Site de Jean-Pierre Depétris


Sign, formal and natural languages



Mon travail

À Travers Champs

L'Ecriture Créatrice


Write Different


email me

What is a sign?

    Everybody knows what a sign is, and has no difficulty to use this word. It is not so easy to explain it. A wet ground may be the sign that rain has fallen. Rain falling may be as well the sign that ground is wet. This example show that a sign, the meaning of a sign, has some relationship with causality, but, by contrast with causality, it works the two ways. If, obviously, rain can be the cause of the wet ground, wet ground can't absolutely be the cause of the rain, however, it may be a sign for it.
   Signs allow us, then, to browse through causal chains, and so to anticipate them.

Notion of meaning system

    There are many definitions of the word "sign" corresponding to different disciplines, but I prefer to let it remaining the most usual and general meaning.
   A cloud may be sign for rain. It is a sign only for who interprets it as a one, that it would be true or wrong. By itself, the cloud may be nothing more than a cause.
   The cloud which one is a sign for me may be a one for another. I show him the cloud and, even if he doesn't know my language, he walks faster, that is a sign for me he made the same interpretation. In this example, there is no agreement according to any convention. A pragmatic relationship with facts is enough. The meaning system is wholly supported by the actual climatic system.
   No meaning system, probably, as sophisticated as it may be, could emancipate itself from such a pragmatic relationship with facts. For example, the word "cloud" lets me evoke the cloud even when I have no one to show; however, I must be able, as my interlocutor, to see a sign in the actual cloud in order to let my sentence provide a meaning.
   I might tell somebody: "look at the clouds", and he might answer me "what about?"
   I could explain him that those clouds mean that rain is coming, but he might shrug his shoulders, or do something like that, and mind I joke or relate to some superstition.
   May we conclude that he understands me better than if he wouldn't know my language, and would walk faster as soon as he should glance at the sky?

   Scales and the concept of weight are another example. How could we conceive what a weight is, without having discovered the lever and its specific application that is the scales? What would be the meaning of weight otherwise?

What a language is.

   A language is a system of signs. This systematic aspect is important, and is specially pregnant with the function of communication. So, man often concludes that communication is the main function of language. I rather mind that communication is not a function of its own, but its condition.
   In my example, the cloud is for me a sign of the rain, even if I am alone and if I don't communicate with anybody. The climate is itself systematic enough. What is the matter if what is a sign for me is not a one for another? The coming or not of the rain is already an answer. The pragmatic relationship that signs manage with facts is widely able to provide answers, feed back, without having to suppose any interlocutor nor message.

   I pierce holes in a reed, and I play notes. I can establish a systematic relationship between the distances separating the holes, their physical characters in general, and the eight of a sound. To do that, I need nobody.
   It is only when such relationships are perceived, created, discovered, invented... that a communication between interlocutors can occur and be conceived. The systematic nature of signs articulation in a language is soon widely maintained by the systematic articulation of facts in actuality. Agreements with common conventions are less needed than man often minds. For example, the number of days in a year is not given by convention.
   Then, communication would be rather a cooperation in order to build such systems. Their construction would not be conceivable without cooperation.

What natural languages are?

   Natural languages are specific forms of languages, composed with some dozens of signs, according to a principle of dual articulation.

   What is a dual articulation? Signs that compose the system have no meaning by themselves, they constitute a first level, which are the phonemes, with which we build the signs of a second one, and by means of which we set up sentences providing meanings.
   A natural language has generally a sound form and a graphic one. We can imagine some others, tactile for example, as the writing for the blind, or knots on ropes, as used by ancient civilizations. In the first case, signs of first level are phonemes, in the second, they are letters.
   These two sets of first level signs may be equivalent. In Arabic, phonemes and letters are scarcely the same. In French, the 26 letters (to which special characters must be added) correspond to 36 phonemes, which may need two or three letters, and may be written differently (o, au, eau...)

   In fact, we have a dually dual articulation: a former, between two levels of signs, the ones used to compose the others, and another, between a phonetic structure and a graphic one.
   "Language" comes from the Latin word "lingua", which means "tongue", that lets us mind that it is phonetic before being writing. Many people can speak without being able to write, and not the contrary. However, writing is not only the graphic transposition of sound signs. Conversion obviousness and easiness hint that language is somehow independent of what kind of signs it uses. This easiness is moreover relative. We are sometimes in pain when we try to render in writing the resources handled by speech, or the opposite. How to write the kindness of a tone? How to pronounce italics?

Former page
See also: Actual functioning of thought