"You have not read about thirty pages and
you're becoming caught up in the story"¹.

"The very existence of texts that can not only be freely interpreted but also cooperatively generated by the addressee (the 'original' text constituting a flexible type of which many tokens can be legitimately realized) [...]"².
  This is how The Role of the Reader by Umberto Eco opens. It is a very dense incipit, interesting for the many statements that are made and the others taken for granted, that are implied. Let us try to analyze them one at a time, seeking out their implications for the reader-translator.
  Up to now, we have spoken of reading as an intersemiotic translation process, having a verbal text as a prototext and a mental text as a metatext. When Eco states that the reader is able to generate texts - by her interpretive cooperation - it is obvious that the notion of "text" is far decentralized from the concept that envisions it as a consistent set of signs printed on paper or emitted in the air.
  Interpretive cooperation is part of the text and the text is not complete without taking into account if, when and how it is interpreted.
  Parenthetically, Eco then implicitly uses the translation metaphor when he speaks of the original and its actualizations, or tokens. Reading is a translation process for Eco too, who sees the prototext as a type and its interpretations or sense-making readings as actualizations of that type.
  If the meaning of the text is complete only when the text is read, it is clear that a well-rounded writer tries to prefigure the model of her reader. In this way, she imagines what can be the model of actualization of her text. Otherwise, the sense of her text is trusted to the casual meeting of her text with any empirical reader.
  By "empirical reader" we mean a given concrete reader reading a text, one of the many concrete actualizations of the abstract notion of "reader". "Model reader" is instead the one that, apart from the author, is able to interpret a text in a similar way to the author who generated it.

Here's the tricky part: this point is not applicable only to literary texts, as superficially one could be led to believe. The choice of the model of reader is implicitly made by choosing the language in which the text is coded, in what style and with what register, and with what degree of specialization. For example, the text by Eco we have quoted from many times foresees a model of reader much more specialized (a researcher in semiotics, for example) than this translation course, addressing a much wider public.

Some texts are interpreted according to the author's prediction, while others are decoded in cultural contexts completely different from the ones that have been foreseen. Some authors foresee decoding under different conditions from those postulated by their own coding strategy, while others do not consider such eventuality, however frequent it may be. Eco proposes differentiating authors according to this criterion.

In other words, a closed text foresees only one form of decoding. All those not foreseen are not 'legitimate', from the author's point of view. If only the reader refers to cultural conventions different from the ones stiffly implied, decoding produces results that are completely different from those conceived by the author's strategy. In the quoted sentence, Eco says that closed texts are the most 'open': it is a play on words of course, a little provocation. The stiffer a set of rules is, the greater the possibility to transgress. By analogy, the narrower a narrative strategy, the more probabilities there are to encounter unforeseen decoding, which actually renders these texts extremely open.
  Vice versa when the author can conceive a sufficiently flexible model of reader within the strategy with which she produces the text to match a high number of empirical readers, this amounts to saying that the openness of a text is an intrinsic, genetic trait of that text. Consequently, the field of legitimate decoding is much wider and, in the same time, the limit of aberrant decoding is much more rigid. In this case, the text is however much more closed to decoding not foreseen by the flexible textual strategy.

  

Bibliographical references

CALVINO I. If on a Winter's Night a Traveller, translated by William Weaver, London, Vintage, 1998, ISBN 0-7493-9923-6.

ECO U. The Role of the Reader. Explorations in the Semiotics of Texts, Bloomington, Indiana University Press, 1995, ISBN 0-253-20318-X.

---

¹ Calvino 1998 p.25.
² Eco 1995, p. 3.
³ Eco 1995, p. 7.
⁴ Eco 1995, p. 12.
⁵ Eco 1995, p. 8.