Editing Indo-European languages (section)

=== Tree versus wave model ===
{{See also|Language change}}
The "[[tree model]]" is considered an appropriate representation of the genealogical history of a language family if communities do not remain in contact after their languages have started to diverge. In this case, subgroups defined by shared innovations form a nested pattern. The tree model is not appropriate in cases where languages remain in contact as they diversify; in such cases subgroups may overlap, and the "[[wave model]]" is a more accurate representation.<ref>{{Citation |last=François |first=Alexandre |contribution=Trees, Waves and Linkages: Models of Language Diversification |editor1-last=Bowern |editor1-first=Claire |editor2-last=Evans |editor2-first=Bethwyn |title=The Routledge Handbook of Historical Linguistics |pages=161–189 |publisher=[[Routledge]] |place=London |year=2014 |isbn=978-0-415-52789-7 |contribution-url=http://alex.francois.free.fr/data/AlexFrancois_2014_HHL_Trees-waves-linkages_Diversification.pdf |ref=francois}}</ref> Most approaches to Indo-European subgrouping to date have assumed that the tree model is by-and-large valid for Indo-European;<ref>{{cite journal |title=From August Schleicher to Sergei Starostin: on the development of the tree-diagram models of the Indo-European languages |last=Blažek |first=Václav |journal=[[Journal of Indo-European Studies]] |year=2007 |volume=35 |issue=1–2 |pages=82–109}}</ref> however, there is also a long tradition of wave-model approaches.<ref>{{cite book |title=Les dialectes indo-européens |language=fr |trans-title=The Indo-European dialects |publisher=Honoré Champion |last=Meillet |first=Antoine |year=1908 |location=Paris}}</ref><ref>{{cite book |title=I dialetti indoeuropei |publisher=Paideia |last=Bonfante |first=Giuliano |year=1931 |location=Brescia}}</ref>{{sfn|Porzig|1954}}

In addition to genealogical changes, many of the early changes in Indo-European languages can be attributed to [[language contact]]. It has been asserted, for example, that many of the more striking features shared by Italic languages (Latin, Oscan, Umbrian, etc.) might well be [[areal features]]. More certainly, very similar-looking alterations in the systems of [[long vowel]]s in the West Germanic languages greatly postdate any possible notion of a [[proto-language]] innovation (and cannot readily be regarded as "areal", either, because English and continental West Germanic were not a linguistic area). In a similar vein, there are many similar innovations in Germanic and Balto-Slavic that are far more likely areal features than traceable to a common proto-language, such as the uniform development of a [[high vowel]] (*''u'' in the case of Germanic, *''i/u'' in the case of Baltic and Slavic) before the PIE syllabic resonants *''ṛ, *ḷ, *ṃ, *ṇ'', unique to these two groups among IE languages, which is in agreement with the wave model. The [[Balkan sprachbund]] even features areal convergence among members of very different branches.

An extension to the ''[[Donald Ringe|Ringe]]-[[Tandy Warnow|Warnow]] model of language evolution'' suggests that early IE had featured limited contact between distinct lineages, with only the Germanic subfamily exhibiting a less treelike behaviour as it acquired some characteristics from neighbours early in its evolution. The internal diversification of especially West Germanic is cited to have been radically non-treelike.<ref>{{Cite journal |last1=Nakhleh |first1=Luay |last2=Ringe |first2=Don |last3=Warnow |first3=Tandy |author3-link=Tandy Warnow |title=Perfect Phylogenetic Networks: A New Methodology for Reconstructing the Evolutionary History of Natural Languages |name-list-style=amp |date=2005 |journal=[[Language (journal)|Language]] |volume=81 |issue=2 |pages=382–420 |doi=10.1353/lan.2005.0078 |citeseerx=10.1.1.65.1791 |s2cid=162958 |url=http://www.cs.rice.edu/~nakhleh/Papers/NRWlanguage.pdf}}</ref>