Language Evolution and Systemic Typology

Gertraud FENK-OCZLON & August FENK, University of Klagenfurt, Austria

Starting point: Languages develop in accordance or in co-evolution with cognitive functions. At any point of time the constraints of our cognitive mechanisms are constraints on diachronic development, typological differentiation, and cross-linguistic variation. “Systemic Typology” takes into account that each language goes through self-organizing processes optimizing the interaction between its (phonological, morphological, syntactical) subsystems and the interaction with its ‘natural' environment, i.e. the cognitive and the articulatory system or the social-communicative environment [1].

In this paper, we focus on the ‘psychological present’ and a significant negative cross-linguistic correlation between the number of syllables per clause and the number of phonemes per syllable [2, 3].

Procedure: Native speakers of 51 languages from all continents (19 European, 32 Non-Indo-European) were asked to translate a matched set of 22 simple declarative sentences encoding one proposition in one intonation unit into their mother tongue. Furthermore, they were asked to count the number of syllables in normal speech. The number of phonemes was determined by the authors, assisted by the native speakers and by grammers of the respective languages.

Results: The 51 languages in our recent sample show a considerable variation in the mean number of syllables per clause, ranging from 4.64 in Thai up to 10.96 in Telugu. The mean number of phonemes per syllable is 2.24, ranging from 2.79 in German to 1.76 in Hawaiian. The results of a cross-linguistic correlation between the syllable complexity and the number of syllables per sentence: r = –0.73 (p < 0. 01). This correlation has proved to be very robust; from the first computation [2] in a sample of 26 predominantly Indo-European languages up to now 51 predominantly Non-Indo-European languages it has shown only little variation within a range from –0.77 to –0.73.

Discussion: The negative cross-linguistic correlation between number of syllables per clause and number of phonemes per syllable indicates time limits being effective on clause length – time limits regarding the psychological present (~2 sec) as well as the breath-cycle –, which force a trade-off between the length of syllables in number of phonemes and the length of clauses in number of syllables. Thus, the segmentation of natural languages and especially the limited size of clauses can be viewed as reflecting self-organizing processes accounting for such physiological and cognitive constraints.

Furthermore, recourse will be made to other significant correlations within the framework of Systemic Typology: Languages showing high syllable complexity tend to have a high number of words per clause, a low number of syllables per word and VO word order, whereas languages with simple syllable structure tend to have a low number of words per clause, a high number of syllables per word and OV word order. Higher syllable complexity is, moreover, associated with a larger phonemic inventory and with a higher proportion of monosyllabic words.

Reference:

[1] Fenk-Oczlon, G. & Fenk, A. (1999). Cognition, quantitative linguistics, and systemic typology. Linguistic Typology, 3, 151-177

[2] Fenk-Oczlon, G. & Fenk, A. (1985). The mean length of propositions is 7 plus minus 2 syllables – but the position of languages within this range is not accidental. In: d’Ydevalle. G. (ed.), Cognition, information processing, and motivation. North Holland: Elsevier Science Publishers B.V. 355-359

[3] Fenk-Oczlon, G. & Fenk, A. (2010). Measuring basic tempo across languages and some implications for speech rhythm. Proceedings of the 11^th Annual Conference of the International Speech Communication Association (INTERSPEECH 2010), Makuhari, Japan, 1537-1540