Abstract :
[en] In the Germanic languages, two general strategies are in use to form the past tense of a verb. The first and oldest, coined the strong inflection by Jacob Grimm (Grimm 1819), is based on a vowel alternation and assumed to be descendent from an Indo-European system of aspect (e.g. English sing-sang, Dutch zing-zong, Ringe 2006). The second and youngest, contrastively called the weak inflection, employs a dental suffix and constitutes a Germanic innovation (e.g. English kick-kicked, Dutch praat-praatte). In most Germanic languages, the weak inflection has become dominant and is still expanding, with the strong inflection being driven back to an ever declining number of strongholds, such as highly frequent verbs (e.g. zit-zat) and isolated ablaut patterns (e.g. the relatively successful Dutch first class of kijk-keek). The reason for this success is seemingly obvious, with the weak inflection enjoying a higher type frequency, and the strong inflection being deprived most of its regularity during centuries of destructive sounds laws (Lieberman et al. 2007; Carroll, Svare and Salmons 2012). Still, neither was the case when the fledgling weak inflection first entered the Germanic language (Ball 1968; Bailey 1997). In order to investigate how this weak newcomer could topple a dominant strong system, we have designed an agent-based simulation. In doing so, we will show how the weak inflection’s property of general applicability, i.e. its ability to in principle be applicable to all verbs irrespective of their stem vowel, could have sufficed to propel it from a position inferior in type and token frequency to the strong inflection to its current leading role. In short, the simulation shows how this property is capable of explaining (i) the gradual rise of the weak inflection, (ii) why low frequency verbs were the first to become weak, and (iii) why certain strong classes have been more successful is staving off the weak inflection than others.
