Uto-Aztecan, “A website for Uto-Aztecan Studies,” is a welcome new addition to the internet, and just the sort of thing that the internet is ideal for. Brian D. Stubbs, who created the site, writes:

Welcome to Uto-Aztecan.org, a website devoted to the comparative study of the Uto-Aztecan (UA) language family. Located in the southwestern United States and western Mexico, UA consists of some 30 related Native American languages descended from a common parent language that linguists now call Proto-Uto-Aztecan (PUA). Hopi, Ute, Pima, and Aztec/Nahuatl are among the better known of UA languages. The valuable works of many linguists are listed in the bibliography and some are discussed in the introduction accessible above and will be cited here increasingly over time, but the initial offerings are portions of the book, A Uto-Aztecan Comparative Vocabulary (by brian stubbs), available on this website, intended to encourage and facilitate the comparative study of Uto-Aztecan languages.
The book presently contains some 2650 Uto-Aztecan cognate sets (groups of related words), which is only the latest plateau of progress or new foundation for future research. After three decades of studying UA and compiling the book, I realized that an undertaking as large as a language family has no end, like running a race without a finish line. Each new discovery creates rows of ripples of adjustments to so much else, and there is no end to new discoveries[…]

You can read more at the Introduction page, and of course sample the various offerings of the site. (Thanks, Yoram!)


  1. marie-lucie says

    This is just what I have been waiting for !!!!

  2. The naming of languages is rather inconsistent and also riddled with The Whig Interpretation of History. Wouldn’t “Old English” be better named as, say, Insular West Germanic?

  3. marie-lucie says

    dearieme, I am not sure what you mean by your first sentence, especially “the Whig interpretation of history”.
    “Germanic” is a family, not a language, and “Insular West Germanic” would be the name of the subfamily which includes English and Scots. “Old English”, formerly called “Anglo-Saxon”, refers to a specific, attested language, spoken at a defined period, which is a known stage in the evolution of English.
    “Uto-Aztecan” is a family which includes the Ute, Aztec, Hopi, etc languages. Proto-Uto-Aztecan (like other Proto-X’s) refers to the putative, reconstructed ancestor of the whole family.

  4. Calling something “Old English” inplies that it has a duty to become in turn Middle English, Young English, Infantile English, etc. It must witness the Renaissance, Reformation, Glorious Revolution and the Great Reform Act.
    Moreover, that name privileges the so-called Anglo-Saxons: Old Danish and Old Norman French also contributed to the essence of Infantile English. There’s no need to rate our Kraut invaders any higher than our Scandowegian or Frog invaders.
    Anyway, it’s ambiguous: “Old English” in English can also mean “Former English” which is plain silly. Bah, I don’t like it; sweep it away!

  5. dearieme, you are entitled to your opinion, but perhaps your definition of “Old English” is wider than that of historians of the English language. There are names for distinctive periods in the history of English, and for distinct dialects before eventual standardization (at least of the written form of the language). Old Danish and Old Norman French contributed to the development of the language beyond the period of “Old English”, once speakers of those languages settled on English soil and eventually mixed with English speakers. If you think “Old English” favours those of Anglo-Saxon origin, then so does plain “English”, named for the “Angles”.

  6. The Whig interpretation of history, an expression invented by Herbert Butterfield in 1963 to delineate a particular school of historical interpetation. Butterfield used the expression to describe those he accused of interpreting the past solely through the lens of the present. “Old English” can be criticised as a name, if I may be so presumptuous as to argue what I believe is dearieme’s case, because it is derived “backwards” from the name “Modern English”: “Insular West Germanic” (or perhaps “Southern Insular West Germanic”), it might be argued, is a better name because it describes where the language at that time had come from and was situated, rather than where it was going to.

  7. “Old English” in English can also mean “Former English” which is plain silly. Bah, I don’t like it; sweep it away!
    That is its name, whether you like it or not. You might as well (as m-l says) try to change the name of the English language itself.

  8. marie-lucie says

    I think I see the point, but “Insular West Germanic” would be quite appropriate for the modern group including English and Scots as well. These languages have not stopped belonging to the Western branch of the Germanic family, even with the additional contributions from Scandinavian (North Germanic branch) and French (Western Romance).
    “Anglo-Saxon” is not used so much nowadays, because it seems to imply that the language before the Norman Conquest was radically different, and that between it and later English there was a language shift, perhaps similar to that from Gaulish to French. The name “Old English” emphasizes the continuity: after all, King Alfred already spoke “englisc”. It is always possible to talk about, for instance, “9th century English” or *14th century English”, but such labels are cumbersome when a distinct period extends over more than one century.

  9. Shrewd chap, that Zythophile.

  10. Yes, and he knows his beer.

  11. You might as well … try to change the name of the English language itself.
    Mencken tried to rename it (or at least the “American language”) “Anglo-American”. As an Australian, of course, I take great umbrage at being left out of the club 🙂

  12. Whig view of history – 1931, I think, not 1963.

  13. Whig view of history – 1931, I think, not 1963.
    Picky, picky! And, of course, correct.
    Wikipedia says of Butterfield:
    “Interestingly, after The Whig Interpretation of History he continued to write history with a Whiggish style. He stated that in fact it was too hard not to portray any historiography Whiggishly.”

  14. How about a compromise? We’ll call it Olde Englysshe.

  15. marie-lucie says

    That would be naming it in Middle English.

  16. I thought you weren’t allowed to say “Interestingly” on Wikipedia. Insert your own joke here.

  17. Me, I want to change “Whiggishly” to “waggishly” in that passage.

  18. marie-lucie says

    What would be the opposite of “Whig” or “Whiggishly” in this context?

  19. Wikipedia has this wonderful quote from Macaulay:
    “I shall relate how the new settlement was, during many troubled years, successfully defended against foreign and domestic enemies; how, under that settlement, the authority of law and the security of property were found to be compatible with a liberty of discussion and of individual action never before known; how, from the auspicious union of order and freedom, sprang a prosperity of which the annals of human affairs had furnished no example; how our country, from a state of ignominious vassalage, rapidly rose to the place of umpire among European powers; how her opulence and her martial glory grew together; how, by wise and resolute good faith, was gradually established a public credit fruitful of marvels which to the statesmen of any former age would have seemed incredible; how a gigantic commerce gave birth to a maritime power, compared with which every other maritime power, ancient or modern, sinks into insignificance; how Scotland, after ages of enmity, was at length united to England, not merely by legal bonds, but by indissoluble ties of interest and affection; how, in America, the British colonies rapidly became far mightier and wealthier than the realms which Cortes and Pizarro had added to the dominions of Charles the Fifth; how in Asia, British adventurers founded an empire not less splendid and more durable than that of Alexander.
    … (T)he history of our country during the last hundred and sixty years is eminently the history of physical, of moral, and of intellectual improvement.”
    I’m so old I still believe all this 1066 and All That stuff. The opposite? Jacobitism? Bonapartism? Pessimism? Common sense? Balance?

  20. Wow, how the mighty have fallen since.
    So, does “Whig”, etc refer to a teleological vision of history? constant improvement towards an ideal end? but in the early stages of the historical study of languages, the prevalent opinion was that modern languages showed decay and degradation from the ideal represented by Latin, Greek and Sanskrit. The idea that change or evolution in language does not mean either improvement or degradation is a modern one. “Old”, “Middle”, etc refer to historical stages defined by certain complexes of linguistic features, not to necessary episodes in the pursuit of some ultimate goal.

  21. The opposite? Jacobitism? Bonapartism? Pessimism? Common sense? Balance?
    Macaulay’s opposites may be contradictory, but isn’t the contrary of Whig and Protestant simply Tory and Catholic? In this context? Who knows? I can’t see any relevant sense in which the term ‘Old English’ exhibits Whiggishness. I think dearieme has simply mistaken a claim to historical certainty for a claim to historical necessity.

  22. Fallen a bit, haven’t they, m-l? He could write, though, couldn’t he?
    I think Victorian Britons would have held Latin and Greek as examples of perfection from which we had fallen, but happily that was all before the real start of history, which was the start of British history, which was the Anglo-Saxon conquest.
    Whigs traced the success of Britain to the Glorious Rev, the Act of Settlement, and the Bill of Rights.
    Everyone, Whig or Tory, traced it back further in a succession of improvements to the savage, noble, but somehow respectable Anglo-Saxons stepping decorously from their ships at Ebbesfleet. I think they would have had difficulty with the conception that modern English was anything other than an evolutionary peak rising from Anglo-Saxon and Middle English foothills.

  23. m-l: The opposite of “Whig” in this context, at least in the U.S., is “fair and balanced.” 🙂
    He Of The Chinesian Cityworld of Aojou Nanbien: Sidney J. Baker wrote a book in homage to Mencken called The Australian Language; alas, I have not read it.

  24. John Cowan says

    I have now, and (up to 1945) it delivers what it says on the tin, even though Baker’s style is nothing at all compared to Mencken’s.

  25. Apparently nobody but marie-lucie cared about the Uto-Aztecan site itself (now defunct; I’ve substituted an archived link), but I just ran across Magnus Hansen’s An evaluation of the Nahuatl data in Brian Stubbs’ work on Afro-Asiatic/Uto-Aztecan, a fairly devastating takedown of Stubbs’s attempts to… well, I’ll let him tell it:

    In this blogpost, I analyze the use of Nahuatl data, in Brian D. Stubbs self-published manuscript “Exploring the Explanatory Power of Semitic and Egyptian in Uto-Aztecan”. This work compares Proto-Uto-Aztecan with Semitic and Egyptian and seeks to find signs of ancient contact between early near-eastern and Egyptian peoples and Uto-Aztecan speaking Native Americans. It finds a lot of such signs, in fact more than 1500 Uto-Aztecan forms that Stubbs claims are cognate with Semitic or Egyptian forms.

    You can read the whole thing for the details, but that in itself is enough to make me shake my head sadly. His work in Uto-Aztecan historical linguistics itself, however, seems to be well regarded.

  26. Nahuatl Scholar posts so irregularly that it falls off my radar, and then I read for a day when I rediscover it.

    The final paragraph:

    One thing that makes me uncomfortable is the fact that Stubbs worked on this simultaneously with working on his catalogue of Uto-Aztecan reconstructions. I cannot help but worry that the reconstructions of UA forms there, may be subtly compromised by Stubbs unconsciously trying to make them fit with his Semitic data. I will have to use it with a degree of apprehension in the future.

    Brian Stubbs used his name and reputation as a scholar of Uto-Aztecan to write a piece of pretend-scholarship meant to bolster up the faith of the LDS. How could that not reflect badly on his other works?

    Also, there”s a joke on Egyptian here, but it’s probably old and tired to Pharao himself.

  27. David Eddyshaw says

    From that excellent post:

    Published in the “Journal of Book of Mormon Studies” in 2019, the review by Chris Rogers, an expert in historical linguistics and the Xinka language of El Salvador and also a professor of linguistics at Brigham Young University, is much more critical.

    Rogers points out some serious flaws in the work: Primary of these is that it does not stick to the established bilateral method of comparing languages with languages and proto-languages with proto-languages, but that it frequently cherrypicks so that a form in any Semitic or Egyptian variety can be compared with a form in any Uto-Aztecan language. As has been demonstrated time and time again this multilateral method hugely increases the risk of mistaking chance resemblances for cognates, and makes it possible to prove virtually anything.

    Preach it, brother!

    To be fair, Stubbs wouldn’t be the only scholar who combines rigour within his actual area of expertise with utter silliness outside it, especially when it comes to long-range proposals; most don’t have the extenuating circumstance (if you call it that) that they are unduly influenced by their religious beliefs: they are just wrong about exactly where their expertise ends.

    I get the impression (due to knowing nothing about it) that comparative Uto-Aztecan is (relatively) simple, as far as such things go, inasamuch as there is a lot of uncontroversially cognate material to work with, along with excellent documentation by some of the best field linguists ever; Stubbs may have been lulled into a sense of false security about the many pitfalls of comparative work.

    “Bulverism” is a useful concept. Though I have an uncomfortable feelling that I have more often Bulvered than been Bulvered against. It’s always tempting to impugn someone’s motives rather than prove them wrong. Easier, and more fun, too … must watch that.

  28. Stubbs wouldn’t be the only scholar who combines rigour within his actual area of expertise with utter silliness outside it, …

    Yeah, Isaac Newton believing in the occult; Conan Doyle and the paranormal; …

    Given the vast amounts of money LDS has at its disposal, and the difficulty of getting academic tenure in recherché subjects, I might allow Stubbs would pay lip-service to LDS as a means to pursuing his ‘actual area of expertise’.

    But doesn’t this utter silliness (and indeed it seems from Rogers’ piece that it spreads to other scholars of U-A languages) invalidate the whole area of study? Do I believe any claim whatsoever about U-A languages? How do I know it isn’t all just made up out of whole cloth, in some sort of recondite scheme to validate Joseph Smith?

    Lizard people; insemination by aliens; child porn rings run from pizza parlours; vaccines containing nanochips … Is there any silliness some people won’t believe! Yes I’m guilty of Bulverism — indeed a useful concept. But life is too short to go round debunking every craziness. IOW Rogers seems to have done a more thorough job than would be needed to dismiss such preposterous claims. I’ll take his word for it.

  29. As far as I know Stubbs has been doing his comparative UA work voluntarily and is retired anyway. This is not “lip service” to the LDS. On the other hand, I see no reason to think that any of his LDS-inspired comparisons has influenced the comparative UA dictionary. The etymologies, many of which are not certain (as with any such work) are reasoned in detail, and alternatives and problems are mentioned. It would be hard to hide any such bias in these circumstances.

  30. from Rogers’ piece …Rogers seems to have done …

    Errk. That should be “Hansen’s piece”. (Rogers is one of the LDS-aligned reviewers.)

  31. @Y, from the follow-up-discussion

    Now I [Hansen] worry that this is indeed a sign that he [Stubbs] has been keeping his potential semitic cognates in mind when he organized the vocabulary, otherwise I really don’t see why the two roots *napa “drunk/alcohol” and *na’pul “prickly pear” should be considered related.

    This is not “lip service” to the LDS.

    Indeed it can’t be. That suggestion from me was a counter-factual: if his motivation were merely lip-service, he wouldn’t go to the bother of publishing a manuscript [allegedly] tracing thousands of forms.

    Stubbs must actually want to believe in Joseph Smith’s nonsense — to the extent Stubbs has abandoned his critical faculties to produce this bogus ‘validation’.

    Isaac Newton’s work on mechanics had to match the motion of anything from cannonballs to planets to the tides — or he’d very quickly have been ridiculed. Then Newton must somehow have been able to divide his mind for the occult stuff. In contrast, Stubbs can get away with religiously-induced nonsense about (Proto-)Uto-Aztecan, because who’s going to take the time and care to ridicule him?

    I’m impressed, reading those follow-up points, at Hansen’s care and patience.

  32. David Eddyshaw says

    As some of the LDS-aligned ccmmenters point out (not all their points are wrong) what Stubbs conceives himself to be doing with the UA/AA stuff is not actual comparative reconstructive work but a purported demonstration of contact and borrowing, which is how he legitimises (in his own eyes) things like inconsistent correspondences.

    That is probably a clue to what has gone so badly wrong, without needing to assume actual bad faith on S’s part. To study borrowing adequately is significantly harder than doing comparative work.

    You need to be familiar with a greater range of languages and language typology, you need to be even more aware of the possibility of pure chance resemblances because the check of regular phonological correspondences no longer works straightforwardly, and above all you need to exhibit an actual plausible social context for the borrowing. You need massive extra helpings of rigour just to achieve similar levels of reliability to a comparativist.

    I can well imagine even a good comparativist getting seriously out of his depth in this without actually realising that he has done so.

    Lameen has done deeply impressive work on borrowing in the Sahel and Sahara, which shows me just what a challenge it must be to do this properly.

    I suspect that there has been a good deal of borrowing between the various Oti-Volta languages, but actually demonstrating that it has occurred in any specific case can be anything from bloody difficult to impossible. Even with the “easy” case of words that are obviously ultimately from Arabic it can be impossible to trace just how they got to (say) Kusaal: you end up making what could easily be quite baseless assumptions about precolonial cultures and their mutual relationships. Only occasionally do you strike lucky with a sound change (say) characteristic of just one potential intermediary language.

  33. Isaac Newton believing in the occult was well within his “actual area of expertise” – and, relative to the state of the art in 17th c. Europe, can hardly be described as “utter silliness”. The fact that an idea didn’t work out is no reason to dismiss it (with the immense condescension of posterity) as something whose silliness should have been obvious from the start.

  34. David Marjanović says

    child porn rings run from pizza parlours

    From the cellar of a pizza parlour that doesn’t have a cellar! Why do most people keep omitting the best part!

    Then Newton must somehow have been able to divide his mind for the occult stuff.

    Not at all. For the most part, I figure, he thought they just didn’t have any implications for each other. And to answer the question of why the orbits didn’t decay (due to friction in the air, IIRC, because people still believed in the horror vacui), he simply said it’s a miracle – God restores the orbits periodically.

  35. In case this rebuttal by Brian D. Stubbs has not been mentioned in this thread:

    “Answering the Critics in 44 Rebuttal Points”

  36. PlasticPaddy says

    He may have felt the occult ideas and the scientific stuff was related, just the example of Galileo and worse, Bruno would cause him to be very discreet; his declaration that God added the energy to prevent orbital decay may even have been tongue-in-cheek. I am also reminded of John Nash, who when asked by a colleague how Nash could believe that space aliens were putting coded messages in newspapers for him, said something to the effect that those ideas seemed to originate in the same part of his brain as the mathematical ideas.

  37. Editor’s note placed above Stubbs’ rebuttal:

    # As might be expected, Stubbs’s efforts have drawn criticism from some, but not all, of his linguistic peers. This article represents a response by Stubbs to those criticisms. Stubbs’s works are admittedly complex and highly technical. They are, therefore, difficult, and it can take quite a bit of work for a reader to assimilate and understand the implications of his arguments. That very complexity and difficulty, though, precludes dismissal of Stubbs’s works out of hand. #

    The attitude expressed by that last sentence is what makes academia so attractive to certain elements. At this blog I once accused Heidegger, and certain other German casters of prosodic whammy, of writing to establish their own exegetical industry.

    But it’s not a point I would insist on, because that would draw me into the epigonal fracas. Whether Heidegger cast whammies or pearls is of no importance to me. I just look the other way.

    “Remarks are not literature, Hemingway” said G. Stein. And a good thing too.

  38. @Lameen: The fact that an idea didn’t work out is no reason to dismiss it (with the immense condescension of posterity) as something whose silliness should have been obvious from the start.

    I agree. But many people are afraid of falling into “relativism”. They have not considered that such a fear is historically of late date, and thus itself parasitic on “relativism”. Sophistry may be a teeny bit younger than the desire for certainty, but the young’uns are not all bad.

    My current reading, Die Soziologie vor der Geschichte by Knöbl, is instructive on attempts to avoid “relativism”, “historicism” etc, by Dilthey, Troeltsch et al, although condescending towards Luhmann. He seizes on “the” and “but” in Luhmann, and misses all the rest. It’s like grabbing Dolly Parton’s tits to diss her songs.

  39. David Eddyshaw says

    From Stubbs’ (attempted) rebuttal:

    but my own three-day investigation into Athapaskan and various East Asian languages convinced me that Athapaskan came from East Asia.

    I think this neatly proves my point about him not actually realising when he’s out of his depth. Three days!
    When the gods wish to destroy a linguist, they first madden hm by pointing him toward long-distance relationships …

    I think the main valid point that Stubbs makes there is that some of the critics (Chris Rogers, anyhow) mistakenly supposed that he was trying to establish a genetic relationship between UA and AA, which he wasn’t. But that doesn’t really help …

    For example: he doubles down on his idea that Nahuatl personal pronouns are borrowed from (or modelled on) the Aramaic imperfective of “to be” and that UA borrowed the Hebrew vav-consecutive … Just how, sociolinguistically, would that have come about (even leaving aside all the other problems …)?

    Interestingly, Stubbs seems to believe that the form of Semitic which influenced UA was Northwest Semitic, early enough that Aramaic and Canaanite were not yet distinct, and in particular that Proto-Semitic final short vowels were still preserved. But the Book of Mormon apparently dates “Lehi”, the Atlantic-crosser, to the reign of Zedekiah, about 600 BCE … and quite what the point of comparisons of UA with Egyptian might be then, I can’t quite see … better reconstruction of the Hebrew of 600 BCE? Moreover, the vav-consecutive is a distinctively Hebrew phenomenon. There is no reason to suppose that it goes back to Northwest Semitic in general. Ugaritic uses the yaqtul form as both jussive and, in poetry, as past (continuing the original state of affairs reflected in Akkadian) but it doesn’t require a preceding wa for the form to have a past sense,

    To my mind, Stubbs’ attempted rebuttal really does undermine one’s faith in his actual scholarship in general; I’m not sure that I can maintain my charitable view that he was merely guilty of going beyond his area of expertise.

    I notice his excursion into genetics; I suspect this is a well-trodden LDS trope.

  40. To my mind, Stubbs’ attempted rebuttal really does undermine one’s faith in his actual scholarship in general

    I had the same reaction. He seems so convinced of his own point of view he has a hard time grasping why anyone might object.

  41. I was once talking with an Indo-Europeanist who turned out to be deeply skeptical about Austronesian, thinking it must be some sort of Nostratic Horror. I said no, our sacred Comparative Method has been applied to AN for an even longer time than it has to IE, and with the same rigor: the existence of Proto-Austronesian is quite secure. (Insert jest here on the semantic consonance of PIE and PAN.)

    What’s different is that it’s easy to see which languages are Austronesian (“What’s your word for ‘five’?” “Lima.” “You’re in.”), grouping them is much more difficult, since physical proximity isn’t worth much. Once we get beyond Oceanic (which might be called the Bantu of Austronesian), “infinite are the arguments of mages”, and all the poor Wikipedians can do is present several trees labeled by their proposers, with dates, because the proposers often change their minds. It’s kind of like the “fallen leaves” non-proposal for Trans-Himalayan (Sino-Tibetan), where we don’t have a tree, just 40 or so leaves on the ground which fell off the unknown branches.

  42. He seems so convinced of his own point of view he has a hard time grasping why anyone might object.

    Michael Faraday wrote in a lecture given in 1819 but not published until 1870 (emphasis added):

    Nothing is more difficult and requires more care than philosophical [i.e., scientific] deduction, nor is there anything more adverse to its accuracy than fixidity of opinion. The man who is certain he is right is almost sure to be wrong, and he has the additional misfortune of inevitably remaining so. All our theories are fixed upon uncertain data, and all of them want alteration and support.

    Ever since the world began, opinion has changed with the progress of things; and it is something more than absurd to suppose that we have a sure claim to perfection, or that we are in possession of the highest stretch of intellect which has or can result from human thought.

    I was suspicious of that fixidity, but the OED lists it, though it says “rare” and under Etymology it sniffs “badly < fix adj. or fixed adj., after fluidity.” Two of its three quotations are from scientists, though, and who knows, it might have caught on as a technical term, leaving fixity to the humanists.

    (Strange that this popped up in my RSS feed today; I didn’t go looking for it!)

  43. David Eddyshaw says

    It seems to be quite common for a top-level classification to be well-accepted but subgrouping to be far from agreed on. Afroasiatic is like that (in fact even Semitic is like that to some extent.)

    Within Volta-Congo, there is little doubt that all the “Gur/Voltaic” languages, even in Manessy’s maximalist use of the term, really are related to one another; but whether they actually form a single branch of Volta-Congo is a very different matter.

    The subclassification of Oti-Volta given by the tree here


    is hopelessly wrong: Buli/Konni are, beyond any reasonable doubt, actually the closest branch to Western Oti-Volta by quite a margin, followed by Yom/Nawdm, which are not particularly close to the Gurma languages at all.

    Adams Bodomo’s classification wrongly puts Buli in as part of “Central Mabia” (i.e. WOV), reflecting the fact, at least, that it evidently is pretty close; but separates Nabit from Kusaal and Talni. Nabit is actually extremely similar to Toende Kusaal, and I’m pretty sure is regarded as a separate language only because the Nabdema don’t self-identify as Kusaasi and are part of the Farefare poilitical system.

    (Also, if a “native Dagbani speaker” tells you that Dagbani is mutually comprehensible with Dagaare and Wali, I would advise you not to put him on the spot by inviting him to demonstrate. Or to buy that used car from him.)

  44. Bulverism

    While searching for the meaning of this, I happened upon the title of what must be a rude anecdote, hardly concealed by the decent obscurity of Franconian dialect: Dschilli-Bulver am Harald san bessdn Schdigg.

    I feel this is relevant to the topic of mutual intelligibility. You may understand what someone says more easily than when they write it – and vice versa. Only after taking this hurdle might you disprove the claim, and only then impugn the motives if so inclined, says Lewis.

  45. I was once talking with an Indo-Europeanist who turned out to be deeply skeptical about Austronesian, thinking it must be some sort of Nostratic Horror.

    I wonder if he had known Isidore Dyen, who could inspire anyone with doubts about the validity of his area of expertise.

  46. David Eddyshaw says

    As for lexicostatistics, Dyen was convinced that quantitative evidence served as a solid method for classifying languages. His consistent reliance on this rather controversial methodology was demonstrated by the fact that his lexicostatistical classification of Austronesian languages, published in 1965, was followed by one of his last major publications, a lexicostatistical classification of the Indo-European language family in1992. He produced the latter study despite the fact that many colleagues had already straightforwardly rejected this method as a valid tool of comparative linguistics after the 1965 book. Nonetheless, he defended his views with all his energy and did not avoid open confrontations.


  47. Yup, that’s our boy. (“Izzy dyin’?” we used to ask each other in the ivied halls of Yale.)

  48. David Eddyshaw says

    I see that his lexicostatistical account of Indo-European is online:


    The avowed objective seems to be to prove that his lexicostatistical method is valid, rather than telling us anything new about Indo-European. Notably, it fails to produce an Indo-Iranian branch; in the discussion, they attribute this to extensive borrowing from Arabic in the Iranian group and to the evidence for Indo-Aryan being based on ancient written sources excluded from the scope of the study. It also seems to make the primary division in Slavonic into Slovenian versus The Rest and gives peculiar results for English and Frisian within Germanic (not altogether surprisingly.) As a demonstration of the pitfalls of lexicostatistics, I think it’s quite interesting …

  49. I don’t think my IE-ist was old enough to have known Dyen personally, but of course I may be quite wrong — on the Internet, nobody knows if you’re an old fart. It also occurs to me that my phrase “Nostratic Horror” is an unconscious echo of “Dunwich Horror”.

  50. David Eddyshaw says

    The Shadow over Moscow …

  51. Of Stubbs’s lengthy rebuttal, I only read the ‘drunk/cactus’ one, and it really doesn’t inspire confidence. He only digs himself in deeper.
    Maybe he did have the Semitic connection in mind when he linked the two in the dictionary, maybe not. In either case, it’s not justified. However, the dictionary is full of examples of so-so matches experimentally put together, as an encouragement for further work. The dictionary is comfortable being indecisive and approximate, and its usefulness depends on recognizing that UA historical linguistics is very much a work in progress.

    @DE: I get the impression (due to knowing nothing about it) that comparative Uto-Aztecan is (relatively) simple, as far as such things go, inasamuch as there is a lot of uncontroversially cognate material to work with, along with excellent documentation by some of the best field linguists ever; Stubbs may have been lulled into a sense of false security about the many pitfalls of comparative work.

    Kinda, kinda not. There are a lot of these approximate matches, which suggests that the historical phonology of UA is very imperfectly known. Maybe it’s where IE comparative linguistics was in the late 19th century.

    Some branches of the family are better documented. The southern languages, many of which still have many speakers, have been and are being better documented than the northern ones. As historical linguistics goes, there have ever been maybe a dozen linguists, total, who have done significant work on the family (and several more working on Nahuan internal relations). The field is dormant (as is much of N. Am. historical linguistics elsewhere): recent work amounts to some person every few years publishing a paper or two.

  52. @DE From Stubbs’ (attempted) rebuttal:

    but my own three-day investigation into Athapaskan …

    I wondered if @DE was cherry-picking that evidence. But reading Stubbs further, he condemns himself far worse. That three days was at the _start_ of Stubb’s career; Other linguists later provided evidence for this claim [of correspondences], which received considerable, but not universal, acceptance.18 Note 18 cites a parade of crackpots, including Edward Vajda. (Who is controlling the wikipedia entry that says meekly “not universal, support among professional linguists”? Vajda could at least point to extra-linguistic evidence of contact from Asia.)

    @DE, earlier re the strength of evidence needed for an explanation from borrowing: You need to be familiar with a greater range of languages and language typology, you need to be even more aware of the possibility of pure chance resemblances because the check of regular phonological correspondences no longer works straightforwardly, …

    So how did Stubbs spend several decades more in academe, studying all sorts of phyla [**] of languages, and it never occur to him “oh, this is a lot harder than you might guess at first”.

    I’m not an academic/never had the privilege of devoting a career to study/have only picked up facts of language from lurking amongst Hat’s generous zones and messuages — but my bullshit detectors went off straight away. Has Stubbs never heard of Goropius or poor Edo Nyland?

    [**] Drawing from another parallel thread: it looks like Stubbs’ whole education has been in Mormon enclaves; Uni of Utah being the only place he knows — where perhaps there is no fostering of critical thinking? Of course they’d teach Semitic languages. But Stubbs seems to know _only_ U-A and Semitic. Would that be the only diet at any Linguistics course outside Utah? (Stubbs’ biog seems to think Hebrew is representative of “Old World languages” — as if Indo-European is much of a muchness.)

    I think it could reasonably be asked whether Stubbs’ whole career is hell-bent on ‘proving’ the LDS claptrap; studying only Semitic and U-A would give a huge confirmation bias.

    To me this more or less condemns religion-controlled education as a thoroughly bad thing, that the State should not be supporting but actively abolishing.

    Are there any reliable linguists studying U-A languages? Should I have any trust whatever in alleged findings about it? Is it all invented out of thin air to ‘validate’ religious mania? Had we better throw it all away and get some actually trustworthy people to record these languages before the remaining native speakers disappear?

    Stubbs: the best Uto- Aztecanists in the world, most holding PhDs in linguistics, have all received my work by now. These men and women have known me for decades, and none of them has spoken to me of incorrect characterizations of linguistic concepts.

    How may of “these men and women” are Mormons?

    How many know a wide spread of language phyla outside U-A and Semitic?

    If a (formerly) respected colleague were to start producing ‘original research’ (as wikip calls it) of this nature, how many would just keep their heads down and avoid the aggravation of trying to combat it? (“Tsch, tsch, and he used to show such promise …”) Stubbs himself says 75% made no comment. I take it “the best UA specialists in the world” would amount to a tiny sample size.

    _Is_ Stubbs merely claiming borrowing of sundry lexemes?

    … shows how Semitic or Egyptian provides the underlying forms that explain seven of nine phonological puzzles that Uto-Aztecanists have not been able to solve since Sapir’s establishment of the language family in 1913/1915. Finally, it addresses the vowel correspondences, the medial consonant clusters, the grammatical and morphological parallels, and ends with unusual semantic combinations preserved in UA.
    [Stubbs’ description of ‘Exploring the Explanatory Power of Semitic and Egyptian in Uto-Aztecan’ 2015]

    Isn’t that title alone claiming more of a systematic relationship? (I agree he’s not claiming a genetic relationship. Something more like Vietnamese taking over the whole Sinitic number vocabulary.)

    Ought a correlation of 400 sets [between one Semitic branch and U-A] be ignored? Even 400 sets is two or three or four times what many Native American language families were founded on.
    [Stubbs’ attempted rebuttal to Hansen, quoting ‘Explanatory Power’.]

    Does 400 sets amount to more than suggestive/’needs further looking into’? — not enough to say “founded on”. So “Native American families” research is founded on sand(?) The whole field has a very low threshold for ‘evidence'(?)

    I [Stubbs] say, no one has ever had the UA liquids figured out. In UA, the liquids and nasals are an as-yet-unresolved puzzle. Some think PUA had no liquids (that PUA *n and *t are the source of later liquids); others think PUA had one liquid …

    And yet they’ve reconstructed PUA. There seems to me ample wiggle-room to sail a battleship through. Gisajob: I’ll find borrowings between U-A and Basque. We at least know Basque speakers sailed the Atlantic.

    BTW, re Proto-U-A (in which Stubbs also seems to be active): since there’s no writing system, how does anyone reconstruct “an unusual vowel inventory” [wp]? Can I believe any of this? ” It would have been spoken by Mesolithic foragers in Aridoamerica, about 5,000 years ago.” Yeah, at about the vintage of Nyland’s ‘Saharan’.

  53. Oww!
    Glancing at Stubbs’s rebuttal, I found that he reconstructs PUA ‘domestic animal’ (#37).
    He just happens to attach it to the Hebrew word for ‘livestock’.

    That said, AntC, University of Utah is not a “Mormon enclave”, and did and continues to have an excellent linguistics department. Comparative UAnist Wick Miller was there, as was Lyle Campbell for a while. And Vajda is not a “crackpot”, nor are any of the other names mentioned in the Dene-Yeniseian footnote, although yes, DY is not that solid.

  54. David Marjanović says

    Don’t confuse the University of Utah, which is simply the state university of Utah (i.e. public), with Brigham Young University, which is a Mormon institution – though even the BYU (aka WhyBeYou) has a perfectly good geology department with entirely normal paleontologists in it.

    since there’s no writing system, how does anyone reconstruct “an unusual vowel inventory” [wp]?

    Uh, the PIE vowel inventory is plenty unusual…

  55. David Marjanović: From the cellar of a pizza parlour that doesn’t have a cellar! Why do most people keep omitting the best part!

    To me, and I suspect some other liberal Americans of my generation, that doesn’t sound like the “best” part of the whole silly, sordid narrative, because that gag had been used before. It sounds too much like real life is cribbing story ideas from Pee Wee’s Big Adventure.

    @John Cowan: To me (a professional scientist, although one who is, at the moment, commenting on a blog post as a way of avoiding the scientific task of rewriting a manuscript draft sent to him by a graduate student), fixidity sounds at least as good as fixity.

Speak Your Mind