Recent searches led me to several interesting articles by Mark Pagel and co-authors on the evolution of languages. These links may be behind Science or Nature paywalls or subscriptions, so I apologize if you are unable to read them.
The basic point is that:
1.) Researchers are using phylogenetic trees to classify the relatedness of languages and their divergence from earlier languages. This is an old idea. The link above is to a review of recent work using the justifying using these tools due to the genetic-like properties of language.
2.) The proposal is that languages evolve in punctuated bursts, not gradually. Perhaps because populations becames seperated physically, Perhaps because groups wanted to differentiate themselves culturally.
Languages Evolve in Punctuational Bursts by Quentin D. Atkinson, Andrew Meade, Chris Venditti, Simon J. Greenhill, Mark Pagel; Science, 1 February 2008: Vol. 319. no. 5863, p. 588, DOI: 10.1126/science.1149683
Human language as a culturally transmitted replicator by Mark Pagel, Nature Reviews Genetics 10, 405-415 (June 2009) | doi:10.1038/nrg2560
