In the previous unit, we have seen that in general the translation activity must deal with two elements: the cultural distance and linguistic distance between the text to be translated (prototext) and the language/culture of the text to be produced (metatext).

  The practical consequences of such a view of the translation activity are manifold. First, in the education of a translator there must be, besides linguistic expertise, a specific knowledge of one or more cultures of the relevant linguistic areas. We chose the English language to make an example.

  In many non-English-speaking countries, the most widespread foreign language taught in the schools is English and, within the framework of this subject, some elements of British culture are also taught. In colleges and universities, the cultural subjects related to the English-speaking countries deal mostly with British culture. This knowledge is necessary for the prospective translators who will have to deal with British texts.

  If, on the contrary, the prospective translation texts belong, for example, to postcolonial literatures, or to American-English literature, a cultural background regarding those countries is required. Otherwise, the translator will only be able to accomplish that part of her work that has to do with linguistic transcoding.

  There is no general agreement or definition in the interaction between the language and the culture. According to the famous researcher B. L. Whorf, language is not so much a tool through which it is possible to express notions belonging to a culture, as it is a sort of cataloguing system, a systematization of otherwise disorderly knowledge. This view overturns the traditional conception in which, within the language/culture relation, language has the task of formulating acquired knowledge, no matter what the level of linguistic competence. Following Whorf's line of thought, it is language to which molds and systematizes knowledge, if two peoples or two persons speak different languages, they often have different world views, not simply different formulations for the same conceptions ¹. In M. Dummett's opinion, the existence of objects depends upon language; it is language that decides what types of objects are recognized as existent.

  Beyond that, Whorf's theory implicitly places much emphasis on mother tongue learning. In his view, through the mother tongue one learns the ways in which experience can be systematized. According to this theory, the learning of a foreign language also becomes learning of a different world view, of a different concept of culture. In Whorf's opinion, there is no knowledge without a mother tongue; therefore, there is no univocal knowledge in multilingual individuals.

  Beyond this general view of the language/experience/knowledge relations, Whorf's thought is not so pertinent to the specific interests of the translator because, when Whorf deals specifically with translation problems, he studies word-by-word translation². Within the framework of a semiotic theory applied to translation activity, the fact that the word "snow" in Eskimo covers the semantic field of many different English words is of no great interest, nor necessarily means that between an Eskimo and an outsider there must be differences with regards to intellectual processes; it means just that they have different cultural experiences.

  Whorf, therefore, provides us with a new captivating perspective, in which language is not a simple expression tool, but also and above all a cognitive tool. This does nothing to advance the understanding of the translatability conception, however, but for the fact that the interlingual translator should be able to understand a new world view for each lan-guage/culture she learns. In Whorf's opinion, implicitly, to translate means to transfer a Weltanschauung. Whorf does not give us any specific indication about translatability, while Sapir is more explicit about what is translatable and what is not.

  Sapir is definite in his cataloguing of texts in relation to translation. According to the famous linguist, non-linguistic art is translatable, while linguistic art is untranslatable. Another distinction Sapir makes regards the translatability of texts in which the layer where we intuitively catalogue our personal experience prevails (latent language content) and texts characterized by the specific nature of the language in which they are written. The former are of course more translatable, because they are less linked to the specific linguistic structure in which they are formulated³.

  The question of translatability is dealt with by Hjelmslev, who divides languages into two categories: restricted languages, like, for example, artificial mathematical languages, and unrestricted languages, like, for example, natural languages. According the Danish linguist, translatability is guaranteed between unrestricted languages (i.e. between natural languages), and also if we translate from a restricted language into an unrestricted language, but not vice versa:

  A linguist who has made important affirmations on translation theory and who can, therefore, be of great help in establishing what we are talking about when we discuss translation is W. V. Quine. Quine differentiates between home language, the language one speaks in his home, and native language, his mother tongue. Each person realizes early in life that the native language spoken by his countrymen does not always coincide with his home language, and, therefore, in order to understand, he is forced to submit the utterances he hears to a radical translation aiming at the differentiation of the meaning and the pronunciation of the same words depending on their occurrence within the family or in the wider circle of people using that language⁵.

  The fact that each word acquires a different pronunciation or meaning depending on the context in which it is empirically formulated, and the consequent impossibility to state the criteria for a single possible translation of each utterance, give rise to Quine's con-cept of indeterminacy of translation. If we reason that the home language is the one giving the energy to face the other speakers' language, we can become accustomed to the theoretical indeterminacy (polysemy) of linguistic meanings and translation becomes the main tool to learn language with its semantic nuances. A competent speaker is always a good "translator" too, especially in an intralingual and intracultural sense, even if this concept, for obvious reasons, cannot be extended to professional, interlingual translation.

  For Quine the concept of translation refers primarily to intralingual translation. In the next unit, we will examine the thought of other researchers as to the concept of translatability.

Bibliographical references

BROWN R. Words and Things. An Introduction to Language. New York, The Free Press, 1968.

HJELMSLEV L. Prolegomena to a Theory of Language Ed. or. Omkring Sprogteoriens Grundlæggelse, København, Festskrift udg. af Københavns Universitet, 1943.

QUINE W. V. Ontological Relativity, in The Journal of Philosophy, n. 65, p. 7, 1968.

SAPIR E. Language. An Introduction to the Study of Speech. New York, Harcourt, Brace and Co, 1921.

WHORF, B. L. Language, thought, and reality; selected writings. edited by John B. Carroll. Preface by Stuart Chase. Cambridge (Massachussets), Technology Press of Massachusetts Institute of Technology, 1956.