As far as I can tell, the comparative method is the most accepted and widely used method in historical linguistics. Often used in conjunction with the internal reconstruction method, it has proved to be very successful ever since the 19th century. There are, of course, alternatives, such as the mass comparison method, defended by Merritt Ruehlen and others. But such methods are, to say the least, very controversial.

So, is there any reasonable alternative in historical linguistics? By reasonable alternative, I mean a competing method capable of achieving the same results by different means. Or perhaps, capable of going further into the past to reconstruct larger language families?

  • I don't think I have ever heard of anything like that, and I suppose I would have heard of it if it were actively competing. But this is also what your question suggests. I wonder, maybe if you rephrased it to suggest that there is such a method, only you don't know what it is, it would make it easier to remember some long-forgotten marginal information? If I marked the question favourite, I'll get a notification when it's answered, right?
    – kamil-s
    Commented Mar 14, 2012 at 7:48
  • I guess en.wikipedia.org/wiki/Mass_comparison is a point to start. But I foresee heated discussions...
    – JPP
    Commented Mar 14, 2012 at 12:03
  • @JPP that's why I mentioned it in the question as a controversial method ;-) Commented Mar 14, 2012 at 12:50
  • 2
    Unfortunately, languages don't change like isotopes. Every C₁₄ atom is exactly like every other C₁₄ atom, and has the same probability of decay; this is far from the case with chunks of language. The fact that the comparative method works at all is evidence that some chunks last longer than others and can be tracked, like tree-rings, according to known phonological and grammatical processes, to unravel some of the changes and signal degeneration, to a time depth of perhaps 8 kiloyears. That's it, afaik.
    – jlawler
    Commented Mar 14, 2012 at 18:49

1 Answer 1


Given the absence of anything that looks like an alternative to the comparative method (and internal reconstruction) in the historical linguistics literature, I think we can safely say that there is no alternative.

The method of mass (or multilateral) comparison is heavily critiqued, but whether one would subscribe to its use or not, it would not pose an alternative to the comparative method. If we take the comparative method as a way of i) establishing genetic relatedness between languages, ii) identifying systematic sound correspondences between them, in order to iii) ultimately arrive at a reconstructed proto-language, then multilateral comparison concerns itself only with the first step.

For example, Greenberg (1996: 131–132):

It seems to be widely supposed that multilateral comparison and the comparative-historical method are in some way contradictory. This is perhaps most starkly stated by Nichols (1990:477, n. 1) as follows: "Greenberg (1987) makes it clear that he believes such grouping cannot be reached by the standard comparative method; a wholly different method, mass comparison, is required." Where I supposedly said this is not specified. In fact, just the opposite is stated: "... my remarks are not intended as an attack on the validity of comparative linguistics or on the importance of undertaking reconstruction. Rather, the discussion is meant constructively as a way of taking first steps where the comparative method has not been applied for want of an assured basis in genetic classification" (LIA, p.3).

I said much the same thing in regard to the relationship between the two much earlier (1957:44), namely, that the methods outlined there did not conflict in any fashion with the traditional comparative method and that they might be viewed as the first step in the method itself, for we cannot begin systematic reconstruction until we know which languages to compare. Whether classification by multilateral comparison is to be viewed as a necessary preliminary step in order to define valid genetic units without which comparative linguistics cannot proceed or as the first step in the method itself - an alternative which seems more natural - is a matter of definition.

The heart of the dispute then seems to be whether a classification of genetically related languages can be proven by way of multilateral comparison, or whether the relatedness can only be established by identifying systematic sound correspondences and/or a reconstructed proto-langauge. Either way, mass comparison cannot be counted as a full alternative to the comparative method, since it does not address the reconstruction of proto-languages.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.