Seminar: Frequency and stability in language systems

The frequency and stability of linguistic variants seems to be tied: the more frequent a given variant is, the less it tends to change over time. For example, frequent verbs are more likely to maintain an irregular past tense form over time. Despite the evident relationship between frequency and stability, it is still unclear what specific factors underlie this relationship. In other words, why, exactly, do frequent linguistic variants tend to be more stable than their less frequent counterparts? This talk will examine several factors which may affect verb regularization, particularly as observed in English. First, English has seen considerable growth in the population size of it its speakers characterized by a large increase in the number of non-native speakers in particular. I will present experimental evidence showing that native and non-native speakers of English exhibit different patterns of regularization in a wug-style task (where participants generate a novel past tense form for a non-word, e.g., I wug every day, what did I do yesterday?). These results provide new empirical evidence that the regularization behavior of non-native speakers differs markedly from that of native speakers, lending weight to the notion that regularity in language may be influenced by non-native speakers in specific ways. Second, the English language itself has grown considerably over the last 200 years or so. The Corpus of Historical American English (CoHA) covers the period between 1830 and 1990, shedding light on both the increase in population as well as an increase in the size of the language itself (i.e., vocabulary size). CoHA shows that even over a relatively short period of time, we can observe not only processes of regularization, but also irregularization. However, by and large, the system is stable, and the growth of the vocabulary affects regularity considerably more than internal dynamics of (ir)regularisation.