< previous page page_61 next page >

Page 61
day. The provocative self-evidence of my last argument suggests that we can prove it only by showing that any of its alternatives is selfcontradictory.
Let us suppose that there is a theory that literally (not metaphorically) asserts that every interpretation is a misinterpretation. Let us suppose that there are two texts, Alpha and Beta, and that Alpha has been proposed to a reader in order to elicit the textually recorded misinterpretation Sigma. Take a literate subject X, previously informed that any interpretation must be a misinterpretation, and give him or her the three texts Alpha, Beta, and Sigma. Ask X if Sigma misinterprets Alpha or Beta. Supposing that X says that Sigma is a misinterpretation of Alpha, would we say that X is right? Supposing, on the contrary, that X says that Sigma is a misinterpretation of Beta, would we say that X is wrong?
In both cases, to approve or to disprove X's answer means to believe not only that a text controls and selects its own interpretations but also that it controls and selects its own misinterpretations. The one approving or disproving X's answers would then act as one who does not really believe that every interpretation is a misinterpretation, since he or she would use the original text as a parameter for discriminating between texts that misinterpret it and texts that misinterpret something else. Any approval or disproval of X's answer would presuppose (i) a previous interpretation of Alpha, which should be considered the only correct one, and (ii) a metalanguage which describes and shows on which grounds Sigma is or is not a misinterpretation of it. It would be embarrassing to maintain that a text elicits only misinterpretation except when it is correctly interpreted by the warrant of other readers' misinterpretations. But it is exactly what happens with a radical theory of misinterpretation.
There is another way to escape the contradiction. One should assume that every answer of X is the good one. Sigma can be indifferently the misinterpretation of Alpha, of Beta, and of any other possible text. But at this point, why define Sigma (which is undoubtedly a text in its own right) as the misinterpretation of something else? If it is the misinterpretation of everything, it then is the misinterpretation of nothing. It exists for its own sake and does not need to be compared with any other text.
The solution is elegant, but it produces a small inconvenience. It destroys definitely the very category of textual interpretation. There are texts, but of these nobody can speak. Or, if one speaks, nobody can say what one says. Texts, at most, are used as stimuli to produce other texts, but once a new text is produced, it cannot be referred to its stimulus.

 
< previous page page_61 next page >