just as written letters are not the same for all humans neither are spoken words. But what these are in the first place signs of, the affections of the soul, are the same for all; and what these affections are likenesses of, actual things, are also the same.
(De Interpretatione, 2002, 16a3)
This much-quoted passage summarises not only Aristotleâs theory of language but also the concept of language generally attributed to the classical paradigm.
Generalising somewhat, the classical paradigm is therefore commonly associated with the idea that thought comes first and is not only universal but also independent from language, whereas language is secondary to thought. And although these views tend to be dismissed as simplistic, they have left a permanent imprint on our mainstream notions of language and, consequently, translation.1 The very concept of equivalence, every translatorâs good old friend and worst foe, is an offshoot of this classical conception of language (more in Section 3.4).
In these classical theories, language is thus seen as referential and universal. And, decisively for our purposes here, they place the source of meaning outside language. In Cratylus (Plato 1921, 439b), Plato even cautions us to seek knowledge outside language, here in Harold Fowlerâs translation: ârealities are to be learned or discovered ⊠not from names but much better through themselvesâ.
Therefore, within the classical paradigm, there was no need for multiple languages â we might as well say that there was no linguistic diversity as we understand it today. For the Greeks, the only true language was Greek â non-Greek-speakers were barbarians [from ancient Greek ÎČÎŹÏÎČαÏÎżÏ, an onomatopoeic representation of dissonant, meaningless sounds]. For Romans, initially there were only Greek and Latin; later, only Latin. And although there was significant diglossia2 for the learned, with a vernacular being used at home and Latin in âhigherâ domains of society, Latin was perceived as universal, the only true language (Trabant 2010, 883â884). Greek, and then Latin, were seen as an instrument to articulate meanings which stemmed from a universal source. This linguistic monocentrism prevailed in Europe until the late Middle Ages.
Later theories of language went on relentlessly trying to pinpoint this universal source of meaning, with some attributing it to God, others to a âthird realmâ between our subjective ideas and reality, yet others to a universal language underneath all natural tongues. There have been numerous attempts both to systematise natural languages, thus ridding them of their âmessinessâ, and to devise fully logical artificial languages or metalanguages. Underpinning these projects is the belief that speaking multiple languages is a nuisance, a murk that obfuscates true meaning, a pointless enterprise, for surely there is no need to employ multiple instruments to represent the same meanings. We will investigate some of these projects in the coming sections.
The legacy of the classical paradigm finds its most influential and lasting expression in Ferdinand de Saussureâs theory of language â though it must be noted that Saussure broke away from these universalist traditions of direct association between language and reality. In fact, he emphasised that languages are not mere lists of terms, and that there is no reality, no pre-established ideas before and outside language; put differently, for Saussure, each language articulates its own meanings. This notwithstanding, he retained two key binary oppositions, which in turn might explain the lasting influence of universalist ideas on our mainstream notion of language (Rodrigues 1999, 187; Lages 2007, 211). The first is form versus substance, whereby âformâ can be understood as languages or words, whereas âsubstanceâ refers to meaning. The second dichotomy is speech versus thought, along with the idea that speech is the direct representation of thought (see Saussure 1986, 115â117). Both dichotomies are crucial to any universalist notion of language (and translation; see Section 3.4), as we will see in what follows.