June 2, 2013

Basque linguistics: Frank criticizes Lakarra

I mentioned recently how the pope of Basque linguistics, Joseba Lakarra, is being more and more criticized. Here however I will briefly discuss another such criticism.

Frank, Roslyn M. 2011. Repasando a Joseba Lakarra: Observaciones sobre algunas etimologías en euskera a partir de un acercamiento más cognitivo (Reviewing Joseba Lakarra: Observations on some etymologies in Basque language from a more cognitive approach). ARSE 45: 17-64. Available at Academia.edu (in Spanish only).

On one side the alleged Latin/Romance alleged etymologies by Lakarra are self-exposed as a total fraud. Some self-explanatory examples:
  • turpe (Lat. foul, dirty) → *durpe*burdeurde (pig, boar; metaphorically only: dirty - the common word for dirty is zikin)
  • timor (Lat. fear) → … *dirbur*birdurbildur (more commonly beldur: fear¹)
  • hierba (Sp. grass, herb) > *erbar > *berar > belar / bedar (grass, there's actually a real Latin/Romance derived word: zerba = herb)
  • caninu (from. Lat. caninus: rel. to dogs) → *ahinu*ahiun*ha.inhagin (molar)
  • *(la)grima (Sp. lágrima: tear; grima: coll. pity) > *girma > *girna > *nirga > nigar (more commonly negar: cry, tears, negar egin: to cry, to express emotion by tears)

One of the many problems with this last 'etymology' is that Basque language hates double consonants (with the exceptions: tx [ch], ts, tz) and that therefore the hypothetical *grima would become *girima or *kirima first of all, again unattested.

But in general all the Lakarran hypothetical etymologies are self-defeating, needing of several unlikely intermediate variants, each of them extremely unlikely, and would not stand any minimally serious scrutiny. The problem is that there is very little of that.

Another criticism is that he uses almost only his own work, often not even published, as evidence of his own conclusions. This is mere pseudoscience but somehow he gets away with it.

Other issues that Lakarran conjectures (calling that 'garbage in - garbage out' speculation "theory" would be way too generous) pose are grammatical. Unlike the late North American linguist Larry Trask, who establishes that Basque has primarily a SOV grammatical structure, for Lakarra the proto-Basque had no grammar whatsoever: no SOV, no ergative, no agglutination, no verbal inflections... In other words: for Lakarra proto-Basque speakers were probably not even human yet, but that proto-Basque is only estimated to have existed some 3000 years into the past, so...

For Lakarra, proto-Basque would not be agglutinative but isolating. This isolating characteristic of some languages is actually restricted to East Asia (the most notable example is Chinese, especially classical Chinese). However real Basque has a clear agglutinative tendency, what seems highly inconsistent with Lakarra's formulation.

The bulk of the study however only touches Lakarrism somewhat obliquely, dwelling in some depth on the etymology of hatzapar² (animal claw), which Lakarra forces to derive from Sp. garra (same meaning) but Frank considers 100% Basque, deriving from the more widespread hatzamar (i.e. the toe that means ten, the big toe), and the grammaticalization of gai, originally ability, able, potential (also matter, substance) but now inserted in many words like zerga(i)tik (why, from zer(en)-gai-tik) or -gale (feeling of the prefix: edagale: thirsty, etc.)

The study ends with a criticism of the group Monumenta Linguae Vasconum, which is led by said Joseba Lakarra (and co-participated by Blanca Urgell, Gidor Bilbao, Ricardo Gómez, Julen Manterola Agirre, Mikel Martínez and Céline Mounole). This group is preparing, at the snail pace of a mere 500 words per year, with public financing, an etymological dictionary of the Basque language, which will be no doubt founded only on Lakarra's own speculations. Since 2007 Lakarra is member of the Academy of the Basque Language, Euskaltzaindia, what means that his work is de facto being backed by this influential organism.


Notes (my own ideas):

¹ Beldur (fear) surely derives from bel(tz): black in fact, possibly: *bel-adur: black humor.

² I would also consider *hatz-adar (toe-horn, toe's horn) as possible origin of both forms.


  1. *(la)grima (Sp. lágrima: tear; grima: coll. pity) > *girma > *girna > *nirga > nigar

    This one is particularly hysterically funny. Does he really think "la" is a prefix? It used to be "da", as in PIE *dáḱru- --> dacrima --> lacrima (the latter two, Latin). Germanic also has preserved the leading d/t sound: Engl. tear, Germ. Träne.

    1. He doesn't even seem able to explain the loss of initial la- at all.

      I'd rather be tempted to link (if anything at all) 'negar' with that PIE '*dakru' you mention, although it'd require some phonemic shifts, they seem coherent on first sight: I would hypothesize a shared proto-word like maybe *mækaR for both terms.

      *mækaR → *bakR' → *dakru (consistent with IE tendency to suppress vowels forming tr, kr, gr, kl, gl, br, bl etc.)
      *mækaR → *nekar' → negar (consistent with Basque tendency to retain such vowels or even insert them, suppressing kr, tr, gr, kl, gl, br, bl, etc.)

      I'm no linguist though but I'm tempted by the idea of a very remote Basque-PIE connection on several reasons. If real, it seems reasonably normal that a shallow linguist like Lakarra could feel some connections but be unable to explain them properly because such a remote connection is beyond their imagination.

  2. Heck, if I were a linguist (which I am not), I would derive this from something like [dl]ac[jr]-ougk --- the lactating eye. Here, [ld] and [rj] are some fricatives that are intermediate to the indicated consonants.

    1. This sounds far fetched to me, sincerely. One of the main errors of Lakarra is to imagine that polysyllabic words derive from root monosyllabic forms, as if language would have been invented just a few thousand years ago and not have hundreds of thousands, maybe millions of years of background. Most probably something as emotionally intense and basic as 'tear' was always a fundamental word on its own right, with maybe the occasional exception and therefore not related to any other fundamental concept since the night of time. It should be in Swadesh lists along with other fundamental words like 'hand' or 'blood' or 'we' or 'walk'...


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).