Weird languages?

« previous post | next post »

Tyler Schnoebelen, "The Weirdest Languages", Idibon blog 6/21/2013:

The World Atlas of Language Structures evaluates 2,676 different languages in terms of a bunch of different language features. These features include word order, types of sounds, ways of doing negation, and a lot of other things—192 different language features in total. […]

The data in WALS is fairly sparse, so we restrict ourselves to the 165 features that have at least 100 languages in them (at this stage we also knock out languages that have fewer than 10 of these—dropping us down to 1,693 languages).

Now, one problem is that if you just stop there you have a huge amount of collinearity. Part of this is just the nature of the features listed in WALS—there’s one for overall subject/object/verb order and then separate ones for object/verb and subject/verb. Ideally, we’d like to judge weirdness based on unrelated features. We can focus in on features that aren’t strongly correlated with each other (between two correlated features, we pick the one that has more languages coded for it). We end up with 21 features in total.

For each value that a language has, we calculate the relative frequency of that value for all the other languages that are coded for it. So if we had included subject-object-verb order then English would’ve gotten a value of 0.355 (we actually normalized these values according to the overal entropy for each feature, so it wasn’t exactly 0.355, but you get the idea). The Weirdness Index is then an average across the 21 unique structural features. But because different features have different numbers of values and we want to reduce skewing, we actually take the harmonic mean (and because we want bigger numbers = more weird, we actually subtract the mean from one).

Tyler offers a spreadsheet with the full set of languages and scaled feature-values behind the ordering.

There's a fairly large prior literature on how to quantify similarities or differences among languages, though usually the goal is to look for relationships rather than to identify conformity vs. freakiness.

An older information-theoretic measure can be found in Patrick Juola, "Cross-entropy and linguistic typology", CoNLL 1998. A recent effort, pointed out to me by Pieter Muysken, is Harald Hammarström & Loretta O'Connor, "Dependency-Sensitive Typological Distance" (those are slides from a Workshop on Comparing Approaches to Measuring Linguistic Differences, 24-25 October 2011, University of Gothenburg).

The problem here, in my opinion, is that there's no clearly-good way to decide how to choose and weight "typological" features of this type. And since Schoebelen used only 21 WALS features in that blog post, a few more or less would make a pretty big impact.

It's easy to list dozens of plausible features that happen not to be covered in WALS, and seem roughly as important as (say) whether comitative and instrumental are the same or not. For example, if we counted auxiliary-like uses of  a verb meaning do or make, English would (according to a presentation by Aaron Ecay at a historical syntax workshop this past weekend) join Korean (Hagstrom 1995), a Northern Italian dialect (Benincà and Poletto 2004), and perhaps the earliest attested Icelandic (Viðarsson 2009) as one of a minority of languages with this feature, thereby boosting our freakiness coefficient.



57 Comments

  1. Howard Oakley said,

    July 2, 2013 @ 2:27 pm

    I find it puzzling that a language such as Turkish – which can perpetrate
    döğüṣtürtülmeyebiliyormuṣsunuzdur
    as a single verb meaning
    it is presumably the case that you sometimes were not made to fight
    (Göksel & Kerslake 2005, 8.2 7 page 74)
    – is rated highly non-weird. Perhaps having English as my first language has made my outlook weird…
    Howard.

  2. Rod Johnson said,

    July 2, 2013 @ 2:35 pm

    But what makes that weird? Un-English-like, sure.

  3. Marek said,

    July 2, 2013 @ 2:57 pm

    Aren't non-weird languages weird by not having at least a couple of weird features?

  4. Theodore said,

    July 2, 2013 @ 3:46 pm

    @Marek: That depends on the frequency at which those "non-weird" languages occurred in the sample :)

    @Rod Johnson's: I have wondered about various ways one might quantify the "distance" between a pair of languages in a way that's still meaningful in comparison to the distance between other pairs of languages.

  5. leoboiko said,

    July 2, 2013 @ 4:47 pm

    > The data in WALS is fairly sparse, so we restrict ourselves to the 165 features that have at least 100 languages in them (at this stage we also knock out languages that have fewer than 10 of these—dropping us down to 1,693 languages).

    Surely the sparseness isn't random, and there's a good chance that the most interesting outliers lie precisely in the less well-described end? And then there's the possibility that typological features that can model the most unusual phenomena haven't even been codified yet. I'd also like to borrow a point by Haspelmath: even if all current ~6500 or so languages were described in equal footing, which is far from being the case, that would still be a very small and biased sample of all languages that have ever existed.

    That being said, I've always thought that the way English changes word order for questions is just weird. Sorry guys. It is. The funniest thing is when you're reading an English textbook to another language and they go to great pains to explain that you don't have to switch anything for questions, just raise the intonation or add a particle.

  6. leoboiko said,

    July 2, 2013 @ 4:50 pm

    (To avoid misunderstandings: I'm being facetious. I don't think "weird" is a good word to describe linguistic features, even if they are infrequent in the available samples.)

  7. Rubrick said,

    July 2, 2013 @ 4:52 pm

    I thought it was well established that the weirdness of a language was strictly a function of how much verbing it had undergone.

  8. marie-lucie said,

    July 2, 2013 @ 5:19 pm

    leoboiko: I've always thought that the way English changes word order for questions is just weird

    Surely English is not the only one that does that. What is weird about English questions is the obligatory use of an auxiliary if there is not one in the corresponding statement.

  9. John Lawler said,

    July 2, 2013 @ 5:21 pm

    @Rubrick: If that's not totally ironic, consider polysynthetic languages like Lushootseed (Puget Salish), where virtually every root is verbal, with nouns made to order.

    Even question words in Lushootseed are verbs:

    gʷat ‘(to) be who?’ ~ gʷat-ciʔiɫ ‘Who is she?’
    ʔәxíd ‘(to) be how?’ ~ ʔәs-ʔәxíd-čәxʷ ‘How are you[sg]?’
    čad ‘(to) be where? ~ čad-әxʷ-tiʔiɫ ‘Where is he now?’

    (ʔәs- ‘temporary stative aspect’, -әxʷ ‘now’)

    See page 2 here for more weirdness, in a number of languages.

  10. garicgymro said,

    July 2, 2013 @ 7:27 pm

    "For example, if we counted auxiliary-like uses of a verb meaning do or make, English would (according to a presentation by Aaron Ecay at a historical syntax workshop this past weekend) join Korean (Hagstrom 1995), a Northern Italian dialect (Benincà and Poletto 2004), and perhaps the earliest attested Icelandic (Viðarsson 2009) as one of a minority of languages with this feature, thereby boosting our freakiness coefficient."

    Ecay didn't mention Welsh?

  11. Lauren Gawne said,

    July 2, 2013 @ 7:59 pm

    It would not have attracted nearly as much attention if it were titled "One of the more statistically improbably language in WALS" instead of "The weirdest languages".

  12. Rod Johnson said,

    July 2, 2013 @ 8:33 pm

    I always thought the canonical term was "funny languages."

  13. Jerry Friedman said,

    July 2, 2013 @ 10:27 pm

    John Lawler: Rubrick was referring to a Calvin and Hobbes quotation, "Verbing weirds language."

  14. Xmun said,

    July 3, 2013 @ 1:48 am

    @Marie-Lucie: "What is weird about English questions is the obligatory use of an auxiliary if there is not one in the corresponding statement."

    Not quite obligatory, though of course that's the general rule. Consider, for example, Rupert Brooke's line:

    Stands the Church clock at ten to three?

  15. Pete said,

    July 3, 2013 @ 3:03 am

    Xmun: That example is not from modern standard English, which I think is the dialect we're talking about here.

  16. Rohan F said,

    July 3, 2013 @ 3:06 am

    Not quite obligatory, though of course that's the general rule. Consider, for example, Rupert Brooke's line: Stands the Church clock at ten to three?

    Poetic and literary conventions permit archaicising constructions in many contexts, but I seriously doubt any native English speaker would consider "Stands the Church clock at ten to three?" grammatical in modern English idiom.

    Apart from be – and even do requires auxiliary do in questions ("Does he do it?") – I would have thought the only remaining exception is lexical have (as "Have you any steak?", "Have we the time?"), and even that's obsolescent in many English varieties.

  17. michael farris said,

    July 3, 2013 @ 3:42 am

    From a European standpoint I would say two weird things about English are

    1 third person singular being the most marked verb form (when it's usually the least marked)

    2 it doesn't have an equivalent of the basic present tense found in most European languages

    I have no problem with calling languages (or language features) weird. IME all but the most unmotivated language learners are drawn to weird and not enthused by the typical.

  18. Jongseong Park said,

    July 3, 2013 @ 4:18 am

    What would the un-weirdest possible language look like, either based on WALS data or preferably a much more extensive set of features? I'm sure an enterprising conlanger could create a list of the criteria that such a language should satisfy (e.g. the most typical word order, the most typical phoneme inventory, etc.) and make one up from scratch.

  19. Alex Bollinger said,

    July 3, 2013 @ 4:21 am

    French changes word order. You don't have to, you can just change intonation or add "Est-ce que" to the beginning, but you can only get away with doing that some of the time if you want to sound more fluent and educated.

    But totally agree, it's weird to talk about the weirdness of a language. Is there even an agreed upon, formal definition of "weird" in the literature of any science? Language, which is highly correlated with national identity and culture and ethnicity, seems like the worst place to start testing value-laden, controversial terms like "weird."

    (Then again, which way does the value go? I think speaking a weird language is better, but then that's just me.)

  20. Jongseong Park said,

    July 3, 2013 @ 4:43 am

    What I haven't seen mentioned yet here is the problem of weighting data points from different languages, which brings up the perennial problem of dialect vs language. At one extreme for the sake of argument, you could count just one Berber language, say, or Tibetan, in which case a unique feature would be considered highly weird. But if such a feature is shared by all the Berber languages, or by all the Tibetic languages, then depending on how you count the languages you could end up with several dozen data points of the same feature, in which case it looks a lot less weird.

    Also, should the number of speakers be taken into account? If a feature is unique to Mandarin and English, that may look weird if we weight all data points equally, but it would be a feature that is natural for a significant portion of the world population.

  21. Pete said,

    July 3, 2013 @ 4:45 am

    Jongseong Park: Yes, you could create a conlang with all the commonest features and the most unremarkable value for each variable, and that would be the "least weird" language possible…but generally conlangers are more interested in giving their languages as many cool features as possible, so it probably wouldn't be a very popular project in the conlang community!

  22. Pete said,

    July 3, 2013 @ 4:56 am

    Jongseong Park, re your comment at 4.43am:

    That's right. Also, shouldn't some account be taken of features shared by related languages, when calculating the weirdness of a feature? For example if you treat all the Berber languages as separate then that would tend to understate the weirdness of their shared features, for example the very rich initial consonant clusters. It's hard to imagine how you'd correct for this.

    Regarding the number of speakers, there are certain features that are strongly correlated with number of speakers. For example the richest consonant inventories are often found among big global languages like English, Arabic and Madarin, while a highly complex morphology tends to be associated with a very small speech community. So maybe you'd want to control for number of speakers rather than weighting by it? I'm not sure what the answer is.

  23. Xico said,

    July 3, 2013 @ 6:01 am

    What caught my eye is that Spanish and Portuguese exhibit widely different scores (0.79 and 0.17 respectively) and yet these languages are almost identical.
    Looking at some of the differences reported in WALS, I can see three problems:
    — Some of data is simply wrong. For instance, Portuguese is described as using a question particle in polar questions. The closest thing to a question particle would be "é que" (or just "que" in Brazil), similar to French "est-ce que", but in Portuguese it is not used in polar questions, only in wh-questions.
    — A very large number of features are just not specified for one of the languages. Although some of these cases do correspond to differences (Uvular consonants), most cases (almost all in fact) do not. If this can account for such a wide difference in scores (0.6), it a serious methodological flaw.
    –Some differences seem unimportant. For instance, one genuine difference is the presence of nasal velars. Spanish has them, Portuguese doesn't, but where Spanish has a nasal velar after a vowel, Portuguese usually has a nasal vowel instead of the two. The perceptive difference between a nasal vowel and a vowel followed by a nasal velar is not that great.

    [(myl) The questions that you raise are serious and to the point. This is partly a traditional problem for typological surveys, namely that the "facts" in them (even if supported by references, as the WALS facts are) are often questionable or even arguably wrong. And it's partly a traditional problem for the use of typological features to measure similarity or difference among languages: the features and their values, even if entirely accurate, cannot easily be evaluated on a common scale of importance (whether historical, perceptual, or whatever).]

  24. Joseph F Foster said,

    July 3, 2013 @ 7:53 am

    Re query by Park and related comments on relation of number of speakers to "weirdness".

    The unit of analysis in Linguistics, as in Anthropology whence much the Cross-Cultural / Cross-Linguistic tradition of study sprung, is the language (or the society), not the individual. So if Mandarin Chinese is an exception to some otherwise overwhelmingly true generalization about language, it is not, as one sophomore in a class of mine once proposed, "a pretty big exception".

    Nope. It is no bigger exception than a language with only, say 10 thousand speakers, would be.

  25. Jongseong Park said,

    July 3, 2013 @ 8:37 am

    Joseph F Foster:
    That may be true for linguistic analysis, but once we're discussing slippery concepts like weirdness for just a silly exercise it wouldn't necessarily be a given. We could be talking about a teen-social-groups kind of weirdness, where members of the most numerous social group will be considered normal by default even if it's the odd one out when comparing social groups one to one.

  26. Mark P said,

    July 3, 2013 @ 10:48 am

    @leoboiko – It's quite common in English to ask a question by intonation. In German, on the other hand, it seems to be less so. I remember someone asking a question in a German class in German by using intonation, and the instructor (actually a German) had to think a moment before he realized that what seemed to be a statement was actually a question.

  27. Brett said,

    July 3, 2013 @ 10:58 am

    @Mark P: It's been a while since I had the chance to ask many questions in German, but I remember learning that you could ask questions in German using intonation only (no reversal of word order), and I'm pretty sure that I did so with native German speakers and had no noticeable problems being understood. On the other hand, they may have been momentarily confused and I just didn't notice. And I can't really comment on how common that kind of question is in German. (Even in English, indicating a polar question with intonation only is much less common than also using word order—although much more common than questions with word order inversion but no modal, like "Stands the Church clock at ten to three?" above, which is distinctly archaic.)

  28. Raempftl said,

    July 3, 2013 @ 12:41 pm

    @Mark P, @Brett: I'm a German native speaker and would agree with Mark P. that intonation questions are rare in German. And if they are used, most of the time they will be used as a retoric device indicating sarcasm or disbelief.

    Consider the following example:
    Scheint die Sonne? – Does the sun shine?
    Die Sonne scheint? – The sun is shining, oh really?

    Acutally, German does not make use of intonation as much as other languages. German students of English have to practice intonation. As apparently using German intonation makes them sound "boring" and bossy.

    My personal theory is that this is due to the abundance of modal particles in German which often have the same function as intonation in English and other languages.

    For all I know those modal particles are a rather weird feature of German.

  29. Jim said,

    July 3, 2013 @ 1:00 pm

    "Even in English, indicating a polar question with intonation only is much less common than also using word order—"

    And often the two are not exactly equivalent."

    "Are you going?" sounds like a simple request for information to me, but "You're going?" sounds like like an expression of disbelief.

    Something similar is going on in Mandarian and I think in Cantonese too, where there are two forms of polar questions, one with -ma as a final particle and one with V neg V eg. "Ni3 qu4 bu qu4". The explanation I got was that the V neg V form was more emotionally neutral but the -ma version was a little querelous.

  30. Ellen K. said,

    July 3, 2013 @ 1:05 pm

    Seems to me there's a difference between these two questions:

    You want to go to the store?
    Do you want to go to the store?

    While the first could be used with the meaning of the 2nd, it can also mean something entirely different, that can't be conveyed with the 2nd. The first can be used for verification, but not the 2nd.

    Does something like that apply in other languages?

  31. Jason said,

    July 3, 2013 @ 1:39 pm

    Weirdness is to me relative to language family.

    English is the only language in the European branch of the Indo-European family that lacks grammatical gender. That one fact makes it extremely weird, IMO. On the other hand, the presence of serial verbs in, say, Paamese is weird if you are comparing it to a set comprised mostly of major (European dominated) global languages, but within its own family of Oceanic languages, serial verbs constructions are utterly commonplace.

  32. JS said,

    July 3, 2013 @ 2:48 pm

    @Jim, The explanation I got was that the V neg V form was more emotionally neutral but the -ma version was a little querelous.
    Surely the opposite must be closer to true? And incidentally, I've been noticing younger Mandarin speakers who seem completely to have abandoned ma in favor of intonation.

  33. Jim said,

    July 3, 2013 @ 3:02 pm

    JS, probably not, since the V neg V form is so much more common and thus likelier to be neutral. Incidentally, intonation has always been a third way of asking a question, but again, not in the simple information gathering mode, but more as a way to confirm some earlier informration or statement, rather then same as in English.

    Ellen, that's the same reading I have on those two options.

  34. Tracy said,

    July 3, 2013 @ 3:32 pm

    I had never heard of WALS before — now I'm going to get nothing done for the rest of the day . . .

  35. julie lee said,

    July 3, 2013 @ 4:02 pm

    @Howard Oakley

    The 13-syllable verb in Turkish which translates into a complex sentence in English is most interesting.

    It recalls numerous Chinese 4-character (4-syllable) expressions that are compressions and often treated as one word. One example:
    ZUI-SHENG MENG-SI 醉生夢死 ("drunk-life-dream-death") which literally means "living as if life were but a state of drunkenness and death were but a dream". Paraphrased into English: "Eating, drinking, and making merry while Rome burns and the writing is on the wall", or "everyone going about business -as-usual while the nation is heading towards collapse and Earth is racing towards doom", and the like. I can't think of an expression in English that expresses this in 4 syllables.

  36. JS said,

    July 3, 2013 @ 6:51 pm

    @Jim– Interesting… your claim about relative frequency is surprising to me, but there you go. Is it based on a corpus count?
    Whatever the case may be as regards frequency, though, I feel confident that VnegV stands out as the marked form, being particularly assertive or insistent. Not sure how to defend this intuition, but perhaps note S daodi V bu V? vs. *S daodi V ma?

  37. Carl said,

    July 3, 2013 @ 11:13 pm

    "I can't think of an expression in English that expresses this in 4 syllables."

    The obvious answer is, "Eat, drink, and be merry," which is five syllables, hardly any different. I find nothing weird about the fact that a language with thousands of years of poetic history can easily make allusions.

  38. JDS said,

    July 3, 2013 @ 11:47 pm

    @Carl, I think "fiddling while Rome burns" is closer to the Chinese saying than "eat, drink, and be merry."

    "Eat, drink, and be merry" doesn't seem right to me for two reasons. 1) That's really only half the saying and you need the other half "…for tomorrow we die" to get close to the meaning "while the world is headed for collapse" that Julie mentioned. And 2) I don't actually think this saying implies that the world is headed for collapse. I've always understood it to mean individuals should enjoy their life because who knows what will happen, with no implication about the state of the world at large.

  39. Jason L. said,

    July 4, 2013 @ 12:43 am

    Apart from be – and even do requires auxiliary do in questions ("Does he do it?") – I would have thought the only remaining exception is lexical have (as "Have you any steak?", "Have we the time?"), and even that's obsolescent in many English varieties.

    "Have you got any steak?" isn't obsolescent in any variety I'm aware of, although it is dispreferred to "Do you have any steak?" in (most of?) North America. "Have got" is weird in its own way, however.

  40. Jason L. said,

    July 4, 2013 @ 12:52 am

    Raempftl: Acutally, German does not make use of intonation as much as other languages. German students of English have to practice intonation. As apparently using German intonation makes them sound "boring" and bossy.

    My personal theory is that this is due to the abundance of modal particles in German which often have the same function as intonation in English and other languages.

    For all I know those modal particles are a rather weird feature of German.

    You mean stuff like "doch" and "mal"? When a "doch" or a "mal" gets tossed in and when it doesn't was something that was completely opaque to me when I was learning German.

    Perhaps related is that a native-English-speaking acquaintance reported that when she said "oder was?" in contexts like, (trans.) "would you like me to send an email or phone you, oder was?", she found she came across as flippant or sarcastic, which was confirmed by my native-German-speaking ex.

  41. Xmun said,

    July 4, 2013 @ 12:54 am

    @Pete and Rohan F
    Of course "Stands the Church clock at ten to three?" is not everyday modern colloquial English, but does that mean it isn't "grammatical" or "modern standard English"? I don't think so. Poetic diction is part of modern English (I don't think it's yet time to find some other term than "modern" for the language Rupert Brooke spoke and wrote).

    But, granted, I'm not quite sure of the exact force of "standard" or for that matter of "grammatical" in your comments.

  42. Jason L. said,

    July 4, 2013 @ 12:57 am

    Jason (not me): English is the only language in the European branch of the Indo-European family that lacks grammatical gender.

    Afrikaans?

  43. eva said,

    July 4, 2013 @ 5:01 am

    @Jason L. – I sympathize with your German learning woes. I'm a native speaker, but having us try to define the approriateness criteria for stuff like that was a favourite way of many lecturers I've known to make freshmen reevaluate what they actually know about their own language.
    Frankly, I still find it easier to explain the usage of "like" in col. English than where to place a "ja" in a German sentence.

  44. Eric P Smith said,

    July 4, 2013 @ 9:18 am

    @Rohan F:
    You assert that ‘be’ and ‘have’ are the only exceptions to the rule that questions in English require an auxiliary. How dare you? Need I say more?

  45. Matt_M said,

    July 4, 2013 @ 10:51 am

    @Eric: what definition of "auxiliary" are you working with that would mean that "dare" and "need" in these constructions are not auxiliaries?

  46. Raempftl said,

    July 4, 2013 @ 11:25 am

    @Jason L: Yes, that's exactly what I'm talking about. I think the only way to learn how to use them is by talkin to natives and immitating.

    But they also present a problem for German foreing language students. Since most other languages don't have modal particles, they are at a loss when trying to find an appropriate translation. But of course very often the appropriate translation would be a change in intonation.

  47. Yastreblyansky said,

    July 4, 2013 @ 12:22 pm

    There's a proof that all numbers are interesting: If there were uninteresting numbers, then there would be the smallest uninteresting number, and that would be pretty interesting, and then the smallest uninteresting even number, and so on. It should be obvious that all languages are weird.

  48. Eric P Smith said,

    July 4, 2013 @ 12:58 pm

    @Matt_M:
    I think there is more than one approach here. CGEL distinguishes auxiliary verbs from lexical verbs on the basis of their syntactic properties, and on that basis it classes dare and need (as in How dare you? and Need I say more? as auxiliaries. Oxford Dictionaries defines an auxiliary verb as "a verb used in forming the tenses, moods, and voices of other verbs" and on that basis it excludes dare and need. Wikipedia observes that the status of dare and need as auxiliaries is debatable.
    I have to admit I like the CGEL approach (though that admission weakens my previous comment!)

  49. Was ist eine seltsame Sprache? | Aus dem Hollerbusch said,

    July 4, 2013 @ 4:55 pm

    […] so selten auftreten. Vielleicht würde eine andere Auswahl das Ergebnis deutlich verändern, wie Mark Liberman im LanguageLog zu bedenken gibt. Und würde man nach Sprechern gewichten, würde das Ergebnis wiederum anders […]

  50. Ellen K. said,

    July 4, 2013 @ 6:27 pm

    Though the meaning is different (in American English), isn't "Have you got any steak?" grammatically the same as "Have you gotten any steak?"?

    And if I recall correctly, "have got" is what British use for the perfect tense, where we Americans would say "have gotten". ("Have got" in American English being equivalent, meaning wise, to plain "have".)

    In other words, it doesn't require "do" because it functions like a perfect tense.

  51. Piotr Gąsiorowski said,

    July 6, 2013 @ 12:12 pm

    Jason: English is the only language in the European branch of the Indo-European family that lacks grammatical gender.

    Afrikaans is of course another. There is no such thing as "the European branch" of IE, but if you mean IE languages located at least partly in Europe, Armenian might qualify as well. What is really exceptional about English and Afrikaans is not so much the fact they have no noun gender (there are lots of such languages, including Basque, Finnish, Estonian, and Hungarian), but that they nevertheless distinguish gender in personal pronouns. Strangely enough, that is an extreme typological oddity, shared only with a handful of other languages.

  52. Language Log » Weird languages? | FiloBlogia said,

    July 8, 2013 @ 6:31 am

    […] Diigo http://languagelog.ldc.upenn.edu/nll/?p=5006 This entry was posted in blogak, Euskara, tipologia. Bookmark the permalink. ← […]

  53. Azterketa estatistiko baten arabera, euskara munduko hizkuntza arrunt eta korrienteenetakoa da – Sustatu – Interneteko albistegia | FiloBlogia said,

    July 8, 2013 @ 6:53 am

    […] Weird languages? | Language Log This entry was posted in blogak, Euskara, tipologia. Bookmark the permalink. ← Language Log » Weird languages? […]

  54. Keith said,

    July 8, 2013 @ 6:55 pm

    JDS said,

    July 3, 2013 @ 11:47 pm

    @Carl, I think "fiddling while Rome burns" is closer to the Chinese saying than "eat, drink, and be merry."

    Or if you want to be more laconic and limit to four words, "Nero fiddles, Rome burns". This reminds me of Latin slogans like "ars longa, vita brevis".

    K.

  55. Buber's Basque Page : Regarding the weirdness (or non-weirdness) of Euskara said,

    July 8, 2013 @ 9:35 pm

    […] German and English were much weirder, in comparison with other languages, than Euskara. Well, another blog, this one from the Language Log at the University of Pennsylvania describes some issues with this […]

  56. Vanya said,

    July 9, 2013 @ 9:22 am

    What is really exceptional about English and Afrikaans is not so much the fact they have no noun gender , but that they nevertheless distinguish gender in personal pronouns

    Japanese also has no noun gender but does distinguish for personal pronouns, including the first person, which must be fairly exceptional.

  57. Azizi Buberek said,

    May 6, 2014 @ 6:29 pm

    A pretty weird language that I know is spoken by about only 43 people, called Tristian. Its alphabet consists of 5 letters: Σ, θ, Δ, β, Ε.

    Often times the words in Tristian are really long and complicated. Here's an example:

    English :How are you today?
    Tristian :ΕθΣΣΔβΕ ΣββΔθβΕΔΣΔθΣΣβ?

RSS feed for comments on this post