by Umberto Eco
In reality, fictional worlds are parasites of the actual one, but they are in effect “small worlds” which bracket most of our competence of the actual world and allow us to concentrate on a finite, enclosed world, very similar to ours but ontologically poorer. Since we cannot wander outside its boundaries, we are led to explore it in depth. It is for this reason that Sylvie is such a magical work. It indeed requires that we know and pretend to know something about Paris and the Valois, and even about Rousseau and the Medici, because it names them; yet it demands that we walk in that limited world over and over again without wondering about the rest of the actual world. In reading Sylvie we cannot deny that there is a horse, but we are not requested to know everything about horses. On the contrary, we are obliged to muse over and over again about the woods of Loisy.
In an essay published long ago, I wrote that we know Julien Sorel (the main character of Stendhal’s Le Rouge et le noir) better than our own father.6 Many aspects of our father will always escape us (thoughts he kept quiet about, actions apparently unexplained, unspoken affections, secrets kept hidden, memories and events of his childhood), whereas we know everything about Julien that there is to know. When I wrote that essay, my father was still alive. Since then I’ve realized how much more I would have liked to know about him, and I’m left to draw feeble conclusions from lackluster memories. Stendhal, however, tells me everything about Julien Sorel and about his generation that I need to know for the purposes of this novel. What I’m not told about (for example, whether he liked his first toy, or—as in Proust—whether he tossed and turned in his bed while waiting for his mother’s goodnight kiss) is not important.
(By the way, it can also happen that a narrator tells us too much—that is, tells us what is irrelevant to the course of the story. At the start of my first lecture, I ironically quoted poor Carolina Invernizio because she once wrote that at the Turin railway station “two nonstop trains were meeting, one about to leave, the other about to arrive.” Her description appeared to be a silly case of redundancy. But on second thought, I must confess that this piece of information is not as redundant as it seems. Where is it that two trains that meet do not both leave immediately after arriving? In a terminal station. Carolina was implicitly informing us that the Turin railway station was a terminus, as in fact it still is. Yet the reason we are entitled to consider her remark, if not semantically redundant, at least narratively useless, is that such a detail is simply not essential to the development of the story: the events that follow do not depend on the characteristics of the Turin station.)
I mentioned at the beginning of this lecture a reader who checked in the newspapers of the actual Paris and who discovered a fire that my book did not mention. He did not accept the idea that a fictional world has a more modest format than the actual world. Now let me tell you another story concerning that same night in June 1984.
Two students from the Ecole des Beaux Arts in Paris recently came to show me a photograph album in which they had reconstructed the entire route taken by my character Casaubon, having photographed, at the same time of night, each of the places I had mentioned. Since the text describes in detail how Casaubon comes up out of the city drains and enters, through the cellar, an Oriental bar full of sweating customers, beer kegs, and greasy spits, they succeeded in finding the bar and took a photo of it. It goes without saying that that bar was an invention of mine, even though I designed it thinking of the many bars of that kind in the area, but those two students had undoubtedly discovered the bar described in my book. It’s not that they had superimposed on their duty as model readers the concerns of the empirical reader who wants to verify that my novel describes the real Paris. On the contrary, they wanted to transform the “real” Paris into a place in my book, and of all that they could have found in Paris, they chose only those aspects that corresponded to my descriptions.
They used a novel to give form to that shapeless and immense universe which the real Paris is. They did exactly the contrary of what Georges Perec did when he tried to represent everything that happened in the place Saint-Sulpice in the course of two days. Paris is far more complex than the locale described by Perec and the one described in my book. But any walk within fictional worlds has the same function as a child’s play. Children play with puppets, toy horses, or kites in order to get acquainted with the physical laws of the universe and with the actions that someday they will really perform. Likewise, to read fiction means to play a game by which we give sense to the immensity of things that happened, are happening, or will happen in the actual world. By reading narrative, we escape the anxiety that attacks us when we try to say something true about the world.
This is the consoling function of narrative—the reason people tell stories, and have told stories from the beginning of time. And it has always been the paramount function of myth: to find a shape, a form, in the turmoil of human experience.
Nevertheless, the situation is not so simple. Until now my talk has been haunted by the ghost of Truth, and you must admit that this is not a notion to be taken lightly. Usually we think we know pretty well what it means when we say that something is “true” in the actual world. It is true that today is Wednesday, it is true that Alexanderplatz is in Berlin, it is true that Napoleon died May 5, 1821. On the basis of such a concept of truth, scholars have widely discussed what it means for an assertion to be “true” in a fictional framework. The most reasonable answer is that fictional statements are true within the framework of the possible world of a given story. We assume it is untrue that Hamlet lived in the actual world. But let us say we are grading the term paper of an undergraduate majoring in English literature and we find the wretched student has written that at the end of the tragedy Hamlet marries Ophelia. I bet that any reasonable teacher would claim the student has said something untrue. The statement would be untrue in the fictional universe of Hamlet, just as it is true in the fictional universe of Gone with the Wind that Scarlett O’Hara marries Rhett Butler.
Are we sure that our notion of truth in the actual world is equally strong and clear cut?
We think we usually know the real world through experience; we think it is a matter of experience that today is Wednesday, April 14, 1993, and that at this moment I’m wearing a blue tie. As a matter of fact, it is true that today is April 14, 1993, only within the framework of the Gregorian calendar, and my tie is blue only according to the Western division of the chromatic spectrum (it is well known that in the Latin and Greek cultures the boundaries between green and blue were different from the ones that obtain in our own culture). At Harvard, one can ask Willard Van Orman Quine to what extent our notions of truth are determined by a given holistic system of assumptions, Nelson Goodman about our many different ways of worldmaking, and Thomas Kuhn about the notion of truth with respect to a given scientific paradigm. I hope they would admit that it is true Scarlett married Rhett only in the universe of discourse of Gone with the Wind, just as it is true I am wearing a blue tie only in the universe of discourse of a given Farbenlehre.
I do not want to play the role either of a metaphysical skeptic or of a solipsist (it has been suggested that the world is overpopulated with solipsists). I realize there are things we know via direct experience, and if one of you told me that an armadillo had appeared behind me, I would turn around instantly to see whether the information was true or false. I think we can all agree that there are no armadillos in this room (provided we agree on the socially accepted zoological taxonomy). But usually our struggle with the notions of truth and falsity is more complicated than that. We know now that there are no armadillos in this room, but in the coming hours and days such a truth will become a little more arguable. For instance, when these lectures of mine are published, readers will accept the idea that on April 14, 1993, there weren’t any armadillos in this room and will do so not on the basis of their own experience, but on the basis of their conviction that I am a serious person and that I have accurately reported the situation in this ro
om on April 14, 1993.
We believe that, so far as the actual world is concerned, truth is the most important criterion, whereas we tend to think that fiction describes a world we have to take as it is, on trust. Even in the actual world, however, the principle of trust is as important as the principle of truth.
I don’t know through experience that Napoleon died in 1821. Moreover, if I had to depend on my own experience I couldn’t even say that Napoleon ever existed (as a matter of fact, somebody once wrote a book to demonstrate that Napoleon was a Solar Myth). I don’t know through experience that there is a city called Hong Kong, and I don’t even know through experience that the first atomic bomb worked by fission and not by fusion; I actually don’t know very much about how atomic fusion works. According to Hilary Putnam, there is a “linguistic division of labor” which corresponds to a social division of knowledge: I delegate to others the knowledge of nine-tenths of the real world, keeping for myself the knowledge of the other tenth.7 In two months I really will be going to Hong Kong; I’ll buy my ticket certain that the plane is going to land in a place called Hong Kong, and I’ll thereby manage to live in the real world without having to behave neurotically. I’ve learned that for a lot of things, I’ve been used to putting my faith in other people’s knowledge. I confine my doubts to some specialized sector of knowledge, and for the rest I put my trust in the Encyclopedia. By “Encyclopedia” I mean the totality of knowledge, with which I’m only partly acquainted but to which I can refer because it is like an enormous library composed of all books and encyclopedias—all the papers and manuscript documents of all centuries, including the hieroglyphics of the ancient Egyptians and inscriptions in cuneiform.
Experience, and a long series of decisions by which I have placed trust in the human community, have convinced me that what the Total Encyclopedia describes (quite often in contradictory ways) represents a satisfactory image of what I call the real world. In other words, the way we accept the representation of the actual world scarcely differs from the way we accept the representation of fictional worlds. I pretend to believe that Scarlett married Rhett, just as I pretend to take as a matter of personal experience that Napoleon married Josephine. Obviously the difference lies in the degree of this trust: the trust I give Margaret Mitchell is different from the trust I give historians. Only when I read a fable do I accept that wolves speak; the rest of the time I behave as if the wolves in question are those described by the latest International Congress of the Zoological Society. I do not want to discuss here the reasons I put more trust in the Zoological Society than in Charles Perrault. These reasons exist and are pretty serious. But to say that these reasons are serious does not mean that they can be clearly spelled out. On the contrary, the reasons for which I believe historians when they tell me that Napoleon died in 1821 are far more complex than the reasons for which I am sure that Scarlett O’Hara married Rhett Butler.
In The Three Musketeers we read that Lord Buckingham was stabbed by one of his officers, called Felton, and so far as I know this is considered a historical truth; in Twenty Years Later we read that Athos stabbed Mordaunt, the son of Milady, and this is considered a fictional truth. That Athos stabbed Mordaunt will remain an undeniable truth so long as there exists a single copy of Twenty Years Later—even if in the future someone invents a post-post-structuralist way of reading. In contrast, a serious historian must remain ready to assert that Buckingham was stabbed by someone else, if by chance a future researcher in the British Archives proves that all previously known documents are false. In such a case we would say that it is historically untrue that Felton stabbed Buckingham, but the same fact would remain fictionally true.
Apart from many important aesthetic reasons, I think that we read novels because they give us the comfortable sensation of living in worlds where the notion of truth is indisputable, while the actual world seems to be a more treacherous place. This “alethic privilege” of fictional worlds also provides us with some parameters for challenging farfetched interpretations of literary texts.
There have been many interpretations of “Little Red Riding Hood” (anthropological, psychoanalytical, mythological, feminist, and so on), in part because the story exists in several versions: in the Brothers Grimm text there are things that are not in Perrault’s, and vice versa. It was reasonable to expect an alchemical interpretation, as well. In fact, an Italian scholar has tried to prove that the fable refers to the processes of extracting and treating minerals. Translating the fable into chemical formulas, he has identified Little Red Riding Hood as cinnabar, an artificial mercury sulfide which is as red as her hood is supposed to be. Thus, within herself, the child contains mercury in its pure state, which has to be separated from the sulphur. Mercury is very lively and mobile, and it is no accident that Little Red Riding Hood’s mother warns her not to go poking about everywhere. The wolf stands for mercurous chloride, otherwise known as calomel (which means “beautiful black” in Greek). The stomach of the wolf is the alchemist’s oven in which the cinnabar is transformed into mercury. Valentina Pisanty has made a very simple comment: if, at the end of the story, Little Red Riding Hood is no longer cinnabar but mercury in its pure state, how can it be that when she steps out of the wolf’s belly she’s still wearing a red hood? There is no version of the fable in which the little girl steps out wearing a silver hood. So the fable doesn’t support this interpretation.8
You may infer from texts things they don’t explicitly say—and the collaboration of the reader is based on this principle—but you can’t make them say the contrary of what they have said. You can’t ignore the fact that Little Red Riding Hood at the end is still wearing her red hood: it is precisely this textual fact that exempts the model reader from being obliged to know the chemical formula for cinnabar.
Can we rely on the same degree of certainty when we speak of truth in the actual world? We are sure that there are no armadillos in this room to at least the same extent we are sure that Scarlett O’Hara married Rhett Butler. But for many other truths we must rely on the good faith of our informers, and sometimes on their bad faith. In epistemological terms, we cannot be sure that Americans landed on the moon (whereas we are sure that Flash Gordon reached the planet Mongo). Let’s for a moment be extremely skeptical (and mildly paranoid): it could have happened that a small bunch of conspirators (say, people from the Pentagon and various TV channels) organized a Big Fake. We—I mean, all other TV watchers—simply trusted those images telling us that a man had landed on the moon.
There is, however, a strong reason that makes me believe Americans really did reach the moon: it is the fact that the Russians did not protest and did not make any accusations of fakery. They had the ability to prove that it was a hoax and they had every good reason to do so. They did not. I trusted them, so I strongly believe that Americans reached the moon. But in order to decide what is true or false in the actual world, I must make some difficult decisions about my trust in the community. Furthermore, I must decide which portions of the Total Encyclopedia are to be trusted, while rejecting others as unreliable.
It seems that with fictional truths things go easier. Even a fictional world can be as treacherous as the actual one, however. It would be a wholly comfortable environment if it had to deal only with fictional entities and events. In that case, nobody would have any anxious moments over Scarlett O’Hara, because the fact that she lived at Tara is easier to check than the fact that Americans landed on the moon.
But we have ascertained that every fictional world is based, parasitically, upon the actual one, which the fictional world takes as its background. We can skip a first question—namely, what happens when the reader brings into the fictional world wrong information about the actual world. We can assume that such a reader does not behave like a model one, and the consequences of this remain a private and empirical affair. If someone reads War and Peace believing that in the nineteenth century Russians were governed by the Communist party, it will be hard for him or her to u
nderstand the story of Natasha and Pierre Besuchov.
I have said, though, that the profile of the model reader is designed by and within the text. Obviously Tolstoy did not feel obliged to inform his readers that the battle of Borodino wasn’t fought by the Red Army, but he provided his readers with enough information about the political and social situation of czarist Russia in that period. Don’t forget that his novel opens with a long dialogue in French, and this tells the reader a lot about the situation of the Russian aristocracy at the beginning of the nineteenth century.
In fact, not only are authors supposed to take the actual world as the background of their story, but they constantly intervene to inform their readers about various aspects of the actual world they may not know.
Suppose that Rex Stout in one of his novels were to say that Archie, hailing a cab, asks the driver to take him to the corner of Fourth and Tenth Streets. Suppose further that Rex Stout’s readers fall into two categories, those who do not know New York and those who do. Let’s disregard the first category—they are eager to swallow everything (in Italian translations of American detective novels, such expressions as “downtown” and “uptown” are regularly translated as “città alta” and “città bassa”—“high city” and “low city”—so that most Italian readers think American cities are all like Tiflis, Bergamo, or Budapest, half on the hills and half on the plain or along the river). But I think that most American readers, knowing that New York City is like a world map where the streets are the parallels and the avenues are the meridians, would react like that reader to whom a hypothetical Nerval said that the coach was not drawn by a horse. As a matter of fact, there is in New York (in the West Village) a point where Fourth Street and Tenth Street intersect, and all New Yorkers know it, except the taxi drivers. I believe, however, that if Stout had had to narrate this event, he would have explained this fact (maybe by inserting an amusing comment) and the reason this intersection can really exist, being afraid that a reader from San Francisco, Rome, or Madrid might not be aware of this and might think that Stout was joking.
