Centrum för kognitiv semiotik (CCS)

Humanistiska och teologiska fakulteterna | Lunds universitet


Are We Justified to Impose the Indo-European Interpretive Grid on Complex Sentences of "Exotic" Languages? -- No!

Esa Itkonen (University of Turku, Finland)

Within the Indo-European tradition, hypotaxis is divided into the COMP, REL, and ADV types, on the (loose) analogy of the three word-classes ‘noun’, ‘adjective’, and ‘adverb’, with the respective functions of (SUBJ or OBJ) argument, noun-modification, and verb-/clause-modification. The general validity of trichotomy, which continues to be adopted even by the best representatives of functional-typological linguistics, will be questioned in this talk.

Paratactic constructions cannot, by definition, have any formal distinctions corresponding to COMP, REL, and ADV. Thus, there cannot be any corresponding semantic meanings. At most, there can only be such pragmatic meanings. But why should they conform to this trichotomy?

  1. We need to distinguish between parataxis without hypotaxis (e.g. Mohawk) and parataxis besides hypotaxis (e.g. Mandarin Chinese).
  2. If a language has only one type of hypotactic (= subordinating) clause, it must, by definition, lack the three semantic meanings COMP, REL, and ADV: “It is important to note that the separation of different ‘types’ or ‘uses’ of these subordinated clauses [in Rembarrnga] should be seen as an artefact of the English translation” (McKay 1988: 8). The situation is somewhat similar in Ancient Tibetan, with action nominalizations performing the main subordinating functions.
  3. It may also be the case that even if a language (e.g. Diyari) has a tripartite subordinating structure, it does not correspond at all to COMP, REL, and ADV.
  4. A case opposite to Rembarrnga is represented by a language like Hua, with a subordinating structure much too complex to be squeezed into the straight-jacket of COMP, REL, and ADV.

The importance of this issue is evident from the fact that there is a great number of language with no (formal) equivalents to the logical connectives ‘or’ and ‘if - then’. To put it formulaically, these are languages with no ‘semantic logic’, but only with ‘pragmatic logic’.

The thesis of this paper was already outlined in Itkonen (2009). A wealth of relevant examples is provided in Itkonen (2005) and (2008-2010).


  • Austin, P. 1980. A grammar of Diyari, South Australia. Cambridge UP.
  • Beyer, S. 1992. The classical Tibetan language. Albany: State University of New York Press.
  • Haiman, J. 1980. Hua: A Papuan language of the Eastern Highlands of New Guinea. Amsterdam: Benjamins.
  • Itkonen, E. 2005. Ten non-European languages: An aid to the typologist. University of Turku: Publications in General Linguistics 9.
  • Itkonen, E. 2008-2010. Maailman kielten erilaisuus ja samuus (‘The diversity and the unity of the world’s languages’ in Finnish). 3rd ed., in three volumes. University of Turku: Publications in General Linguistics 12-14.
  • Itkonen, E. 2009. The true nature of typological linguistics. In J. Zlatev et al. (eds): Studies in language and cognition. Newcastle: Cambridge Scholars Publishing.
  • McKay, G.R. 1988. Figure and ground in Rembarrnga complex sentences. In P. Austin (ed): Complex sentence constructions in Australian languages. Amsterdam: Benjamins
  • Mithun, M. 2009. Re(e)volving complexity: adding intonation. Manuscript.