Imperfect language learning reduces morphological overspecification: Experimental evidence

Abstract

It is often claimed that languages with more non-native speakers tend to become morphologically simpler, presumably because non-native speakers learn the language imperfectly. A growing number of studies support this claim, but there is a dearth of experiments that evaluate it and the suggested explanatory mechanisms. We performed a large-scale experiment which directly tested whether imperfect language learning simplifies linguistic structure and whether this effect is amplified by iterated learning. Members of 45 transmission chains, each consisting of 10 one-person generations, learned artificial mini-languages and transmitted them to the next generation. Manipulating the learning time showed that when transmission chains contained generations of imperfect learners, the decrease in morphological complexity was more pronounced than when the chains did not contain imperfect learners. The decrease was partial (complexity did not get fully eliminated) and gradual (caused by the accumulation of small simplifying changes). Simplification primarily affected double agent-marking, which is more redundant, arguably more difficult to learn and less salient than other features. The results were not affected by the number of the imperfect-learner generations in the transmission chains. Thus, we provide strong experimental evidence in support of the hypothesis that iterated imperfect learning leads to language simplification.

A bifurcation threshold for contact-induced language change

“…This mechanism-change triggered by L2-difficulty-is supported both by empirical studies in L2 acquisition and artificial language learning (e.g. Margaza & Bel 2006;Berdicevskis & Semenuks 2022) and by large-scale typological studies which suggest a negative correlation between the proportion of L2 speakers and language complexity (Lupyan & Dale 2010;Bentz & Winter 2013;Sinnemäki & Di Garbo 2018;Sinnemäki 2020).…”

“…The role played by second-language (L2) speakers in language change has attracted increasing interest in recent years, uniting research traditions in historical linguistics (Weerman 1993), sociolinguistic typology (Trudgill 2004;, language complexity (Kusters 2003;2008; McWhorter 2011) and second-language acquisition (Sorace & Serratrice 2009;Berdicevskis & Semenuks 2022). If L2 learners generally struggle to acquire certain linguistic features in a target-like fashion, and if L2 learners are sufficiently prevalent in a speech community in a particular historical situation, then features of this kind, namely L2-difficult features, may be expected to be lost from the language over extended time.…”

Cultural evolution of music and language

“…The language side has already discovered population size can matter the evolution of key features of language (e.g. Berdicevskis & Semenuks, 2022;Raviv et al, 2019). Therefore, it is not unreasonable to expect that how people learn and transform music also varies under different population conditions.…”

“…Motamedi et al, 2019;Schouwstra, Smith, & Kirby, 2020;Fay et al, 2022); but also visual color sequences (Cornish, Smith, & Kirby, 2013), drawings (Fay et al, 2010Garrod et al, 2010;Theisen-White, Kirby, & Oberlander, 2011), and continuous sounds created using slide whistle (Verhoef, Kirby, & De Boer, 2015) or leap motion (Eryilmaz & Little, 2017) have been studied as signalling devices. Moreover, several studies have explored effects of population structure on language structure and learnability, such as group size (Raviv, Meyer, & Lev-Ari, 2019;Raviv, de Heer Kloots, & Meyer, 2021), network structure (Raviv, Meyer, & Lev-Ari, 2020), and proportion of imperfect or non-native learners (Berdicevskis & Semenuks, 2022), as well as novel communicative environments through virtual reality (Nölle, 2021). Overall, methodologies in the experimental language evolution literature provide promising means to refine theories on how individual level cognitive biases, as well as types of transmission and population structure, shape the structure and learnability of resulting languages.…”

Sociolinguistic Typology Meets Historical Corpus Linguistics

“…For instance, Berdicevskis & Semenuks (2022) demonstrated, via an iterated learning experiment modelling generational transmission, that morphological simplification and thus change correlates with the number of ‘short‐time’ learners (i.e. learners with a reduced amount of linguistic input) involved in a transmission chain, with redundant agent marking on verbs being the most vulnerable feature.…”