Conversazione con André Martinet
André Martinet/1 Communication is our basic relevancy
André Martinet/2 Language articulates what we feel into a succession of items
André Martinet/3 Pregiudizi linguistici
André Martinet/4 Cosa c'è dietro le parole
As it seems that there is a general consensus on the importance of the study of phonology in the Nineteenth Century and, conversely, as it is not the same for the study of syntax, which has a very complex and fragmentary background, I would like to know why you chose the term syntax for that part of your study dedicated to the analysis of the monemes?
No, first of all, you seem to hint upon the problem of the connection betwwen the phonology side of the picture and the rest of the picture.
Now, I would say the first thing we have to remember is that in contraddistinction to practically all linguists and in any case in contraddistinction to the hjelmslevians, I don't operate with isomorfic systems. I don't think that whatever is good for the signifiant is good for the signifié (they oppose signifiant-signifié in the saussurian way).
I have a sign and my sign has signifiant-signifié and my point of view is that they shouldn't be put on the same level, because as soon as the unit is a significant unit, the meaning of that unit is decisive, not the form. While those people who operate with signifiant-signifié think that phonology covers the signifiant and, well, the rest should be covered by I don't know what name, but that doesn't make any difference. You know, with the double articulation you start from the sentence, that is, a text or a segment of a text, in which the relationships have a status. They have sense within this status.
So, that's the first part of it and then the second part of the double articulation is the second articulation, which is very different, which has no bearing upon meaning at all. I see no relationship between the two sides, not the way the hjelmslevians do, because they want to find on the one side what they find on the other side. I don't care about that. I don't care about presenting something which is not symmetrical. Nature is not symmetrical and language is not symmetrical, because if you have the double articulation, if you know very well that on the one hand you have a lot of units, it is necessary to have a lot of distinct significant units, because you have to describe the world, and on the other hand you have a minimal number of distinctive units because those are the tools, plain tools with no function, no value, except in reference to expressing the world.
You see, we should really try to understand what we are operating with. What strikes me is... take for example something I mentioned yesterday*: my objection to Hjelmslev is that, because in a language like latin and most European languages, substantives can be used in the singular or as a plural, and, at the same time, used in a certain case, he concluded that there was a natural relationship between case and number, which is ridiculous. It just happens to coincide there.
And this, of course, would not strike people if, instead of endings, you had had prepositions, because the prepositions would be there. And if the number had been indicated by means of numbers, you would again have something totally different.
So, from the point of view of signification, which is foundamental when we tackle problems of the first articulation, we should, of course, reckon with the fact that those coincide in the form... usually in such a way - a cohincidence is such - that you cannot tell what is number and what is case. Take for example rosarum in Latin. well, which is the stem? You don't even know. The stem would be rosa- and the ending would be -arum. What is the plural and what is the case? It's just amalgamation. Functionally... it is just morphology and we shouldn't fall victims to morphology. And that's what those people, those hjelmslevians, do all the time.
They insist on the substance. I don't mention substance, because I know very well I would have to operate with substance at a certain point. if I don't operate with substance, I cannot tell what's the difference between this and that. They are so abstract in their approach and they don't care about it. At a certain point they don't care about saying what distinguishes this from that and, later on, they have to create a new science in order to explain why finally they reckoned with substance. There you are. And the very fact that they create something they call semiotics, induces a lot of people to prepare a semiotic which has nothing whatever to do with what real semiotics should be.
Have you read my wife's book La Clée puor la semiology? She presents the writing system in a semiotic system. We, as linguists, should consider a language as basic and everything connected with language should belong to linguistics and not to semiotics. Semiotics is the science where all the systems which may be used for communication are studied and compared with each other.
So, semiotics is interdisciplinary...
Exactly, but we linguists start with languages. We should be interested in linguistics and not try to start with semiotics before we have finished with languages. Languages are a semiotic system, but a very special one. It is the only one where you can find just those elements which we present in the definition of a language.
*Comunicazione di André Martine al convegno Hjelmslev oggi: Une relecture de Hjelmslev, San Marino 12 ottobre 1993