Retour vers liste

Détail de la contribution

Auteur: Andrea RAVIGNANI

Co-Auteur(s): Mauricio MARTINS, Language Research Laboratory, Lisbon Faculty of Medicine, Portugal & Department of Cognitive Biology, University of Vienna, Austria Archishman RAJU, St. Stephen's College, Delhi University, India W. Tecumseh FITCH, Department of Cognitive Biology, University of Vienna, Austria

The Role of Quantitative Modeling in Language Evolution

Abstract/Résumé: Mathematical models and computer simulations are a flourishing and indispensable area of research in language evolution. Here we highlight critical issues in using and interpreting models and suggest viable approaches. First, contrasting models can explain the same data and similar modeling techniques can lead to diverging conclusions. The most striking contrasts emerge when similar models or data are used by some to support the biological evolution of language and by some others to defend the cultural evolution of linguistic features. This evident flexibility in interpretation indicates that results should be interpreted cautiously, especially in areas of fervent debate. Second, quantitative techniques similar to those used in modeling language evolution have sometimes proven themselves inadequate in other disciplines. Economics, for instance, has gone through a phase of extreme mathematization. Many models have been put to test and many have been discarded due to failure. Cross-disciplinary collaboration is crucial to avoid inferential mistakes in modeling language, which have previously occurred in other areas. Finally, experimental validation is necessary both to sharpen models' hypotheses, and to support their conclusions. This is not trivial to accomplish, and requires a thorough selection and scrutiny of tacit experimental assumptions. Very often, for instance, only data concerning features of E-languages, rather than I-language, is available. However, data relevant for the study of the cultural/historical evolution of languages cannot be assumed to be relevant to understand the biological evolution of the capacity for language. Our belief is that models should be interpreted as quantitative demonstrations of logical possibilities (“proofs of concept”), rather than as direct sources of data and evidence. Only an integration of theoretical principles, rigorous quantitative techniques and empirical validation will allow research in the evolution of language to make satisfactory progress.