New methodologies for historical linguistics?

Abstract

Recent research claims that analysis of lexical cognate classes for a basic wordlist can reproduce linguistic subgroups within the Austronesian family (Gray et al. 2009). The analysis is open to question in two respects. Primarily, the lexicallybased classification, primed with pre-established cognate classes of the family it seeks to emulate, fails to differentiate shared retentions from shared innovations. Secondly, languages and language families typically disperse through contiguous regions (especially in the Pacific) which means that geography or social distance should be expected crudely to match phylogeny in most cases. The reproduction fails because of local borrowing between branches not closely related to each other. For instance, when we examine disjunct distributions, cases in which the phylogeny does not match a straightforward geographic spread, we can determine which of these (phylogeny or geography) the lexical cognate approach preferentially detects. Where we find a mismatch between geography and phylogeny, Gray et al. 's approach clusters languages based on human geography (that is, social distance), not linguistic subgroup. In all cases of divergence between Gray et al. 's tree and accepted Austronesian trees, the discrepancy is a product of the former representing social distance rather than historical phylogenetic relationships. In summary, the examination of lexical cognate classes is not a valid proxy for the comparative method, though it is a useful heuristic for detecting pairs of languages that are either lexically conservative, or which show the effects of later lexical diffusion (without discriminating between these two outcomes).

Using hybridization networks to retrace the evolution of Indo-European languages

“…Trees with less than four leaves have identical topologies and thus cannot be used to recognize word borrowings [29]. It is important to note that the MLN approach is an automatic approach based on tree topology and the 42 suggestive cases of borrowing recovered by MLN, which include 33 English loanwords identified by Donohue et al [22], cannot be considered as crystal-clear borrowings. They may comprise some false positives, which can be due to parallel semantic development [21], for example.…”

“…The loanword information was taken from the studies of Donohue et al ([22]; Supplementary Material) and List et al ([21]; modified MLN approach). To the 33 English loanwords identified by Donohue et al [22] and 8 possible additional English borrowings suggested by List et al ([21]; Supplementary Material II; Table C), we added the word dust , which might have been borrowed by English from Old North according to the results given by our algorithm for detecting word borrowing events [29]. This English word was also identified as belonging to an irregular pattern, which may be suggestive of borrowing, by the modified MLN approach [21].…”

“…Cases of confirmed borrowing, according to Donohue et al [22] are not colored; cases of possible borrowing, which could have been unrecognized according to List et al [21], are shaded in gray (note that List and colleagues point out that the grayed words can also represent cases of parallel semantic development [21]); an additional suggestive case of borrowing (i.e., dust) is in yellow. When the Old North (ON) group was assumed to be the word donor, the exact source of borrowing in the ON group was not considered since it remains unknown in most of these cases.…”

“…While the use of this list may lead to a certain decrease in the number of loanwords [20], it remains helpful for detecting the most important word borrowing trends [21]. For example, the traditional 200-meaning English Swadesh list includes 33 confirmed loanwords (16.5 %) [22] and 10 additional “irregular phylogenetic patterns” which might be suggestive of unrecognized borrowings [21]. Moreover, in a recent revision of the Albanian Swadesh list 31.8 % of its entries were identified as probable borrowings [23].…”

Typology and Areality

“…When we have no clear subgrouping available from the comparative method, as is the case for Albanian, an isolate within IndoEuropean, the language clusters with its geographic neighbors, and not as a distant branch on its own. While there is a clear genealogical stability associated with some typological features (as described and quantified in Wichmann and Holman, 2009), this may be inflated due to the high correlation that linguistic subgroups have with geography as well (as argued in Donohue et al, 2011, andDonohue et al, 2012). The reader can assess the geographic integrity of the clusters shown in Figure 2 by examining their mapping in Map 2, which shows the large-scale divisions of the languages.…”