Challenges of sampling and how phylogenetic comparative methods help: With a case study of the Pama-Nyungan laminal contrast

Abstract

Phylogenetic comparative methods are new in our field and are shrouded, for most linguists, in at least a little mystery. Yet the path that led to their discovery in comparative biology is so similar to the methodological history of balanced sampling, that it is only an accident of history that they were not discovered by a typologist. Here we clarify the essential logic behind phylogenetic comparative methods and their fundamental relatedness to a deep intellectual tradition focussed on sampling. Then we introduce concepts, methods and tools which will enable typologists to use these methods in everyday typological research. The key commonality of phylogenetic comparative methods and balanced sampling is that they attempt to deal with statistical non-independence due to genealogy. Whereas sampling can never achieve independence and requires most comparative data to be discarded, phylogenetic comparative methods achieve independence while retaining and using all data. We discuss the essential notions of phylogenetic signal; uncertainty about trees; typological averages and proportions that are sensitive to genealogy; comparison across language families; and the effects of areality. Extensive supplementary materials illustrate computational tools for practical analysis and we illustrate the methods discussed with a typological case study of the laminal contrast in Pama-Nyungan.

Evolution and Trade-Off Dynamics of Functional Load

“…Study 2 One consequence of the high levels of phylogenetic signal found in FL, is that statistical analysis, such as the measurement of correlations should be carried out using phylogenetic comparative methods [46]. Phylogenetic Pearson's correlation (Table 4) was significant and negative between FL V and FL C but did not reach significance between FL V and FL P , in accordance with our hypotheses.…”

“…As languages are related to one another, it is not statistically valid to treat crosslinguistic observations as independent [46]. Quantitative phylogenetic methods [47] take genealogical relatedness into account in a principled and statistically sound manner.…”

“…Furthermore, it will be stronger when evolution was similar to Brownian motion, so that the value of the variable had equal probabilities of shifting up or down at any point, as opposed to (for example) being constrained within some tight range, so that at extreme outer values, there was a greater chance for the variable to evolve back towards the central value than further outwards. For a more technical description of phylogenetic signals written for a linguistic readership, see [42,46].…”

“…To place these K values in context, Macklin-Cordes et al [42] examined the lexical Markov chain transition probabilities of biphones (two-segment sequences) in PN and found mean K values of 0.54 or mean K of 0.59 when segments were binned into groups by place or manner of articulation. Macklin-Cordes and Round [46] examined the relative frequencies of dental versus palatal consonants in word initial and intervocalic positions in PN and found mean K values from 0.78 to 1.32 word-initially and from 0.34 to 0.70 intervocalically.…”

“…Phylogenetic signals have recently been shown to be present in phonotactic biphone frequencies, phoneme frequencies, and contextual ratios of places of articulation [42,46,[53][54][55]. These studies reveal that the frequencies of phonological structures pattern with genealogy.…”