This paper presents a critical review of several recent studies that aim to model language contact, language shift and language death mathematically. We will present the models proposed by Abrams & Strogatz 2003, Mira & Paredes 2005,...
moreThis paper presents a critical review of several recent studies
that aim to model language contact, language shift
and language death mathematically. We will present the
models proposed by Abrams & Strogatz 2003, Mira & Paredes
2005, Mira, Seoane & Nieto 2011, Minett & Wang 2007
and Castelló (2010), as well as Castelló et al. (2012). These
papers refer to different contact situations, including Galician-
Spanish language contact. The different models and
corresponding algorithms will be described and critiqued,
distinguishing several degrees of sophistication and adequacy.
The study identifies a series of shortcomings, particularly
in Abrams & Strogatz’ and Mira & Paredes’ papers:
data problems, argumentative and logical errors, and also
inadequate simplifications. Nevertheless, it will be argued
that the more sophisticated models in particular should
continue to be developed, and a “heuristic spiral” between
model developers and sociolinguists pursued, in order to
achieve better models and greater insight into the mechanisms
involved in language change and language shift.
Este artigo consiste nunha reseña crítica dunha serie de
traballos recentes que procuran modelar, baseándose en
cálculos matemáticos, o contacto lingüístico e posibles
escenarios de desprazamento lingüístico (language shift),
é dicir, o paso do uso dunha lingua a outra, ou de morte
de linguas (language death). Presentaremos os modelos de
Abrams/Strogatz 2003, de Mira e Paredes 2005 e de Mira/
Seoane/Nieto 2011, de Minett/Wang 2007 e de Castelló
(2010) e Castelló et alii (2012), con referencia a diferentes
situacións de contacto (incluindo referencias ao contacto
galego-castelán). Os modelos e os algoritmos dos que parten
son presentados e criticados, diferenciando os graos
de sofisticación e adecuación dos diferentes traballos.
Identificaranse varios problemas sobre todo nos traballos
de Abrams e Strogatz e de Mira e Paredes: problemas de
datos, problemas de saltos argumentativos e problemas de
simplificacións. Aínda así, e sobre todo con referencia aos
modelos máis sofisticados, proponse continuar cunha “espiral
heurística” entre modeladores e sociolingüistas para
chegar a modelos máis adecuados e a unha comprensión
mellor dos mecanismos que actúan no contacto lingüístico.
