I haven’t rolled my eyes at the very thought of artificial languages since I had the pleasure of reading Arika Okrent on the topic (see this 2009 LH post), but sometimes they’re just too weird for me, and Babm is such a case. To quote the Wikipedia article:

Babm (pronounced [bɔˈɑːbɔmu]) is an international auxiliary language created by the Japanese philosopher Rikichi [Fuishiki] Okamoto (1885–1963). Okamoto first published the language in his 1962 book, The Simplest Universal Auxiliary Language Babm, but the language has not caught on even within the constructed language community, and does not have any known current speakers.The language uses the Latin script as a syllabary, and possesses no articles or auxiliary verbs. Each letter marks an entire syllable rather than a single phoneme. Babm follows a sound-based rule set, which Okamoto outlines in his book. He states “Nouns are coined from three consonants and one vowel, verbs from one or two vowels between two consonants at the beginning and at the end. Adjectives, adverbs, pronouns, numerals, and propositions have respectively their own peculiar form.”

The language has in common with some 17th-century artificial languages an over-riding concern with taxonomy, and providing a universally consistent set of names for chemicals, etc.; the author’s “scientific” preoccupation is a contrast to the socio-political mandate of Esperanto, although the 1962 book is certainly not lacking in statements about world peace. Okamoto hopes this “simple” language would become universally useful.

The Phonology section tells us that “Every consonant in Babm must be followed by a particular short vowel, with the exception of /k/ which can be followed by any vowel. Vowels that are attached to nouns are short vowels by default, and those not attached to nouns are long, but vowel length can be modified.” The “Consonant Inventory” table shows the inherent vowels: [bo], [co], [de], etc. Oh, and c and k are both /k/, for added fun. I’m sympathetic to the desire to create what one imagines is the ideal language, but I simply can’t grasp the mentality that would think this was a plausible language at all, let alone one that “would become universally useful.”


  1. I read Okamoto’s Babm in my first semester of grad school half a century ago. My impression was that he didn’t quite grasp the alphabetic principle, and that he would have been much better off making a language on these principles but based on kana.

  2. Oh, and c and k are both /k/, for added fun.

    Looks like q is /k/ too, though the example word is “cook” — not sure whether it’s meant to be qook or cooq or qooq.. My vote would be kook.

    The example words for a and z are “armadillo” and “zoloft”.

    I’m tempted to add a {{notability}} tag to that article.

  3. “x” is also /k/, or rather is also a /kV/ syllable.

    Looks like q is /k/ too, though the example word is “cook” — not sure whether it’s meant to be qook or cooq or qooq.

    What it means is that “q” is pronounced /ku/, corresponding to the “coo” of “cook”. The Wikipedia page seems to have managed to make things even less clear than Okamoto’s original text.

    I’m not sure it’s true that /k/ can be followed by any vowel. “c” is /ko/, “k” is /ke/, “q” is /ku/, and “x” is /ki/, but I don’t know how you’d indicate /ka/.

    Okrent does mention Babm, by the way, devoting a bit over two pages to it in chapter two of In the Land of Invented Languages (a lot of that’s quoted example sentences, though).

  4. Well, auxlangers often become the butt of ridicule, as they usually are the kind of people on a mission who earnestly claim that their auxlang is either (1) the easiest to understand and learn or (2) the most logical language ever invented (or both), and often claim to boot that many or all of humanity’s problems could be solved if just everybody would speak their auxlang.

  5. Sure, it’s just that I can usually see at least vaguely where they’re coming from.

    I’m tempted to add a {{notability}} tag to that article.

    If Rodger C studied it in grad school, I think it’s clearly notable. (And in any case I deplore the people who vigilantly doubt the notability of anything that seems silly to them while ignoring the fact that every rivet in Star Trek and every pixel in every video game has its own article. Not lumping you in with them, just venting.)

  6. I like his suggestion that ‘r’ ([ra]) should be used for ‘quote unquote’.

  7. In addition, j is [zi] whereas z is [zo]. So there are 17 consonant phonemes but only 21 phonotactically licit CV syllables.

  8. I am reminded here of a description of an experiment designed to teach “putatively impossible grammatical structures” (cited from Newmeyer 2005):

    The second noteworthy attempt to teach subjects putatively impossible grammatical structures is described in Musso et al. (2003). In the first of two functional Magnetic Resonance Imaging (fMRI) studies, twelve native German speakers learned three grammatical rules of Italian and three UG-impossible rules using the Italian lexicon. The first of these latter rules was the placement of the negative word no always after the first word of the phrase; the second was the formation of interrogatives by the inversion of the linear sequence of words in the sentence; the third demanded that indefinite articles agree with the last noun of the phrase. In the second fMRI study, eleven native German speakers participated in a similar experiment, but this time with Japanese. The first two rules of ‘unreal Japanese’ paralleled those of ‘unreal Italian.’ Since Japanese has no articles, the third artificial rule was different. It involved creating a past tense by adding the suffix -ta, not to the verbal element, as in real Japanese, but on the second word counting from the right, in all sentences.

    I can’t decide which is worse: both Babm and this horseshit is equally insane, but the experiments at least had some purpose. On the other hand, the experiments are just so ming-bogglingly insane, completely ignoring anything that is not narrowly conceived syntax. I mean, “second word counting from the right”? How does that even fall within the domain of grammar rules? It’s the same kind of rule as “every third Thursday after full moon”.

  9. The second word counting from the right, in all sentences.

    This one sounds very easy in an OV language.

  10. I didn’t study Babm, I just ran across it in the wonderful IU Library (before it was the Wells, or even in that building).

  11. Oh! Well, I still say it’s notable enough for WP.

  12. J.W. Brewer says

    I had the same suspicion that Rodger C confirmed — what I read while in college (or in law school, for that matter) was a rather different set of things (with admittedly *some* overlap) than what I studied while enrolled in ditto.

  13. I see that in fact the Babm Wikipedia article survived a 2007 deletion request on the basis that Babm was “widely discussed by linguists in the 60s and 70s”, although a decade later the article itself still doesn’t actually state that. The self-published work of a crank is generally not notable, whether it’s a conlang or Star Trek fan fiction; but of course if the work has been widely discussed by academics then that suffices to make it notable for Wikipedia purposes.

  14. The first of these latter rules was the placement of the negative word no always after the first word of the phrase

    At least some Wackernagel clitics work like that (others appear after the first construct, which is less un-UG).

  15. David Marjanović says

    and three UG-impossible rules […] the second was the formation of interrogatives by the inversion of the linear sequence of words in the sentence


    On the other hand, the experiments are just so ming-bogglingly insane, completely ignoring anything that is not narrowly conceived syntax.

    My superficial impression is that narrowly conceived syntax is most of what UG is about.

  16. Eli Nelson says

    @David: Huh? I can’t find anything in the grammar of Sumerian interrogative clauses that resembles inversion of the order of all the words in the clause. This seems to indicate that most kinds of interrogative sentences had the same word order as declarative ones:

  17. I don’t recall where I encountered Babm, possibly in a Cornell U. library, but I wrote to the author asking about it, and his son (he’d passed away a few years previously) sent me the English-language book on it; Sekaigogakuron (‘world language essay’), about Babm, in Japanese; and another Japanese book by Fisk Okmot, which I offered to the University of Pittsburgh library, which I imagine quickly tossed it. It seemed to be about physics and Babm. I admire the idea, if not the usability.

  18. David Eddyshaw says

    I think the point, insofar as there is one, in the Musso stuff, of invoking the fairytale “UG”, is that the impossible rule counts words rather than sentence constituents. Chomskyites like to imply that (a) the Master discovered the concept of sentence constituents, hitherto shrouded in mystery, and (b) the mere existence of sentence constituents as an actual thing proves the infallibility of the Master.

    I can imagine a perfectly workable language sticking clause-level negative particles in the Wackernagel position. Lingala always puts the sentence negation particle last, and lots of Australian languages have a sort of omniclitic in the Wackernagel position which encodes most of what goes on verbs in the way of flexion in SAE languages; though I can’t offhand think of one that does this with negation, I can see no reason at all why that should be impossible.

    All indefinite articles in Kusaal and Welsh agree with the last noun of the phrase. And also don’t agree. Or agree on Tuesdays but not other days. Whatever you like. There are no counterexamples.

    In any case, it’s a safe bet to assume that anything involving fMRI is completely valueless in understanding actual grammar. Or anything else, really.

  19. Lars Mathiesen (he/him/his) says

    the Wackernagel position: Oh, you did, did you? GIF or it didn’t happen.

  20. David Eddyshaw says

    I tried, but Akismet is so narrow-minded …

    I’m afraid that you’ll just need to use your imagination.

  21. Lars Mathiesen (he/him/his) says

    En serio, I’m reading Wackernagel’s Revenge: Clitics, Morphology, and the Syntax of Second Position, Stephen R. Anderson. Language, Vol. 69, No. 1. (Mar., 1993), pp. 68-98 (DOI:10.2307/416416) with great interest, and some chuckles at the author’s summaries of presumed G/B analyses of which there are usually four incompatible ones. (There would be other analyses to poke fun at now).

  22. David Eddyshaw says

    Kusaal has several obligate second-constituent clitics, but the first constituent always has to be the clause subject, so I suppose they don’t really count as echt Wackernagel clitics. Well, can’t have everything …

    Interesting paper (as you’d expect, from Mr Clitic himself.) Apparently W himself actually did mean “after the first word“, rather than “constituent.”

    Interesting too to be reminded about Sanskrit, and probably PIE, accentlessness of verbs in main clauses: Kusaal basically does this too (specifically, lexical verb tone is neutralised in positive-polarity main clauses), though I can’t think of any way of making this to be anything but an odd coincidence: it’s hard to see any general principle behind it.

    I hadn’t heard of W’s idea about the development of verb-second order in Germanic main clauses. Ingenious …

  23. Are there any languages (verb-final presumably) with a syntactically significant penultimate position, a sort of mirror-image Wackernagel?

  24. David Eddyshaw says

    Hungarian focuses constituents by placing them just before the verb, but it’s not verb-final.

Speak Your Mind