On Literature

LES SEMAPHORES SOUS LA PLUIE

How is space represented in words? This problem has a history of its own, and the rhetorical tradition classifies the techniques of verbal representation of space (as of every other visual experience) under the heading of hypotyposis, or "evidentia," which is sometimes considered the same as, and sometimes judged to have affinities with, "illustratio," "demonstrado," "ekphrasis" or "de-scriptio," "enargeia," etcetera.

Unfortunately, all definitions of hypotyposis are circular, which is to say that they define as hypotyposis that figure through which visual experiences are represented or evoked through verbal procedures. Look at the definitions that come from the major exponents of classical rhetoric, from Hermogenes to Longinus, from Cicero to Quintilian, which I quote from Lausberg without specifying who said what, since one seems to borrow from the other: (i) "credibilis rerum imago quae velut in rem praesentem perducere audientes videtur" (a believable image of something that seems to lead the public into its very presence); (ii) "proposita forma rerum ita expressa verbis ut cerni potius videatur quam audiri" (a form of things proposed by the speaker in words in such a way that they seem to be visible rather than audible); (iii) "quae tam dicere videtur quam ostendere, praesentans oculis quod demonstrat" (things that the speaker seems to display as much as to say, presenting before our eyes what he is trying to show); (iv) "quasi gestarum sub oculis inductio" (a kind of parading before the eyes of things that have been done), and so on.

I have before my eyes (but this time in the literal sense of the expression) the paper on hypotyposis given by Hermann Parret at the Decade de Cerisy, which took place in July 1996, and here too this expedition into the forest of the most modern theoreticians does not seem to yield appreciable results.^* Dumarsais reminds us that hypotyposis means image, or painting, and that it comes about "when, in descriptions, one depicts facts that are spoken about as though what is described were genuinely before our eyes; the speaker displays, as it were, what he recounts..." (Des Tropes), and that for others this figure of speech "allows us to touch reality with our fingers"—a fine metaphor, to be sure, but using one figure of speech to define another is not very helpful. All the more so since, as Aristotle observes, a figure that almost places things before our eyes is a metaphor—and no one will claim that a metaphor is the same thing as hypotyposis. The truth is that if by definition rhetorical figures endow a discourse with brilliance, vivacity, and persuasiveness, and if, according to Horace, one has to admit that poetry is to a certain extent "ut pictura" (like painting), then all such figures surprise the reader or listener and place something before their eyes in one way or another. But if that were the case, and if this metaphor were too generic, where would hypotyposis end up?

Luckily, at those very points where theoreticians are incapable of telling us what hypotyposis is, they are nearly always able to provide us with magnificent examples of it. The first three come from Quintilian (Institutio Oratoria, 8.3.63–69). His first one quotes a line from the Aeneid(5.426) where two boxers "stood up instantly erect on the tips of their toes." The second quotes Cicero's Verrine Orations (5.33.86): "He stood on the shore, wearing sandals, with a red pallium and a long tunic down to his feet, leaning on a vulgar little woman, he, the praetor of the Roman people," and Quintilian wonders if there is anyone so devoid of imagination as not to be able to visualize this scene and its protagonists, and even more than is said, to see their faces and eyes and their obscene caresses, and the uneasiness of the onlookers. The third example, also from Cicero, is a fragment of his speech for Q. Gallius, and refers to a dissolute symposium: "I seemed to see people going in and coming out, some staggering in drunkenness, others yawning from the drunken excess of the day before. The floor was filthy, spattered with wine, and covered with garlands of faded flowers and fish bones."

His fourth example goes like this: "Undoubtedly, in fact, whoever says 'a city has been captured,' associates with that phrase the idea of all the horrors that such a calamity usually entails, but this kind of concise statement does not arouse profound emotion. If instead the concepts contained in a single word have the chance to expand, they will set before you flames spreading through houses and temples, the roar of crashing buildings and the indistinct uniform rumble produced by various sounds, the uncertain flight of some people, the last desperate embraces of others, the howling of infants and women, and the old people who have remained alive, unluckily for them, until that day; and then the devastation of things both sacred and profane, the milling around of those carrying off booty and coming back for more, and the prisoners, in chains, being pushed along, each one by his own torturer, and the mother trying not to let her child be taken away from her, and the struggle among the victors..."

Similarly, Dumarsais suggests the following extract from Racine's Phèdre as an example of hypotyposis:

Cependant sur le dos de la plaine liquide

S'élève à gros bouillons une montagne humide;

L'onde approche, se brise, et vomit à nos yeux,

Parmi des flots d'écume, un monstre furieux.

Son front large est armé de cornes menaçantes;

Tout son corps est couvert d'écaillés jaunissantes;

Indomptable taureau, dragon impétueux,

Sa croupe se recourbe en replis tortueux...

Meanwhile on the surface of the ocean's plains

A watery mountain arises amidst great froth;

The wave approaches, breaks, and spews forth before our eyes,

Amidst the spray of foam, a furious monster;

Its broad forehead is armed with menacing horns,

All his body is covered with yellowish scales;

This indomitable bull, this aggressive dragon,

Has a croup that curves round in tortuous curls...

If we consider the four examples from Quintilian, we see that in the first all that is mentioned is a physical posture (and the reader is invited, as it were, to imagine the scene). The second describes a pose with a certain amount of spite; the solemnity of the red pallium is set against the vulgarity of that little woman ^ muliercula '), leading the addressee to notice this clash. In the third passage what makes the description interesting is not only its greater precision and length but the unpleasantness of the things described (we must not forget that in the classical art of memory a monstrous or terrible image had more chance of being remembered by the speaker and therefore of being evoked at the right moment). The fourth example is not a specific instance, but it suggests what a detailed and moving description of an extended sequence of actions might be like, and I deliberately describe the dramatic sequence of these actions as if I were discussing a cinematic sequence.

In the case of the Racine passage we have something even more complex: the description of various phases of a natural event, but with continuous zoomorphic transformation of each of the wavy forms that are listed. It is difficult to resist the temptation, or the habit, of imagining them visually. It might seem irreverent to cite Walt Disney and Snow White's flight into the woods (though this irreverence would be lessened if we supposed that Walt Disney had poetic procedures of this very type in mind), but in reality both Racine and Walt Disney are simply following one of the most natural tendencies in human beings, namely, giving substance to shadows, or, rather, seeing threatening, animal forms in the shapeless darkness of nature in turmoil (and in this context Parret rightly sees hypotyposis as one of the figures that leads to the production of the sublime).

What I think we can say, however, is that in these examples we are faced with differing descriptive and narrative techniques, which have in common only the fact that the addressee draws a visual impression from them (if he wants to—in other words, if he wants to collaborate with the text). This means I can therefore say that hypotyposis does not exist as a specific rhetorical figure. Language allows us to describe faces, forms, positions, "scenes," and sequences of action, and it allows us to do this continually in everyday life (otherwise we could not even say, "Could you please go to the hardware store and get me something that looks like this"); all the more so, then, does it encourage us to do this for artistic reasons. However, it allows this to happen by way of a multiplicity of techniques, which are not reducible to a formula or rule, as can be the case with real tropes and rhetorical figures like synecdoche, hyperbaton, zeugma, and even—to some extent—metaphor.

All we need to do then is to proceed to a typology of techniques for representing or evoking space. Except that at this point we really need to ask ourselves what is meant by space—and we cannot avoid asking ourselves an analogous question about time.

There is Newtonian space and time, which exist as absolute entities, and Kantian space and time, which exist as pure intuitions and conditions that are a priori of experience. There is the Bergsonian contrast between clock time and the time of internal duration, and there is the measurable space of Cartesian geometry and the lived space of phenomenology. It is not a question of privileging one or the other, since language always allows us to speak of these things, and we can say without difficulty how many millions of miles we need to travel to reach Alpha Centauri or to make us undergo (or, rather, suffer) an interminable journey between Florence and Fiesole, not to mention a voyage around one's own room (in Tristram Shandy the discussion between two characters coming down a staircase takes up three chapters).

If we wanted to rewrite Lessing's Laocoon today (after the invention of new mimetic techniques like cinema), we would have to ask ourselves if a division between temporal arts and spatial arts still makes sense, and—if we regard such a division as still valid—ask how spatial arts can represent time and temporal arts represent space.

In the meantime, there can still be many reflections on how spatial arts represent space. The classic example of this is perspective, where a two-dimensional physical surface produces three-dimensionality as its proper content, and where a minimal portion of representational surface can express a vast expanse of space: this is realized by anyone who, after long contemplation of it in various reproductions, finally sees Piero della Francesca's Flagellation in the Ducal Palace at Urbino and is amazed at how such a small frame can contain what is perceived to be such a vast space.

I have dealt with the question as to how the arts of space represent time, or actually imply the time of their own contemplation, elsewhere. ^* Phenomenology is vast and requires first and foremost an analysis of the various relations between what Genette calls signifying spatiality and signified spatiality (and which for reasons that will become clear later I would prefer to refer to as spatiality of expression and spatiality of content). There are paintings that suggest a sort of freezing of the moment, like the Annunciation by Lotto, where Mary's gesture of surprise is caught in the moment a cat darts across the room, or the slash by Lucio Fontana, a snapshot of the lightning movement of the blade that has cut through the stretched canvas.

But when one thinks carefully about it, there is nothing unusual in the fact that a limited portion of space, which of itself is atemporal, can express an instant. The problem arises when you ask yourself how you can express a long period of time through portions of space. And you discover that, in order to express a long period of time you generally need a lot of space. There are stories in painting that represent a century-long succession of events through a series of frames, as happens with comic strips; and there are others that do so through the repeated visual presentation of the same characters in different hairstyles, situations, and ages; and these are all cases where an abundance of space is required to convey an abundance of time, and not only an abundance of signifying space, but also of the (semantic, not pragmatic) space that the beholder has to traverse. To grasp the flow of time in Piero's series of frescoes on the Finding of the True Cross in Arezzo you have to move, and not only with your eyes, but also with your feet, and you have to walk even more to follow the whole story narrated by the Bayeux tapestry. There are works that require a long time to circumnavigate, and a long period of attention to their minutest details, such as a Gothic cathedral. A sculpture that appears as a little ivory cube can be experienced in a second of contemplation (even though I believe it should be touched and rotated in order to grasp all its facets), but a cube in which every side was of one million by one million kilometers has to be circumnavigated, maybe with the spaceship from 2001: A Space Odyssey, otherwise one would not grasp its megagalactic sublimity.

If, then, to represent a lot of time one needs a lot of space, is it not the case (with the arts of time) that to represent a lot of space one needs a lot of time? Let us start by limiting this topic. We shall not be asking ourselves whether, for instance, music can represent space, even though intuitively we would reply yes. Even if we did not want to be extreme descriptivists, it is hard to deny that Dvorak's New World Symphony or Smetana's Md Vlast conjure up vast expanses, so much so that the conductor is tempted to conduct with expansive, smooth gestures, almost as if to suggest something "flowing," and that the length of these pieces contributes to the creation of that effect. Of course, certain kinds of music dictate a pirouette, others a leap, others a calm walk, and so there are rhythmic structures that determine or represent bodily movements with which we move through space—otherwise dance would not exist.

But let us limit our discussion to verbal discourse, and take up again the distinction between space of expression and space of content, emphasizing the fact that we shall have to deal not so much with the form of expression as with its substance.

We saw before that, however well intentioned Quintilian was, saying of two boxers that they stood upright on the tips of their toes does not strike us as a great hypotyposis, whereas describing the invasion and sack of a city event by event and moment by moment seems more visually evocative. But this latter description implies a certain number of pages (or at least of verses).

As a result, in speaking of spatial expression I would not consider those cases where, on the level of expression, something is said about space that, on the level of content, is not really about space but rather about something else, and it could even be about the flow of time, as happens in expressions such as "the party line," "it is a bleak outlook," "boundless culture," "he followed the discussion step by step," "he is on the wrong track," "in the middle of the night,"—see what Lakoff and Johnson write about these spatial metaphors.^* I am thinking more of the substance of the expression, the quantity of which impinges on the spatiality expressed. I simply want to say that between mentioning a beautiful valley and describing it over one hundred pages, something should happen on the level of content; or, in other words, we ought to see something more of that valley.

Let us examine some techniques of verbal expression of space.

1. Denoting. This is the most simple, immediate, mechanical form, as when one states that there are twenty kilometers of distance between one place and another. Naturally it is up to the addressee, if's/he wants, to associate an image with the conceptual content just acquired (maybe the image of himself as overheated jogger, if the information is about an area devoid of transport), but one cannot say that the statement on its own does anything to force the addressee to picture that space for himself.

2. Detailed Description. Things are already different with a space that is described, as when we say that a square has a church to the right and an ancient palace to the left. Note that the simple mention of the respective positions of the two buildings would already be sufficient to make us recognize the piazza in question, and therefore we would have to say that every description of visible objects is of itself hypotypotic. But let us imagine that in the city where we are told this there are two squares with a church and a palace, and suddenly the force of the description as hypotyposis would diminish. We would then need to supply more details.

Here we come up against a problem of quantity: how many details do there need to be? Enough to encourage the addressee to build up an image in his mind, but not too many, because in that case the image would become impossible to build up. Take the case of this description in Robbe-Grillet s The Voyeur:

The stone rim—a sharp, oblique edge, at the intersection of two perpendicular planes: the vertical wall that rose upright toward the quay and the flight of stairs that went up to the highest part of the pier—at its upper extremity (the high part of the pier) was extended in a horizontal line straight toward the quay.

The quay, which the perspective effect pushed into the distance, sent out on both sides of this main line a group of parallel lines, which, with a precision that was further heightened by the morning light, delimited a series of long planes that were alternately horizontal and vertical: the upper plane of the massive parapet protecting the passage on the water side, its inner wall, the paving on the high part of the pier, and the unprotected side that plunged into the sea in the harbor. The two vertical surfaces were in shadow, the other two brightly lit by the sun: the upper floor of the parapet throughout its entire breadth, and the paved walkway, except for a narrow strip which was in darkness: the shadow projected by the parapet. In theory one should also have been able to see the mirror image of the whole construction in the harbor water, and on the surface, still within the same system of parallel lines, the shadow projected by the high vertical wall going off straight toward the quay.

This passage (the description continues for almost two pages) is interesting because it says too much, and by saying too much it prevents the reader from building an image for himself (unless he is one of those few readers who manages to do so with considerable effort and intense concentration). This seems to point to the conclusion that for hypotyposis to exist, it must not say more than what leads the reader to collaborate with the text by filling empty spaces and adding details on his own initiative. In other words, hypotyposis does not so much have to make us see, as make us want to see. But where can one find a rule? In the case of Robbe-Grillet we could say that the passage fails deliberately in its attempt to make us see (an interestingly provocative act for a book entitled The Voyeur and belonging to the école du regard), because it does not offer a hook on which to hang any priority: everything is as important as everything else.

What does offering priority mean? It could mean emphasizing verbally, in a certain sense almost dictating, an emotional reaction ("in the center of the square there was something terrifying..."), or insisting on one detail at the expense of others.

There are countless examples of this, but I will restrict myself to a passage that has elicited thousands of attempts at translation into visual terms, from Revelation, chapter 4:

And lo, a throne stood in heaven, with one seated on the throne! And he who sat there appeared like jasper, and carnelian, and round the throne was a rainbow that looked like an emerald. Round the throne were twenty-four thrones, and seated on the thrones were twenty-four elders, clad in white garments, with golden crowns upon their heads. From the throne issue flashes of lightning, and voices and peals of thunder, and before the throne burn seven torches of fire, which are the seven spirits of God; and before the throne there is as it were a sea of glass, like crystal. And round the throne, on each side of the throne, are four living creatures...

This is hypotyposis if ever there was one. Yet the description does not include everything. It stops only at surface details; only the elders' clothes and crowns are mentioned, not their eyes or beards.

3. Lists. This is a technique that undoubtedly leads to the evocation of spatial images without establishing priorities. Three examples: one classical, or at least from late Latin, and two modern.

Here is the description of the city of Narbonne in Sidonius Apollinaris:

Salve Narbo, potens salubritate,

Urbe et rure simul bonus videri,

Muris, civibus, ambitu, tabernis,

Portis, porticibus, foro, theatro,

Delubris, capitoliis, monetis,

Thermis, arcubus, horreis, macellis,

Pratis, fontibus, insulis, salinis,

Stagnis, flumine, merce, ponte, ponto;

Unus qui venerere iure divos

Leneum, Cererem, Palem, Minervam

Spicis, palmite, pascuis, trapetis.

Hail, Narbonne, famous for your healthy air,

Both your town and your countryside are good to see,

As are your walls, citizens, periphery, shops,

Harbor, colonnades, forum, theater,

Shrines, capitols, banks,

Baths, arches, granaries, butchers,

Meadows, fountains, islands, salt flats,

Pools, river, merchandise, bridge, and sea;

The only town that can rightly worship the gods

Bacchus, Ceres, Pales, Minerva,

With your vine sprigs, corn ears, pasture land, and olive presses.

And here is a single paragraph from the description of the drawers in Leopold Bloom's kitchen, in the penultimate chapter of Ulysses:

What did the first drawer unlocked contain?

A Vere Foster's handwriting copybook, property of Milly (Millicent) Bloom, certain pages of which bore diagram drawings, marked Papli, which showed a large globular head with 5 hairs erect, 2 eyes in profile, the trunk full front with 3 large buttons, 1 triangular foot: 2 fading photographs of queen Alexandra of England and of Maud Branscombe, actress and professional beauty: a Yuletide card, bearing on it a pictorial representation of a parasitic plant, the legend Mizpah, the date Xmas 1892, the name of the senders: from Mr + Mrs M. Comerford, the versicle: May this Yuletide bring to thee, Joy and peace and welcome glee: a butt of red partly liquefied sealing wax, obtained from the stores department of Messrs Hely's, Ltd., 89, 90, and 91 Dame Street: a box containing the remainder of a gross of gilt "J" pennibs, obtained from same department of same firm: an old sandglass which rolled containing sand which rolled: a sealed prophecy (never unsealed) written by Leopold Bloom in 1886 concerning the consequences of the passing into law of William Ewart Gladstone's Home Rule bill of 1886 (never passed into law): a bazaar ticket, no 2004, of St. Kevin's Charity Fair, price 6d, 100 prizes...

And in this case the list continues for page after page. If we can say about the Narbonne passage that the presentation of architectural elements acts as a cinematic panning shot suggesting a shape (today we would say it suggests at least a skyline), for Joyce all that matters is the gargantuan abundance of irrelevant objects that opens up its unfathomable depths before our eyes, the labyrinthine richness of that drawer. Let us look at the description of another drawer, this one belonging to the old aunt in the "Othys" chapter of Nerval's Sylvie:

She rummaged once more in the drawers. What delights! How everything smelled good, how all the bits and pieces shone and shimmered with lively colors! Two slightly damaged mother of pearl fans, Chinese porcelain boxes, an amber necklace and a thousand frills, among which shone out two little white woollen shoes with buckles encrusted with Irish diamonds.

We certainly see shining in the drawer those nice things of terrible taste that enchant the two young visitors, and we see them because priorities stand out—that is to say that some objects are emphasized at the expense of others. But then why do we also see Bloom's drawer, where no article has a privileged role? I would answer that Nerval wants us to see what is in the drawer, whereas Joyce wants us to see the unfathomability of the drawer. I do not know whether the reader manages to see Bloom's drawer better than Robbe-Grillet's pier, seeing that both are devoid of priorities. We can say, however, that Robbe-Grillet puts together objects that in themselves are not surprising, like a pier, a parapet, some surfaces and lines, whereas Joyce puts together things that are mutually incongruous and for that very reason surprising and unexpected. In some way (I presume) the reader has to prioritize a disordered ensemble of objects, or make some choices (perhaps by focusing on the sandglass), and by doing so collaborates in the task of making mental images.

4. Accumulation. Another technique (which was already suggested by Quintilian's third example) is the excited accumulation of events. The events have to be either incongruous or extraordinary. Naturally elements of rhythm intervene, and this is why the following hypotyposis from Rabelais (Gargantua 1.27) lets us visualize the scene (whereas we would not be able to see the scene from a procedure that was merely one of listing, as in the wonderfully amusing list of "balls" in 111.26 and 28, which is more enjoyable in phonic than in visual terms):

With one lot he spilled their brains, with others he broke their arms and legs, and with a third he dislocated the neck bones, split their kidneys, slit their nose, burst their eyes, battered their jaws, shoved their teeth down their throats, shattered their shoulder blades, splintered their legs, dislocated their thigh bones, and plucked their limbs from their trunks.

If any tried to hide among the thicker vine leaves, he broke the whole ridge of their back, breaking their spine like a dog's.

If others tried to run and escape, he knocked his head away in pieces shattering his lamboidal joint. If any tried to climb up a tree, thinking he was safe, he impaled him through the arse with his stick [...]. Others he would stab beneath the ribs, wrecking their stomachs, and they died on the spot. Others he struck so fiercely on the navel that their tripes spilled out. And with others he pierced their bum-gut through their balls. It was, you must believe, the most gruesome spectacle ever seen [...]. Some died without speaking, others spoke without dying; and some died speaking, and spoke dying.

5. Description appealing to the addressee's personal experience. This technique requires that the addressee bring to the discourse something he has already seen and suffered. This activates not only preexisting cognitive schemes but also preexisting bodily experiences. I would suggest as a key example one of the many passages in Abbott's Flatland:

Place a penny in the middle of one of your tables in Space; and leaning over it, look down upon it. It will appear a circle.

But now, drawing back to the edge of the table, gradually lower your eye (thus bringing yourself more and more into the condition of the inhabitants of Flatland), and you will find the penny becoming more and more oval to your view; and at last when you have placed your eye exactly on the edge of the table (so that you are, as it were, actually a Flatland citizen) the penny will then have ceased to appear oval at all, and will have become, so far as you can see, a straight line.

Part of this technique involves the evocation of interoceptive and proprioceptive experiences on the part of the addressee. In other words, it makes the reader recall experiences in which he has suffered the effort of proceeding through a space. On this topic I would like to cite these two lines from Blaise Cendrars's Prose du transsibérien (by the way, this is a text that, since it has to describe a very long journey, uses many of the techniques I have already defined, from the list to minute description). At a certain point Cendrars reminds us that

Toutes les femmes que j'ai rencontrées se dressent aux horizons Avec les gestes piteux et les regards tristes des'sémaphores sous la pluie...

(All the women I have met rise up on the horizons With the piteous gestures and sad looks of signals in the rain...)

If we bear in mind that in French "sémaphores" are not our city traffic signals (which in French are "feux rouges") but rather signals along the railroad tracks, those who have experienced trains proceeding slowly on misty nights will be able to evoke these ghostly shapes that slowly disappear in the drizzle, as though fading away, while one looks out of the window at the countryside immersed in darkness following the panting rhythm of the carriages (that carioca rhythm evoked by Montale in "Addii, fischi nelbuio").

The interesting problem is, rather, how much these lines can be appreciated by those born in the age of high-speed trains, with their hermetically sealed windows (for which there is no longer even the trilingual notice banning us from leaning out). How does one react to a hypotyposis that summons up the memory of something that has never been seen? I should say the answer is by pretending to have seen it, and on the basis of the elements that the hypotyposic expression supplies us with. The two lines appear in a context that mentions a train traveling for days and days across endless plains, the signals (specifically named) in some way remind us of eyes twinkling in the dark, and the allusion to horizons makes us imagine them lost in a distance that the train's movement can only magnify second by second....In any case, even those who only know today's express trains have seen through the window lights disappearing into the night. And suddenly the experience we have to remember becomes tentatively outlined: the hypotyposis can also create the memory it needs for it to work.

On the other hand, if it is actually true that those who have experienced their first kiss can easily savor a line like Dante's "la bocca mi baciò tutto tremante" (he kissed me on the mouth all trembling), would we say that those who read of a first kiss for the first time cannot pretend to have experienced it? If we try to deny this, then neither Paolo and Francesca's sin nor their trembling would be understandable, victims as they were (in turn) of a fine hypotyposis.

If this is what happens, then we can also record under this rubric the cases where hypotyposis requires us to imagine nonhuman experiences. Such is the case with the "fractalization of space visible only at an ant's pace," for which there is a wonderful example in Eliot's "The Love Song of J. Alfred Prufrock":

The yellow fog that rubs its back upon the window-panes,

The yellow smoke that rubs its muzzle on the window-panes,

Licked its tongue into the corners of the evening,

Lingered upon the pools that stand in drains,

Let fall upon its back the soot that falls from chimneys,

Slipped by the terrace, made a sudden leap,

And seeing that it was a soft October night,

Curled once about the house, and fell asleep.

In this case, if the human traveler is moving too fast to realize what the walls and corners of the streets of London are like, the reader is asked to imagine at what speed fog might travel. This involves slowing down our pace, as it were, while reading, and following every crevice in the wall and edge of the window—exactly as would happen if we were asked to imagine how an ant travels through the curves of a fraction of space that we can cover, in an instant, with the sole of our foot.

It is not my intention to exhaust within the limits of such a brief paper the inexhaustible typologies of hypotyposis. I will suggest only some directions for research.

The different techniques of focalization could be analyzed. For instance, the excellent analysis by Joseph Frank of the agricultural fair and speeches in Madame Bovary, which he carried out in a genuinely pre-semiotic age, should be reread: the three levels of the square, the stage and the room are set up live, as it were, that is to say in a Griffith-style parallel montage, creating a visual effect through this emphasis on simultaneous speeches. ^* One would have to refer, of course, to the two battles of Waterloo, Stendhal's (seen by a protagonist, Fabrizio del Dongo, who is inside it, and who gets lost in the space he crosses haphazardly, while he loses the sense of the global space of the battle), and the one in Les Misérables, seen from on high by an omniscient Hugo, who analyzes the spaces Napoléon does not see. Elsewhere^† I have spoken of the different points of view that gradually create the space in Manzoni's "Quel ramo del lago di Como" (That stretch of Lake Como)—where idiosyncratic syntax corroborates the play of points of view. It would be worthwhile following step by step the vision of the three trees during the carriage ride in A l'ombre des jeunes filles en fleur, in order to grasp a double phenomenon: a successive shifting of point of view and the interspersing of spatial description with other reflections that take time (in reading) and space (in writing) in order to make the sense of a journey real, and justify the slow, progressive change of visual angle.

We could continue, but it is worthwhile trying to say something in conclusion about what links these different manifestations of hypotyposis. We have already alluded to this at various stages, and all we need to do now is pull the threads together. Hypotyposis is not based on a semantic rule, as happens with tropes and rhetorical figures, a rule whereby—if you disregard it—you do not understand what is being said. When metonymy uses the container to stand for the content ("let's drink a glass"), if the addressee ignores the rule, he (pedantically) assumes that he is being invited to sip a solid object. If in using a simile or metaphor someone says that a girl is a little deer, an obtuse person who ignores the rule will note with amazement that the person named is not a quadruped, nor has she horns on her forehead. But on the whole these misunderstandings do not take place, except in comic or surreal stories.

When, however, one is faced with any of the instances of hypotyposis cited up till now, the addressee can easily avoid collaborating and visualizing. He can merely grasp that he is being told that a city is being sacked, that a drawer is full of knickknacks, or that a certain Friar John was a giant-killer. We have even insinuated that in the case of Robbe-Grillet the reader can refuse, indeed may have to refuse, to see anything precise, because the author probably wanted to stimulate this refusal to activate an excess of visuals.

Hypotyposis is, then, a semantic-pragmatic phenomenon (besides, inasmuch as it is a figure of thought, like irony and similar figures, it requires complex textual strategies and can never be exemplified through brief quotations or formulas) and is a prime example of interpretative cooperation. It is not so much a representation as a technique for eliciting an effort to compose a visual representation (on the reader's part).

And indeed, why should we think that words "allow us to see," when they were invented precisely to speak of what is not before our eyes and what cannot be pointed at with a finger? The most words can do (since they produce emotional effects) is to lead us to imagine.

Hypotyposis uses words to make the addressee construct a visual representation. The proof of this is the kind of problems that arise in that exercise which is the opposite of ekphrasis (which is a verbal description of an image), and is the "translation," or visual materialization of what a verbal text allowed us to imagine. Let us go back to the description from Revelation I cited previously. The real problem for all the Mozarabic miniaturists (the illustrators of those splendid commentaries on Revelation known as the "Beati") was that of representing the four living creatures who are above and around the throne (in the Vulgate, which was the only version known to the illuminators, " super thronum et circa thronum"). How can these creatures be above and around the throne at the same time?

An examination of the solutions offered by the various "Beati" shows us how impossible the enterprise is, and gives rise to representations that do not "translate" the text satisfactorily. And this is because the miniaturists, having grown up using the Greco-Christian translation, thought that the prophet "saw" something similar to statues or paintings. But the culture of John the Apostle, like that of Ezekiel, from whose vision John drew inspiration, was a Hebraic culture, and moreover his was the imagination of a seer. Consequently, John was not describing pictures (or statues), but, if anything, dreams, and, if you like, films (those moving pictures that allow us to daydream, or in other words, visions adapted for the layman's state). In a vision that was cinematic in nature, the four creatures can rotate and appear at one moment above and before the throne, at another around it.^*

But in this sense the Mozarabic miniaturist could not collaborate with the text, and in some sense, in his hands and in his mind, the hypotyposis failed to work. Proof therefore that there is no hypotyposis if the addressee does not play the game.

Version of a paper given at the Centro di Studi Semiotici e Cognitivi at the University of San Marino on 29 September 1996, during a conference on the semiotics of space. I began my paper with a reference to Sandra Cavicciolis article "I sensi, lo spazio, gli umori. Micro-analisi di In the Orchard di Virginia Woolf (in VS 57 [1990]), one of the finest analyses of space in literature. On that occasion the author was present in the room. Now that she is no longer with us, I would like to dedicate these reflections to her.