Who Controls the Text?

Umberto Eco's second lecture in the Ellmann series was entitled "Author, Text, and Interpreters." It was more tailored to the academic crowd present to hear him lecture than yesterday's, though it drew in part on the groundwork he laid then.

According to Eco, one of the questions he most has to answer as an author when dealing with his books' translators has to do with perceived ambiguities in the text. He has three sorts of response: (1) "keep it in, I meant it to be that way;" (2) "keep it in, I didn't mean it but it's useful;" or (3) "iron it out, I didn't mean it." His position, for the main, is that creative writers should never interpret their own texts. The reader shouldn't appeal to the author's intention, but to the evidence of the text.

Eco borrows the term "unlimited semiosis" from the American logician C.S. Peirce to express his approach to what he calls the "open text." Peirce says: ""The meaning of a representation can be nothing but a representation ... the interpretant is nothing but another representation ... and as representation, it has its interpretant again. Lo, another infinite series." The text is susceptible to an infinite number of uses (apparently recursive uses). What, then, are the criteria that limit interpretation? Eco draws on the Augustinian notion that you can check the validity of an interpretation by reference to the text. Any interpretation that is incompatible with the text must be rejected.

Eco also distinguished the semiotic approach to the text from deconstruction. Briefly, semiotics holds that the text is a series of multiple, perhaps infinitely many, signs (as indicated above). Deconstruction, to the contrary, holds that the text is mere signifier. And, as we all know, the sign is comprised of the signifier/signified relation.

Eco cops to being an allusive writer. Indeed, in his first lecture he stated outright that he was explicit in his use of intertextuality, and more specifically, "double codes". He loves to drop clues for the "paranoid reader" to track down. However, there are limits to intertext. Thus, on a personal note, he found his early books The Name of the Rose and Foucault's Pendulum opened up a number of chains of associations in their readers. Some of those associations he felt were intended and embodied in the text and, thus, valid. Others, were unintended and unsupported by the text and, accordingly, invalid. There were others, however, which were unintended yet were supported in the text and, reluctantly, valid.

The delight of the lecture (for me) was Eco's anecdotal examples of dealing with his readers. Specifically, he recounted a couple instances of his unconscious inserting references into his text that were unavailable to his conscious creative mind. In particular, after Name came out, he found an old text by Riccoboni (sp?) on a dusty top shelf in his study. It had page corners darkened by Eco's eager fingerprints. In this book, the author had tried to reconstruct from the available evidence the second part of Aristotle's Poetics, i.e., the disappeared "Comedy". Eco had pored over this book as a young man, but had completely forgotten about it when writing Name. In another instance, he unconsciously used the name of a Brazilian singer of the pop song "Guantanamero" as a character in Pendulum. When he discovered this, he remembered having a crush on her at the time.

What was remarkable about this lecture was Eco's ability to maintain a critical detachment from his own creative works, to treat them as textual objects in order help locate the point were meaning must be fixed in the text without violating the text. Yet, he refrained from providing an interpretation of his own works. And he refrained from asserting that the text and its interpretation constituted a perfect system—there must necessarily be a leakage of significance.

In answer to the question we posed above, the text controls both its author and its interpreters. In our own terms, the text not only is battleground in the struggle between the writer and the critic, it is the ultimate proving ground. Intention as well as interpretation find their empirical constraints within its bounds.