"Many feelings distress you as you leaf through these letters"¹.

We have seen some aspects of Saussurean and Peircean thought in the preceding units that, although contemporaries, did not get to know each other. Another language researcher and philosopher, the Austrian, Ludwig Wittgenstein (1889-1951), devised a theory of signification, it seems, independent of the writings of his two precursors. These are not quoted by Wittgenstein, even though we must say that his Philosophical Investigations, his last work, published posthumously in 1953, is written in the form of elaboration of thoughts, without a bibliography of any kind. The philosopher expresses his meditations, relying upon logical thought and not on explicit references to other authors.
  Since Philosophical Investigations is almost entirely devoted to the question of language and of meaning, we cannot overlook it when dealing with the semiosis of reading. We know how complex the interaction of ideas in the semiosphere can be, and we think it is important to try to synthesize the ideas of the most important thinkers on language and compare them, whenever possible, using a common metalanguage that the original works do not have². We will start touching upon the topics that concern us most closely.
  Wittgenstein reflects on the nature of words, affirms that to say that every word means something is the same as saying nothing at all. Words have different functions, as diverse as the functions different tools may have; and there are similarities between the two cases compared³.
  While trying to understand how a word could be understood, the Austrian philosopher distinguishes words used to signify something from words used to signify words themselves. Here is the example he gives:

Such distinction is very important, even though it is often overlooked in practice. In the quoted example, the first "the" is inserted into the utterance as a 'normal' word, as object language, i.e. it has the function of completing the utterance so that it can generate a semiotic act along the sign-interpretant-object path.
  But the case of the second "the" is very different, which is written in quotes exactly to demonstrate that it is not used to trigger a normal semiotic act, but a metalinguistic act, a simple sign-sign reference. We have already talked about that in the first part of the course: it is called autonimy, a word formed by the prefix "auto-" and by the Greek root "ónoma", meaning "name". As a matter of fact it is a word that names itself, that refers to itself, transgressing all signification rules we have come across up to this point.
  And quotes, the delimiters of autonimy, have the precise aim of delimiting the portion of text that does not have a meaning in reference to the world, but has a meaning in reference to signs themselves. So, going on to the practical aspect of the question, it is important that the instances of autonimy are indicated by delimiters. Of course, this rule is in force in every language; it is, therefore, also a universally valid the principle that, when we find an instance of autonimy without delimiters in a text to be interlingually translated, we don't have to add them, because it can be a mark of the author's style. This is not true for purely denotative texts (for example a handbook of instructions), where missing delimiters can be attributable to a simple oversight.
  Another important intuition concerns reading, above all the reading aloud of a written sentence, which is compared to a vocal execution (singing) from a musical score. Wittgenstein speaks of the difference between exact matching of score and singing and lack of correspondence between written text and "'meaning' (thinking) the sentence"⁵. From that we understand how much room this view of language leaves to individual interpretation, to what Peirce would call "interpretant".
  Another very important notion is that of "language-game", a term by which he calls the various types of language, to stress that speaking a language is part of an activity, it is a form of life. Here are some examples of language-game:

As we see, translation is included among the language-games. But how does one proceed in order to attribute meaning to words? One of the procedures defined as not valid is the one of definition by antithesis. He makes the examples of the adjectives "red" and "modest", both impossible to be defined as the contrary of their negation: "not red" and "not modest"⁷.
  Natural languages, differently from the artificial ones like mathematics, are anisomorphic. This means that there is no one-to-one correspondence between meanings and words. Consequently, even notions as "synonym" or "opposite" are rather out of place in a natural language. The network of the potential match is too complicated to give the possibility what the antonym of a word is, especially if one does not specify from which point of view it should be the antonym.
  Saying that "modest" is the contrary of what is not modest, Wittgenstein stresses, is not necessarily wrong, but it is, at least, ambiguous. A fundamental warning that, however, we do not find when we look a word up in the synonyms and antonyms.
  To learn a meaning is compared to the learning of a game, particularly chess. There can be at least two procedures: in the former case, someone explains the rules of the game to the person who wants to learn, and later there is a practical experience. In the latter case, someone observes games of chess not knowing any of the rules and, on the basis of past experiences with other similar games and of her observations, she reconstructs the rules. We think that the latter case is very close to the abductive process as described by Peirce, in which one has to reconstruct a case (the sense of a move) based on hypotheses of rules and results.
  To close this initial overview of Wittgenstein's thought on meaning, we quote one of the first, fundamental conclusions at which he arrived:

This statement makes one think that the denotative meaning that we find easily in a dictionary can be applicable to the little group of remaining cases. In every other case, the meaning is defined by use, and the translator, the reader, as a meaning hunter, must concentrate herself more on the acts of parole in a Saussurean sense than on denotative, codified, fixed meaning available, as we were saying, in a dictionary.

Bibliographical references

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

GORLÉE D. L. Semiotics and the Problem of Translation. With Special Reference to the Semiotics of Charles S. Peirce.Amsterdam, Rodopi, 1994. ISBN 90-5183-642-2.

TOROP P. La traduzione totale - Total´nyj perevod, edited by Bruno Osimo, Modena, Guaraldi Logos, 2000. ISBN88-8049-195-4.

WITTGENSTEIN L. Philosophische Untersuchungen Philosophical Investigations, translated by G. E. M. Anscombe, second edition, Oxford, Blackwell, 1958. ISBN 0-631-20569-1.