Séminaire HTL : Otto Zwartjes

Otto Zwartjes (Université d'Amsterdam)
Some remarks on the ‘grammaire latine étendue (GLE)’: How ‘exotic’ is Latin?

During the Renaissance period in Europe, grammars of the vernacular languages of the nations of Europe began to emerge. Besides Latin and Greek the study of Hebrew was added to the education programme of missionaries and academics. While these developments were taking place in Europe, the production of dictionaries and grammars began to increase in the New World and in Asia, often outnumbering what was being made in Europe (Zwartjes 2011: 4-7). The discovery and exploration of hitherto unknown territories and continents was accompanied by an expanding interest in languages in these territories, whereas in pre-modern Europe showed so little interest in the missionaries’ linguistic documentation. From the perspective of human thought, the political expansion was accompanied by an increasing knowledge of hitherto unknown species in flora and fauna, various societies, religions, and of course, linguistic structures. There seems to be a contradiction between what has been called “grammaire étendue” (Auroux 1994), referring to a Greco-Latin model which had a bad effect on missionary linguistics, since missionaries who applied this system too rigorously were applying concepts which were not developed for these languages (exogrammatisation, cf. Auroux 1994: 122). The model would not be sufficiently equipped in order to describe ‘exotic’ languages, the traditional metalanguage was not sufficiently ‘elaborate’ in order to cover unknown linguistic features, it was too ‘narrow’, it was a model which needed to be ‘extended’. On the other hand, it may surprise us why scholars did not pay much attention to the expression used by missionaries themselves: unanimously they observe that they “reduced” the linguistic structures to “arts”. Although the verb “reducir” has several meanings, (to subdue, to overpower, to put down) it also means “to reduce”, “to decrease”, to diminish”, or more neutrally “to convert”. When we study the languages of the world, Latin (and Greek) is not a very common language. If we compare the typology of Ancient Greek, we can observe that this language is relatively rich in ‘accidents’ (parepómena), according to Dionysius Thrax five for the noun, eight for the verb, seven for the participle, etc. (Auroux 1994: 178-179). It is obvious that for many languages to be described, there was no need at all to “extend” this system at this point. The opposite is true: Missionary grammarians attempted to “reduce” this too “broad” model to a more simplified one. In this paper we shall first describe how to situate a language as Latin or Greek in the context of the world’s languages, using the WALS as our main source. In the second place we shall analyse a representative corpus of Spanish grammars of indigenous Meso-American languages from different linguistic families. The objective is to demonstrate that these grammars were not exclusively products of what has been called “grammaire latine étendue”, adding hitherto unknown categories, but rather an adaptation in which the quantity of ‘accidents’ was drastically ‘reduced’, whenever they were not relevant in these languages, i.e. when these ‘accidents’ were not morphologically marked. These ‘reductions’ occur much more frequently than the so-called ‘extensions’ (there are more ‘empty spots’ than new metalinguistic terms), due to the specific typological features of Latin and Greek. Apart from the obvious ‘extensions’ which almost always took place, another, and not less important linguistic discovery, was that so many “familiar” features were not present at all in these languages. For this reason, we prefer to see the application of the Greco-Latin model as a framework for these languages more as a case of “adaptations”, or “conversions” which could lead both to underdifferentiation and overdifferentiation at the same time.