The art of creating a legend

What distinguishes 'literature' from 'light fiction'? Umberto Eco looks to the past for an answer

Saturday July 20, 2002
The Guardian

I've read that there have been animated discussions in France over the protests of the town of Villers-Cotteret - the birthplace of Alexandre Dumas - at having the ashes of their author moved to the Panthéon in Paris. I fear that in Italy, many would also protest if this great popular narrator (it's a bit of a stretch to ascribe to him this kind of canonisation) were to be buried next to those who are already canonised by way of scholastic decree. But in truth, we are not the only ones who have a difficult time discriminating between literature and the so-called "light fiction".

Certainly, light fiction exists and encompasses mysteries or second-class romance novels, books that are read on the beach, whose only aim is to entertain. These books are not concerned with style or creativity - instead they are successful because they are repetitive and follow a template that readers enjoy.

If this is the case, then did Dumas aim to write light fiction, or did he not even worry about such things - as some of his critical and controversial writings would suggest? He had "slaves" who helped write numerous books and he wrote lengthily to earn more money. But with some works, he was able to create characters we can define as "legendary," who populated the collective imagination, and who are copied and retold as happens with such characters of legend and fairy tales.

Sometimes he succeeded in creating a legend by pure literary ability: The Three Musketeers is quick; it reads like a sheet of jazz music and even when he produces those dialogues, which I have defined as "piecemeal dialogues": two or three pages of short and unnecessary quips (which he does merely for length), Dumas does it with "boulevardier" grace.

And what about The Count of Monte Cristo? I have written previously about how once I decided to translate it. I would find phrases such as: "He rose from the chair upon which he was sitting." Well, which other chair should he have risen from, if not from that upon which he was sitting? All I had to say in my translation was, "He rose from the chair", or even "He rose", as it is already clear he was sitting at a table.

I calculated that I had saved the reader at least 25% reading time by shortening Dumas's language. But then I realised that it was exactly those extra words and repetition that had a fundamental strategic function - they created anticipation and tension - they delayed the final event and were fundamental for the excellent vendetta to work so effectively.

That this was Dumas's great narrative capacity is clear today in rereading his contemporary, Eugene Sue, who at the time was more famous than Dumas. If we reread The Mysteries of Paris - which produced collective hysteria through character identification and also offered political and social solutions - we realise that the added words and phrases make the book heavier than lead to read, and we can read it only as a document, not as the novel it was intended to be.

Therefore, are there virtues in writing which are not necessarily identified with linguistic creation, but are part of rhythm and shrewd dosage, and cross the boundary, albeit infinitesimally, between literature and light fiction? The novel, like a legend, begins in the language, in the sense that Oedipus or Medea are typical characters and are exemplary simply because of their actions even before they become the great Greek tragedies. Similarly even Red Riding Hood or the characters in African or Native-American mythology function as models of life beyond the poetry which overtakes them and creates another layer to them.

Does the novel have to deepen the psychology of its heroes? Certainly the modern novel does, but the ancient legends did not do the same. Oedipus' psychology was deduced by Aeschylus or Freud, but the character is simply there, fixed in a pure and terribly disquieting state.

Particularly in Italy, we are led to identify the novel with prose as art and by a short circuit with poetry, a kind of "proetry". And yet Stendhal used the prose of the civil code; Italo Svevo, it is said, wrote badly, and if we want something "poetic", there is more poetry in Liala than in Alberto Moravia.

The problem is that the novel must "tell a story" and enliven exemplary characters even if it only describes their external behaviour. The psychology of D'Artagnan is amusing, but the character becomes legendary. The psychology of Julien Sorel is complex and therefore I agree that there is a distinction between the historical novel, which makes us understand an entire era through its heroes, and a cloak-and-dagger novel, which takes place in a certain time period but could have easily taken place in another era and would have remained equally appealing.

But here we are not talking about works of art whose greatness and density of layers no one disputes. We are talking about mythical writings, which are another thing. Fundamentally, Pierre Souvestre and Marcel Allais also lengthened their works to make more money. Their stories of Fantomas are not an example of exalted writings, and yet the man became an urban legend who obsessed the Surrealists and others. The Rocambole of Pierre-Alexis Ponson du Terrail still entertains us, but he has not become a legend.

Why? There are dazzling original models and narrative strategies that still need to be studied and compared in depth.

© 2002 Umberto Eco