I'm asking this very much much as a interested layman. As I understand things, the academic linguistics community, by and large, views macrofamily hypotheses - Nostratic, Altaic, etc - rather poorly.

Which of the following, then, best describes the current orthodoxy?

  • Any genetic relationships between the current primary language families are unknown, and most likely unknowable.
  • There are almost certainly no genetic relationships between the current primary language families, because groups of ancient humans probably developed languages independently from each other, and this is reflected in the massive gulf between, say, the Indo-European languages and the Sino-Tibetan languages.
  • Some third position which I'm not yet aware of.

A couple of clarifications to my original question:

  • I suppose I was really coming at this from an origin of language point of view. Do academic linguistics tend to think either (A) language itself was probably developed just once, so all languages must have a genetic relationship, even if it's not always possible to uncover that relationship today, or (B) language itself was probably developed more than once, so there will be languages which have no genetic relationship whatsoever, or (C) are they just as confused about all this as I am?
  • Sadly, I've not been put in charge of any committees deciding what is and what isn't considered orthodox amongst linguists. However, people who've published something about how the larger units of primary language families fit together seems like a good place to start.
  • 2
    I guess it was inevitable that sooner or later someone would utter the phrase "Is agnosticism the current orthodoxy?" in relation to something :) Mar 18, 2020 at 13:54
  • Ha ha. Yes I see what you're getting at now. In these days of plague and panic, can anyone afford to be without at least some form of orthodoxy, even if it is only with respect to the Nostratic hypothesis?
    – Tom Hosker
    Mar 18, 2020 at 13:58
  • 1
    First, I think it's not that the question doesn't make sense, but the answer. If the origin of language is anywhere between ten and thousand-thousand years old, give or take, then the question is well defined, but obviously unanswerable in the specific extent that any individual branch of linguistics would be interested it. Second, I'd one-up all that and say that the questions premisses of a linguistic community is not well defined, and the answer thus rather subjective PoV. Asking a general linguist about macro families (a nullary language family in your set up?) is like ...
    – vectory
    Mar 19, 2020 at 16:04

2 Answers 2


I think the prior question should be, who gets to vote? The difference between agnosticism and dogmatic nihilism, as I interpret the concepts, is that the agnostician simply says "I don't know", and the dogmatic nihilist would say things like "The question doesn't even make sense / the entire foundations of the enterprise are corrupt....".

I have spent most of my life on Bantu languages of the eastern / southern zone (similar in degree of relatedness to Romance), and I have a view of language relatedness – that view could be categorized in the direction of the "dogmatic nihilist" view. I don't dispute the idea that Shona and Logoori are related, but I do have many doubts about claims of specific relations. The reason for the doubt is specifically about methodology, that is, I distrust lexicostatistical methods as proof that A had a more-recent linguistic unity with B than either did with C. The underpinning of my nihilistic tendencies is language contact – languages can borrow; borrowing is not a trivial factor.

I do not work at the top end of these relatedness trees, so while the argument for Niger-Kordofanian doesn't strike me as obviously laughable, I also don't take it to be at all "established" in the way that Indo-European is. Actually, that's what I would also say about Niger-Congo. I do know that in the handful of specialists who do top-level historical linguistics, there is a general belief that much of Niger-Congo is valid, though not the internal details. I also know that earlier, there was a general belief that Niger-Kordofanian is valid, and there has been some back-peddling.

I don't have any informed theory of Altaic, because I don't work in that subdomain, despite dabblings in synchronic Turkic and Mongolian. I have seriously no opinion about "Penutian". However, I do have an opinion about the methodology that is often used in arriving at big language families, which leads me to be suspicious. (Vajda's argument for Yenisean-Dene is sort of the corrective, since he presents a linguistically credible argument that Ket is related to Athabaskan).

To be "fair", I probably should not vote on the Penutian question. Yet I know a lot of people who "vote" (have and express opinions on the topic) who don't have the appropriate factual knowledge. So I am meta-questioning the question – who gets to say what constitutes the current orthodoxy? I think the difference between agnosticism and dogmatic nihilism resides in what degree of familiarity the individual has with the evidence. If you are like me and have not read any of the literature on Penutian and don't know anybody who has, I would say "I don't know", since for me there is no evidence that the claim is true, but I also have no reason to say that no rational linguist could believe that theory. But if you have some small specific knowledge of that domain of research and have concluded that there is no good evidence for the theory, then a dogmatically nihilistic response is not totally inappropriate.

  • 1
    What a wonderful answer. Thank you. I'm going to add a couple of clarifications to my question, in response to what you've written.
    – Tom Hosker
    Mar 18, 2020 at 15:52

Instead of "agnosticism", I would rather say that the "orthodox" point of view is dogmatic nihilism...

  • 1
    Are you poking fun at the question (which is okay with me!) or am I failing to understand what you mean by dogmatic nihilism in this context?
    – Tom Hosker
    Mar 18, 2020 at 14:30
  • 1
    Dogmatic nihilism means that most people in academic milieus reject at all costs the idea that macro-families of greater extent exist. It's not sound scepticism. It's denial and nihilism.
    – user23769
    Mar 18, 2020 at 19:26

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.