prize pattern, structure, and wordplay will rejoice.

(The Wednesday Book Club is an ongoing initiative of mine to write a book review every week. I invite you to peruse the index. For more on Le Ton beau de Marot, keep reading below.)

1.

At over 600 pages, Le Ton beau de Marot is as thorough a case as one is likely to find for the argument at its nucleus: that poetic form is an integral conveyor of poetic meaning. Its backbone is Clément Marot’s cigar-shaped poem “A une Damoyselle malade”, 24 lines and 84 syllables of 16th-century Old French to which Hofstadter refers by the first line, “Ma mignonne”. Over the course of the book, Hofstadter presents over 70 different translations and permutations of “Ma mignonne”, from his own multifarious efforts to those of his colleagues, translators, and students. The translations vary in tone, idiom, semantic liberties, and respect for formal constraints: late in the book, we encounter everything from second-order translations (e.g. French-Italian translations juxtaposed with English explications of the Italian text) to the strained attempts of computer programs designed to translate technical documents.

All of this is a basis for a discussion of just about anything Hofstadter can relate to translation. Among the issues in play:

• To what extent should translators sacrifice fidelity to an original text’s meaning to preserve its formal features—metre, rhyme, puns, lipogrammatic omissions of the letter E—or vice versa?
• When do we preserve the idioms and cultural connotations of a work’s original language, and when do we opt instead for the transposition (“transculturation”) of those occurrences into the idioms of the destination language?
• Why do performers of classical music often get top billing over composers, yet translators’ names hide in obscurity under the original authors? How much credit for authorship do translators deserve?
• How tightly bound to its language must a text be to be truly untranslatable?

This should be different but not unfamiliar territory to anyone familiar with Hofstadter’s earlier works—chiefly, Gödel, Escher, Bach: An Eternal Golden Braid and the compilation of Scientific American columns published as Metamagical Themas: Questing for the Essence of Mind and Pattern. Le Ton beau de Marot is a much wordier tome, as Hofstadter sidesteps the lectures on formal logic and recourse to graphical examples that characterize the other books and specifically writes for a verbally oriented audience. Nevertheless, his predominant interests remain intact: the interplay of form and content, paradoxes of self-reference, and how human cognition grasps the complexities that emerge from interlocking layers of meaning. And, as is characteristic for someone with a deep interest in self-reference, topical puns and anagrams are everywhere to be found.

Le Ton beau de Marot has a very autobiographical bent: there is a story behind every major object of study, and Hofstadter does not shy away from telling us how he first discovered them and imparting his personal aesthetic judgments of their elegance. This appeals to readers like myself who already have a good sense of Hofstadter’s interests from his earlier works and are eager to find out more about the polymath behind the pages, though others reading him for the first time may find him too digressive. But by the end of the bittersweet conclusion, anyone can understand why language means so much to him personally: he maps the marriage of form and content, medium and message, to the joyous marriage that he had himself before it was cut short by his wife’s untimely death.

The strength of the book is the bevy of examples and case studies that fill every chapter. Hofstadter evaluates translations of texts ranging from classics of world literature like Pushkin’s Eugene Onegin and Dante’s Inferno to texts that are defined largely by their formal wordplay—Georges Perec’s E-less La disparition, Hofstadter’s own GEB (beautifully rendered in Mandarin as Jí Yì Bì, “Collection of Exotic Jade”), and sentences like this Dutch specimen:

Dit pangram bevat vijf a’s, twee b’s, twee c’s, drie d’s, zesenveertig e’s, vijf f’s, vier g’s, twee h’s, vijftien i’s, vier j’s, een k, twee l’s, twee m’s, zeventien n’s, een o, twee p’s, een q, zeven r’s, vierentwintig s’s, zestien t’s, een u, elf v’s, acht w’s, een x, een y, en zes z’s.

To my delight, Hofstadter recounts his own attempt to devise a Polish-English translation of the first story in Stanislaw Lem’s Cyberiad, the one about the machine that can create anything starting with the letter n. He compares it to the standard English translation by Michael Kandel (who contributes his own rendering of “Ma mignonne” to the book) to illustrate different solutions to the problem of preserving, among other things, a joke about natrium (sodium), a satirical passage about nauka (science), and the crux of the story, when the machine is commanded to create Nothing.

I was surprised that Hofstadter did not mention one of the later stories in The Cyberiad (and my personal favourite), “Trurl’s Electronic Bard”. The story tells of a poetry-writing machine that spits out alliterative poems about haircuts and romantic pastorals in the language of tensor algebra, and is in some respects a microcosm for the entirety of Le Ton beau de Marot in a fraction of the space. It covers everything: man/machine hostilities, the avant-gardist rejection of crossword-puzzle poetry composed within tight formal constraints, the construction of poetic language upon the edifice of an entire cultural history—it’s all there. As Hofstadter has obviously read The Cyberiad, I am shocked he never mentions “Trurl’s Electronic Bard” once.

2.

My regular readers may recall that about a year ago, I wrote a post about The Cyberiad that lamented the decline of poetic forms that have yet to be exhausted; Hofstadter dedicates an entire chapter to the topic, sparing no harsh words for modern translations of Dante’s Inferno that cast aside its delicate terza rima architecture as if it were a dispensable accessory to the poem’s meaning. “Should [Dante translator Robert Pinsky] someday wind up in one of the nine circles of Hell,” he writes, “I know just the punishment that matches his sins committed on earth: He should be condemned to construct, unto eternity, one tercet after another in perfect, non-slanted English rhymes.”

Central to Hofstadter’s aesthetics is a belief that art is pattern; his personal hero is Frédéric Chopin. In one of Le Ton beau‘s most controversial passages, entitled “The Trendy Pooh-poohing of Pattern”, he expresses his bafflement at how the anti-structural movements of the twentieth century have taken over the tradition of high art:

The collective message radiated by a large clique in the club of today’s poets is that the devices of rhyme and rhythm and such things were nice back in the quaint old days, but in our infinitely more serious and sophisticated day and age, it would be horribly kitschy to resort to such easy, fluffy, childish sound-games.

Over our century the esoteric and obscurantist tendency grew and grew, till it became the norm. As this happened, of course, rhyming and rhythmicality grew tainted with suspicion: to indulge in such things was the signature of doggerel, of light verse, of Ogden Nash-style frivolity. It thus became de rigueur to avoid catchy sonorous patterns, so as to avoid being labeled “light”, “popular”, “accessible”. What a kiss of death! Whereas poets and composers in previous centuries didn’t make being inaccessible a high goal, in our century failing to achieve wide appeal has become a mark of elitism and success.

It is important to note that this is not a populist argument: Hofstadter reserves an equal amount of disdain for modern popular music (rock in particular), which he sees as a far and vulgar cry from the intricate lyrical stylings of W.S. Gilbert, Cole Porter, and Ira Gershwin. What Hofstadter values in his art is complexity and sublimity of formal architecture, in no small part because he conceives of human intelligence in general as the ability to detect patterns and create analogies.

Readers familiar with Gödel, Escher, Bach will remember the passages in which Hofstadter criticizes the works of John Cage (“Composition of Aleatorically Generated Elements”) as lacking in intrinsic meaning and structural qualities, as opposed to pieces by J.S. Bach (“Beautiful Aperoidic Crystal of Harmony”). Cage’s compositions may possess a lot of extrinsic, phenotypic meaning in the academic conversation over the history of Western music, but as genotypic structures they do not stand alone apart from external explanation; for Hofstadter, art should stand on its own merits.

The irony, however, is that most of the translations of “Ma mignonne” that Hofstadter puts on display in Le Ton beau in Marot are incredibly context-dependent. It is easy to appreciate their cleverness if you regard them as solutions to a puzzle, but without reference to the Marot poem and some of its more esoteric formal constraints (among them, a requirement that the poet name himself in the text), most of them are but lightweight trivialities. If you were to set the translations apart from the original poem, they would hardly be worthy of consideration. Hofstadter concedes as much: he acknowledges that his playful style comes off as juvenile next to some of the profound submissions by other translators, which he also includes in the book. As he admits in the chapter on untranslatability, there is a point at which the imposition of constraints comes at the expense of meaningful standalone coherence.

3.

The turf war over the centrality of structure is not a new debate in poetics by any means. If we wind the clock back to 1668, we can observe John Milton’s prefatory comments on the verse form of Paradise Lost:

The Measure is English Heroic Verse without Rhyme, as that of Homer in Greek, and of Virgil in Latin; Rhyme being no necessary Adjunct or true Ornament of Poem or good Verse, in longer Works especially, but the Invention of a barbarous Age, to set off wretched matter and lame Meter; grac’t indeed since by the use of some famous modern Poets, carried away by Custom, but much to their own vexation, hindrance, and constraint to express many things otherwise, and for the most part worse than else they would have exprest them.

Surely, even Hofstadter would agree that the prosody in Paradise Lost is beyond reproach (in good measure because of its strict metricality), though he would probably prefer that it rhymed. As a connoisseur of structurally elegant art that speaks for itself (as we would expect him to be, coming as he does from a background in classical music, where there is no content without form), Hofstadter does not appear to put great stock in literary history. Linguistic history, yes—he has a keen ear for the associative effects produced by the dialects of Shakespearean English and inner-city rap English alike—but as for the evolution of poetic constraints and verse forms themselves, not so much.

I would argue that, for better or for worse, the erosion and eventual collapse of metre and rhyme came about as a result of conventional verse forms accumulating so much cultural baggage that a descent into self-parody was all but unavoidable. We can see this happening to the English heroic couplet as early as 1714, in Alexander Pope’s The Rape of the Lock. Mind you, this did not preclude Romantics such as Coleridge and Byron from continuing the tradition of rhymed verse with a straight face a century later.

The fault that Hofstadter finds in the avant-gardist abandonment of past forms as passé is the notion that constrained compositional spaces have been fully explored. (I’ve addressed this quite frequently myself: see my post on The Cyberiad and my review of jazz critic Stanley Crouch’s Considering Genius.)

My personal position on anti-structural art is conflicted: I, too, am none too pleased that beauty and elegance have yielded to social transgression as the most cherished value of critical discourse. On the other hand, I firmly believe that modern and postmodern works of art, music, and verbal composition are at least worthy of objective analysis that takes them apart, explains how they work, and sheds light on why they command the tremendous academic capital they presently do.

Conversely, no matter what your opinion is on modern music and poetry, the public rejection of it is also too systematic to ignore or attribute to the ignorance of the masses; it should be open to investigation in the field of cognitive science. Contrary to the apologists for obscurantism in the humanities and social sciences, I do not believe that Hofstadter is being merely dismissive of something he doesn’t “get”.

4.

We should also remember that Hofstadter’s rejection of anti-structural art only becomes prescriptive—a dictation of how others should behave—in the context of translation. His dictum is clear: write your own poetry if you must dispense with form, but if you are translating a carefully structured work, treat the medium of expression as part and parcel with the message (as opposed to a dispensable container) or risk losing the distinctive flavour of the source text entirely.

In this debate in translation theory, the individual Hofstadter identifies as his chief opponent is Vladimir Nabokov, whose idea of a translation of Eugene Onegin (originally composed in Onegin stanzas) was a prose explication of the Russian text accompanied by 1,000 pages of notes. A handy reference for scholars of Russian, perhaps—but not, in Hofstadter’s view, a translation: at least, not a translation presentable to an English-language readership that wants to savour the lilting qualities of the Pushkin sonnet, the very property that distinguished the work for its original Russian audience.

One of the unexpected discoveries buried amidst Hofstadter’s excoriation of Nabokov (and, specifically, of Nabokov’s derision towards other Onegin translators who deigned to attend to form) was the light it shed on why Nabokov’s novel Ada, or Ardor bothered me as much as it did. (You may recall from a fortnight ago that Ada was a book I desperately wanted to love, but couldn’t.) The culprit, directly or indirectly, was Nabokov’s attitude towards translatability, which no doubt informed Ada‘s excessive untranslated trilingualism as well as its overwhelming allusive dependence on antecedent literature—the antithesis of the standalone ideals that Hofstadter promotes.

(The latter impediment was the same problem I had with Umberto Eco’s Foucault’s Pendulum when I read it years ago. In one of Le Ton beau de Marot‘s most amusing moments, Hofstadter points out how the name Umberto Eco could be anglicized as “Humbert Echo”—in other words, Humbert Humbert, the protagonist of Nabokov’s Lolita! It would be easy to regard this as one of Nabokov’s characteristic allusions were it not for the fact that Lolita was published well before Eco produced anything of note.)

5.

Readers with an keen interest in artificial intelligence, as I do, will find an additional delight in Le Ton beau de Marot: the manner in which Hofstadter parlays issues of translation into a refutation of John Searle’s Chinese Room argument. For those of you new to the ongoing battle between proponents of artificial intelligence and the philosophers who insist computers can never really think, I’ll provide the briefest summary I can.

The Chinese Room argument runs thus: a computer program that demonstrates complete linguistic facility indistinguishable from humans is still, under the hood, nothing more than a syntactic symbol-pusher. It no more “understands” a language like Chinese than a non-Chinese speaker would if placed in a room to mechanically produce a translation based on dictionaries and set procedures. Searle argues that syntactic manipulation can never give rise to semantic comprehension; ergo, computers are inherently unable to think “about” anything in the way humans do.

The pertinence of translation to the Chinese Room argument is obvious. Hofstadter, a proponent of artificial intelligence and abstract models of consciousness (albeit one who recognizes AI’s current practical limitations), initially defers to the Systems Reply: that it is fallacious to identify with the symbol-pushing human in the Chinese Room when he or she is merely one infinitesimal cog of a complex system that, taken as a whole, is intelligent.

What Hofstadter’s study of translation adds to the discussion is its dismantling of Searle’s clean division between syntax and semantics. As Hofstadter shows, any machine of sufficient linguistic aptitude to pass for a native human speaker (as per the Turing Test) must account for semantics. If it plays with syntactic patterns alone at the naïve, grammatical level, it would fail to pass for an entity of humanlike intelligence in the first place. To demonstrate this, Hofstadter turns the tables on the Chinese Room, and compares a machine translation of “Ma mignonne” (needless to say, with complete disregard for the formal qualities of the poem) with a translation performed by a Chinese speaker with little to no knowledge of French or English, armed with a stack of dictionaries and rule books.

This isn’t to say that Hofstadter has unrestrained enthusiasm for machine translation: his refutation of Searle is a defence of the validity of the Turing Test, but in no way a commendation of MT research’s present direction. Hofstadter criticizes efforts in MT for their exclusive dedication to industrial applications, which has led to odd stylistic guidelines like Xerox’s directive that its technical writers compose in a specific manner designed to avoid tripping up the translation engine. For Hofstadter, this privilege of denotative meaning at the expense of higher-level structures exhibits a lack of respect for the subtleties of natural language, and precludes MT from reaching a stage where the Turing Test is remotely within reach. As the hype-driven enterprise of automated translation reaches out to wider problem domains without pause for refinement in any of them, we gain as little insight into human cognition as we did when Deep Blue defeated Garry Kasparov with brute-force search.

6.

Seeing as how Le Ton beau de Marot was published in 1997, long before the advent of Google Translate’s large-scale implementation of a statistical methodology, I thought I’d plug “Ma mignonne” into Google Translate to get a sense of our present state of affairs.

Here is Clément Marot’s original:

Ma mignonne,
Je vous donne
Le bon jour ;
Le séjour
C’est prison.
Guérison
Recouvrez,
Puis ouvrez
Votre porte
Et qu’on sorte
Vitement,
Car Clément
Le vous mande.
Va, friande
De ta bouche,
Qui se couche
En danger
Pour manger
Confitures ;
Si tu dures
Trop malade,
Couleur fade
Tu prendras,
Et perdras
L’embonpoint.
Dieu te doint
Santé bonne,
Ma mignonne.

And here is “Ma mignonne” according to Google Translate:

My cute,
I give you
The good days;
The stay
It is prison.
Healing
Cover,
Then open
Your door
And so we
Vitement,
For Clement
The request you.
Will fans
Your mouth,
Who goes to bed
Endangered
To eat
Jams;
If you hard
Too sick
Color fade
Thou shalt take,
And lose
Being overweight.
God you doint
Good health,
My cute.

The similarities to the machine-translated samples in Le Ton beau de Marot are striking, and many of the stumbling blocks are the same—the use of “cute” as a noun, for instance; or the archaic formations vitement and doint, which speakers of modern French should still be able to parse. What surprises me is that the corpus-based statistical translation provided by Google Translate far more closely resembles the poem generated by the syntactic, rule-based Globalink Translation System (#66 in Le Ton beau, “My Cute”) than the one produced by the corpus-based, statistical Candide (#67, “My Flapper”).

At this stage in the game, it’s no wonder that the opponents of AI would seize upon these rudimentary, syntactic efforts and extrapolate, from them, the impossibility of machines attaining humanlike levels of linguistic comprehension and production. Of course, Hofstadter also produces examples of how good generated prose can be if one develops and refines it within a specific problem domain. Consider Anthony Davey’s “Proteus”, a program developed to play tic-tac-toe and supply a running strategic commentary that produced lucid sentences like this one: “If you had blocked my edge, you would have forked me, but you took the middle of the one opposite the corner I had just taken and adjacent to mine and so I won by completing my edge.”

The human capacity for language, too, may ultimately be a reflection of our versatility in switching between limited contexts and domains. We read and write in one way when it comes to business letters, and another way when it comes to rhymed verse; one way in French, and another way in English. Not so different, after all, from the architecture of digital computers—the universal Turing machines, the programs that can simulate all other programs.

It’s fascinating to me, as someone with more than a passing interest in random, gratuitous acts of paronomasia, that it is entirely possible to construct puns that work in some dialects and accents, but not others. This may seem like a rather simple observation, but I think it has a certain latent power. It could also be disabling. For instance, if your philosophical rhetoric is founded on punning as a substitute for logic (not saying I mind), the puns had better work. Otherwise, you might be caught, and I’ll either see you in court (in Britain) or ignore you and lie on my cot (in America).

Speaking of which: I’ll see you in hell, Pachelbel.