Using a large database and many alternative methods Dediu and Levinson show that both positions are right: there are universal tendencies for some features to be more stable than others, but individual language families have their own distinctive profile. These distinctive profiles can then be used to probe ancient relations between what are today independent language families.
"Using this technique we find for instance probable connections between the languages of the Americas and those of NE Eurasia, presumably dating back to the peopling of the Americas 12,000 years or more ago," Levinson explains. "We also find likely connections between most of the Eurasian language families, presumably pre-dating the split off of Indo-European around 9000 years ago."
From the paper:
Quite convincing is the evidence that Core Eurasian families (comprising Altaic – or Mongolic + Turkic –, Dravidian, Indo-European, Uralic and the Caucasian families) might form a group (p=0.0013, 5 methods, and , p=0.094, 4 methods, when controlling for geography).The authors were also able to reject the "broad" Afroasiatic group "comprising Afro-Asiatic, Indo-European, Dravidian and Uralic". I think this makes some sense, since Afroasiatic is basically an African language family with a Near Eastern offshoot, so I did not expect it to group with the Eurasian language families.
The Core Eurasian group seems very interesting in light of accumulating evidence about contacts between human groups across Eurasia. Such a group is pushing the limits of what can be inferred using linguistic data, and, perhaps, archaeogenetics might provide some evidence that might be used to plausibly argue for such a relatively broad group.
PLoS ONE 7(9): e45198. doi:10.1371/journal.pone.0045198
Abstract Profiles of Structural Stability Point to Universal Tendencies, Family-Specific Factors, and Ancient Connections between Languages
Dan Dediu, Stephen C. Levinson
Language is the best example of a cultural evolutionary system, able to retain a phylogenetic signal over many thousands of years. The temporal stability (conservatism) of basic vocabulary is relatively well understood, but the stability of the structural properties of language (phonology, morphology, syntax) is still unclear. Here we report an extensive Bayesian phylogenetic investigation of the structural stability of numerous features across many language families and we introduce a novel method for analyzing the relationships between the “stability profiles” of language families. We found that there is a strong universal component across language families, suggesting the existence of universal linguistic, cognitive and genetic constraints. Against this background, however, each language family has a distinct stability profile, and these profiles cluster by geographic area and likely deep genealogical relationships. These stability profiles seem to show, for example, the ancient historical relationships between the Siberian and American language families, presumed to be separated by at least 12,000 years, and possible connections between the Eurasian families. We also found preliminary support for the punctuated evolution of structural features of language across families, types of features and geographic areas. Thus, such higher-level properties of language seen as an evolutionary system might allow the investigation of ancient connections between languages and shed light on the peopling of the world.