Abstract Phoneme inventory size is related to speaker population size, but the cause of this relationship has remained a mystery. The correlation was first reported by Hay and Bauer (2007), who considered several possible causes of this relationship, including ascertainment bias, similarity between relatives, loss of phonemes in small, isolated populations, and maintenance of larger phoneme inventories in larger speaker populations. Here, we test these proposed causes of the relationship and show that the correlation between phoneme inventory size and speaker population size is largely driven by similarity between relatives and neighbours in both speaker population size and phoneme inventory size. However, we also show that phoneme inventory size is related to aspects of linguistic isolation—languages confined to islands or with few bordering languages tend to have lower phoneme inventory sizes than their relatives. We also explore the geographic bias of phoneme inventory databases and show that in some cases there is a positive bias towards regions with larger phoneme inventories. However, we find no evidence that less-studied languages tend to have lower reported phoneme inventory sizes. Our analyses demonstrate the importance of taking sources of non-independence and bias in typological data into account when conducting global analyses of linguistic diversity.
Abstract Phylogenetic reconstruction in historical linguistics now typically relies on cognates sets assembled from multilingual wordlists. While more and more scholars now trust in the robustness of the algorithms underlying the reconstruction of cognate-based language phylogenies, few studies have actually tested to which degree initial choices during data preparation can influence their outcome. Here, we provide first tests that focus on the role that concept translation—the initial stage of wordlist compilation, in which a list of concepts is translated into the target languages prior to identifying cognate sets—plays in phylogenetic reconstruction. Based on a newly compiled comparative dataset consisting of seven wordlists from five language families in which the same language varieties were coded by different authors, we investigate to which degree differences in concept translation lead to differences in phylogenetic reconstruction. Our results show that despite considerable differences in concept translation, lexical distances still show a considerably high correlation across all datasets. However, when comparing individual phylogenies reconstructed with the help of Bayesian inference, we find considerable differences ranging between 0.02 and 0.44 in normalized quartet distances computed from posterior tree samples. An additional cluster analysis that we introduce shows that larger differences in phylogenies do not necessarily correspond to high disagreement in the larger subgroups. A detailed inspection of individual concept translation differences in the Indo-European and Tupian wordlists in our sample further confirms that concept translation differences may specifically impact subgrouping decisions in lower clades of the tree, while major groupings are often unaffected.
Languages have been claimed to globally disfavor ergatively-aligned case marking, possibly reflecting processing preferences. Phylogenetic research on Pama-Nyungan has cast doubt on this, showing a preference for retaining ergativity in the family. In this study, we re-examine the evolution of ergative alignment of case marking in a sample of over 200 languages from eleven language families (Pama-Nyungan among them) in a continuous-time Markov Chain model. Since one language can adopt multiple alignment patterns of case marking in its grammar ('split' alignment), we incorporate a third, split state into the models. In addition, we compare the goodness of fit across various models where clades can show different rates of change, defining evolutionary regimes that can cut across families. The results show that, across all families, the best-fitting model requires two connected regimes: one favoring non-ergative (S=A) and the other favoring split-ergative alignment (S=A and S not equal A). These differences notwithstanding, the stationary probabilities of the model suggest that in the long run, non-ergative alignments are expected to be much more frequent than ergative alignment in case marking.
Abstract Human languages are widely understood to be conventional: for a given meaning, a population of language users expect a certain form to be used. Given the lack of unconventional natural languages with which to compare, there is little direct empirical evidence which explains why linguistic conventionality matters. We carry out a series of emergent language experiments with artificial agents designed to exhibit the effect of conventionality on robustness, learnability, and cognitive efficiency. By experimentally manipulating the interaction dynamics through which agents align their vocabularies, we create conditions that either promote or prevent population-level convergence, resulting in conventional and unconventional emergent languages while holding communicative success constant. We find that conventional languages are indeed more robust, better learnable, and more efficient, when compared with less conventional languages.
Human communication is highly adaptable in the sense that, for instance, new words or combinations of words emerge when there is a communicative pressure. However, another way in which humans can communicate about novel experiences is through meaning extensions of words that are already in their lexicon. One of the key driving factors of such polysemic extension is metaphorical conceptual mappings, which first become usualized, and then conventionalized within a community. In this article, we set out to investigate whether this kind of conceptual mapping can be achieved in an experimental semiotic study and whether the usualization trajectories of concrete and metaphorical meanings differ from each other. We do so by engaging participants in a referential game in which dyads of participants have to communicate about concrete and metaphorical meanings. We find that, over the course of the experiment, participants adapt elements of their kinematics from concrete to metaphorical meanings. We also find that the usualization trajectory is different between the two groups of meanings: both meanings initially undergo reduction in terms of movement path. However, concrete meanings continue to change and become more efficient across all rounds, while metaphorical meanings stabilize earlier and do not change significantly after their initial kinematic changes. In contrast to concrete meanings, we find no significant change in terms of speed and volume for metaphorical meanings. The study provides evidence that both concrete and metaphorical meanings can be expressed through pantomime, but their trajectory of change is different.
