|
 |
Is analyticity normal? | ||
Tags: Linguistics | Grammar | ||
Share with:
| ||
 Language Learning Forum : Philological Room |  | |
| Stolan Senior Member United States Joined 4017 days ago 274 posts - 368 votes  Speaks: English* Studies: Thai, Lowland Scots Studies: Arabic (classical), Cantonese | Message 1 of 11 17 August 2014 at 6:03pm | IP Logged | |||||||||||
|
http://www.academia.edu/7756170/Is_Analyticity_Normal_Implic ations_of_Niger-Congo_and_Sino-
Tibetan_for_Typology_and_Diachronic_Theory I am sharing this because it provokes thought, now many of you know I have some disagreements with John McWhorter but I also believe he has very well thought out insights into languages and is invaluable for linguistics with the common man (despite some misconceptions such as tones not being learnable past 18) Here he brings up an interesting point, analyticity in the Asian sense is quite exceptional in a way. I have disagreements with some things he wrote in this paper and some of the things he said just raise more questions for me. I won't just list my thoughts yet before everyone else here has put forth their ideas, so I ask a few things to get this convo started: Is analyticity normal at all? Is analyticity normal but only in a certain manner? (Compare the mild analyticity of some modern languages and late Egyptian vs the near absolute isolating nature of Old Chinese) IE: Is analytical grammar normal but not an isolating structure behind it? What do you agree or disagree with in McWhorters paper I posted above? Edited by Stolan on 17 August 2014 at 6:05pm 1 person has voted this message useful | ||||||||||||
| tarvos Super Polyglot Winner TAC 2012 Senior Member China likeapolyglot.wordpr Joined 4692 days ago 5310 posts - 9399 votes Speaks: Dutch*, English, Swedish, French, Russian, German, Italian, Norwegian, Mandarin, Romanian, Afrikaans Studies: Greek, Modern Hebrew, Spanish, Portuguese, Czech, Korean, Esperanto, Finnish | Message 2 of 11 17 August 2014 at 10:52pm | IP Logged | |||||||||||
|
Normal according to what standards? If you're asking about normalcy, give me a definition
of normal to work with first and then we can talk. Second: what is the value of knowing that analyticity is normal/abnormal (given a certain reference framework)? Give me that first before you can enter a discussion on the linguistical side of things, and even then I'm inclined to leave it to the linguists here (I am most certainly not a linguist). 1 person has voted this message useful | ||||||||||||
 emk Diglot Moderator United States Joined 5517 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map | Message 3 of 11 17 August 2014 at 11:25pm | IP Logged | |||||||||||
It's an interesting paper, but it contains a huge number of implicit assumptions and potential issues. This section caught my eye:
Here, McWhorter seems somewhat confused about Egyptian, and gotten some of the details from Hodge 1970 reversed. Compare what Hodge writes about prefixes and suffixes in the original paper with McWhorter's summary above: 
Of course, this might be just an editing error. But it should raise a pretty major red flag to anybody who has spent any time with Egyptian. The classic forms of Egyptian almost never use grammatical prefixes (except for the causative s-). The second problem, however, is more troubling: McWhorter is trying to make arguments about typology using a paper from 1970. At that point in history, however, Egyptology was pretty much isolated from the rest of linguistics, and used non-standard terminology and grammatical analysis. Today, Egyptologists and linguists actually speak to each other, and it's much easier to find reliable overviews of Egyptian grammar which use standard linguistics terminology. (Loprieno's Ancient Egyptian: A Linguistic Introduction is an excellent starting point.) If you're going to make typological arguments about the history of Egyptian, you really need newer references, because they're much more accessible to mainstream linguists, and because the traditional analysis of Egyptian verbs is rather non-standard. This is related to one of my major complaints about this sort of typological paper: It's far too easy to misunderstand the evidence provided by any single language, and to cherry-pick specialist papers that support your larger hypothesis. If you want to argue about the historical development of Egyptian, you really need to get down into the details and get a real feel for the language at several points in time. I don't have that feel for Egyptian yet, but even my limited knowledge is enough to make me skeptical about this paper's discussion of Egyptian. And this, in turn, brings me to a more general question: What's so special about "radically analytic" languages, anyway? Middle Egyptian is a highly regular synthetic language. Sure, verbs are inflected, but they're incredibly regular compared to typical Indo-European languages: You have a stem (which can take a couple of forms, mostly involving whether the last consonant gets dropped), an optional tense/aspect marker, and possibly a clitic pronoun. There are a tiny handful of irregular verbs. The exact same clitic pronouns are used with nouns (to mark possession) and with prepositions (when the object is a pronoun). For example, sDm.n=f "he heard" consists of the root sDm "to hear", the aspect marker n, and the clitic pronoun =f "he". You'll notice that the standard transliterations of Egyptian actually separate each of these morphemes with punctuation. In other words, this is not some ugly, fusional language with dozens of pages of conjugation and declension tables and hundreds of irregular verbs. As far as adult learner is concerned, learning to inflect Middle Egyptian is a lot like learning to inflect Esperanto: You can explain the whole system on a few sheets of paper. (This doesn't mean that Egyptian grammar is "easy", any more than English grammar is. The complications lie elsewhere.) So even though Egyptian is not "radically analytic", it certainly doesn't sound like McWhorter's description of the "typical state of human language":
So I don't understand how highly-regular synthetic languages are supposed to fit into McWhorter's overall theme. Theoretically, they should be almost as easy as "radically analytic" languages. And finally, the paper's main argument feels a bit circular:
Basically, he proposes that "radically analytic" languages can only arise through widespread adult acquisition (or perhaps via areal transfer). But wherever we have no evidence of this actually occurring, it's possible to assume that it must have happened thousands of years ago, somewhere in the unrecorded past. So, yeah: I don't know quite what to make about this paper. I don't find the part about Egyptian convincing, a huge fraction of the supporting citations go to other papers by McWhorter, there's an implication that synthetic languages are normally fusional and irregular, and for many of the interesting languages, the interesting events are hypothesized to happened in the "distant past", with no written records. Presumably I've missed a whole bunch of supporting papers, and I'm not really qualified to comment on the ongoing debate here, but it feels like McWhorter is working at a very high level of abstraction. I'd find the argument a lot more convincing if it were presented at book length, with detailed and convincing analyses of the languages involved. Edited by emk on 17 August 2014 at 11:31pm 3 persons have voted this message useful | ||||||||||||
| Stolan Senior Member United States Joined 4017 days ago 274 posts - 368 votes Speaks: English* Studies: Thai, Lowland Scots Studies: Arabic (classical), Cantonese | Message 4 of 11 17 August 2014 at 11:43pm | IP Logged | |||||||||||
I will rephrase these: 1. Can analyticity something that occurs in a language or can it only be brought by some simplification from outside? 2. If analyticity can occur in a non-interrupted language, would you make a distinction between a language that happens uses more auxiliaries, word order techniques, and demonstratives in most areas, than the likes of a typical southeast Asian language which is not just analytical but "telegraphic". 3. The value is gaining further understanding of why languages are the way they are.
Yes, He seems to also equate inflection with many complexities due to sound changes, but its not just that: Georgian is agglutinating yet has extremely irregular verbs while hittite is fusional but esperanto like. This may seem like just trivia but this shows that its not just matter of inflection leading to complexity because of some sound changes, there are things at work he doesn't even mention or admit he is not knowledgable of. But I would like to remind you the syntax also matters, these Southeast Asian languages are quite transparent. McWhorter's explanation does at least satifisfy some part of me who wants to know why they are so free of grammatical exceptions and categories. Mcwhorther refers to them as "telegraphic" because of their bare-bone structure. Do you at least notice this? This is where they may part with Ancient Egyptian . The main context that his works are set in is his idea of linguistic complexity. His books deal with complexity and he has written about complexity. Edited by Stolan on 18 August 2014 at 12:29am 1 person has voted this message useful | ||||||||||||
| emk Diglot Moderator United States Joined 5517 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map | Message 5 of 11 18 August 2014 at 3:44am | IP Logged | |||||||||||
I don't know enough to have an opinion. :-) But honestly, I'm not really convinced that analyticity is especially desirable, even from the perspective of adult learners. Given a choice, I actually think there's a lot to be said for highly-regular synthetic features, such as: - Esperanto's inflection. - Egyptian verb and noun inflections. - French clitic pronouns. The unifying principal in all these cases is that certain positions in a sentence are special, and they may only contain one of a strictly limited number of morphemes. For example, every inflected French verb has a number of special "slots" directly in front of it, as described on Wikipedia: Nom - Neg - Obj - COD - COI - Loc - De - VERB An adult learner with enough exposure can deeply internalize this system. I even love the way these clitic pronouns feel in my brain (I really miss en when I speaking English). And because the positions are so rigid, these pronouns can be very short and slurred, and they still remain comprehensible. It's both learnable and efficient! Now, in French, clitic pronouns are traditionally written as being separated from the verb with spaces. In Egyptian, clitic pronouns are actually even simpler, and they're normally transcribed as being attached to the verb via special punctuation, as in sDm.n=f. But because of the rigid ordering restrictions, you can't really call these independent words, whether or not they're written with spaces. For any given language, how you analyze these morphemes comes down, in part, to convention. So if anyone wants to argue that certain features are especially good or bad for adult learners, I really want to see that analysis carried out with a lot of nuance. I mean, you can spend multiple pages debating whether or not the French word ne is a clitic pronoun—certainly everything to the right of that slot is firmly attached to the verb, but ne can separate from the verb in a way that the COD and COI can't, as when je ne le lui dis pas "I don't say it to him" becomes ne pas le lui dire "to not say it to him." Personally, I'm a fan of careful analysis and corpus linguistics. Large-scale typology always seems a bit shaky to me, because if specialists are still slightly puzzled by the finer points of French clitic pronouns, trying to make huge-scale generalizations about how all human languages behave over thousands and tens of thousands of years seems like a massive undertaking. And of course, we don't have more than 5,000 years of data for any language, and usually a lot less.
I don't know anything about Asian languages. But one hypothesis that needs to be considered is that some language families just have unique and weird features. I mean, several Afroasiatic language share a system of consonotal roots which are inflected using multi-part vowel transfixes. I mean, it's a very cool system, but how weird can you get? So if somebody tells me, "There's a group of languages over here which all share this really interesting feature," I'm not going to be terribly surprised. So I'm hard to sell on any particular theory. :-) I want both careful analysis of specific languages, and evidence of a pattern that extends across language families and regions. My other major reservation with a lot of McWhorter's work is that he seems to divide the world into two categories: 1. Child language learners. 2. Adult language learners. But I live in a multilingual household and I know a bunch of other multilingual families. And based on my personal experience, I tend to divide the world into four categories: 1a. Heritage learners. (Kids learning a non-community language.) 1b. Native childhood learners. 2a. Adults with limited exposure and low integrative motivation. 2b. Adults with massive exposure and high integrative motivation. I've seen all four categories up close, in my own household and among our friends. In my personal experience, groups (1a) and (2a) will often fail to acquire a language beyond a basic level—they'll learn whatever they need to get by, and nothing more. Groups (1b) and (2b), however, will reach very high levels. In the long run, group (1b) will come out ahead, thanks to accent and some possible advantages with the nastier kind of grammar. Or to put it another way, age matters a bit, but massive exposure and high integrative motivation matter a lot more. An adult who identifies deeply with a new culture will do far better than a child who can't see a reason to learn a language that they never hear from anybody but a few family members. But for the sake of argument, let's imagine I buy into McWhorter's raw data. In this case, I'd say that radical analytic languages might be the result of learners in group (2a)—and possibly (1a), but I don't know what evidence might exist in that case. 2 persons have voted this message useful | ||||||||||||
| Stolan Senior Member United States Joined 4017 days ago 274 posts - 368 votes Speaks: English* Studies: Thai, Lowland Scots Studies: Arabic (classical), Cantonese | Message 6 of 11 18 August 2014 at 4:22am | IP Logged | |||||||||||
I hate to be off topic but I tried to send a private message to you so we could continue a discussion about this without risking bringing this thread off topic but it says you've reached the maximum amount allowed to be received. Is that normal? 1 person has voted this message useful | ||||||||||||
| tarvos Super Polyglot Winner TAC 2012 Senior Member China likeapolyglot.wordpr Joined 4692 days ago 5310 posts - 9399 votes Speaks: Dutch*, English, Swedish, French, Russian, German, Italian, Norwegian, Mandarin, Romanian, Afrikaans Studies: Greek, Modern Hebrew, Spanish, Portuguese, Czech, Korean, Esperanto, Finnish | Message 7 of 11 18 August 2014 at 10:24am | IP Logged | |||||||||||
|
A few things I would like to remark on in general here (I don't want to go into a 20
page dissertation on linguistics because I am not one of them): 1. The system emk describes using the consonantal roots and prefixes regarding Ancient Egyptian is a crucial feature of most Semitic languages (at least the ones I have studied). I have only very limited Egyptian (though I have a textbook) but I have studied Modern Hebrew in more detail and I agree with emk that its grammar is very logical and elegant, much more so than most languages I have encountered. There is much to be said for a Hebrew-esque way of structuring your language. 2. Is it important to really know why analyticity occurs? It's a feature of some languages, and knowing why satisfies mostly our historical curiosity, but applying it (even if not in the absurd) to language learning as a measure of complexity or difficulty seems to me to be an arbitrary way of defining complexity; what about all that context that you need to know to decipher certain texts? I find Asian languages to be very context-dependent. There is much to be said for making that explicit instead. Difficulty is not a matter of intrinsic complexity as much as it also deals with motivation and mostly the discrepancy with the learner's natural way of thinking (influenced by their base language, which in my case is Dutch and to an extent English, so I certainly have an IE bias here). 3. Further understanding is a nice bonus, but I doubt it allows you to draw any conclusions worth remembering concerning language learning. It's a nice theoretical discussion, but even notwithstanding any academical flaws that emk so generously pointed out, I fail to see even how something so seriously grounded in theory could bridge the gap to something I generally consider an applied skill (language learning and use to me is something you apply and use). 1 person has voted this message useful | ||||||||||||
 Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6688 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 8 of 11 18 August 2014 at 12:39pm | IP Logged | |||||||||||
|
The conventional divisíon of languages into polysynthetic, agglutinating, flexional and analytic is misleading because many languages have elements that point into several directions. For instance German and Russian and Greek and many other supposedly flexional languages use prefixes in a way that clearly fulfills the criteria for agglutinative structuring. Their word endings are (mostly) flexive, but their word beginnings are definitely agglutinating.
If you want to see a circular movement that passes like a clockwork through four stages with uniform grammatical typologies you will be disappointed - and this will also make it difficult to find purely analytical languages. When a language is getting close to this stage there will already be processes underway that transform isolated elements into affixes or endings. Like the unstressed pronouns in French, which only are one step from becoming agglutinative elements within the word boundaries for verbs. Or take Icelandic, where the nouns can have clearly flexional person/number/case endings followed by postclitic articles (derived from demonstrative pronouns), which in one and only one case change the form for the preceding ending (*-um+um -> -unum). Within the Germanic languages we have the highly inflected High German in the middle and a host of much less inflected languages like Dutch, Low German and Danish around it. It is difficult to see any simple explanation for this - Danish has never been a creole, iinvaded by wawe after wawe of second language learners. But Germany has constantly had foreigners running around on its territory. Even more perplexing: English lost the bulk of its inflection during the dark years where the upperclass spoke French. But in Latvia a similar situation with a German-speaking nobility and clergy left the peasants with a heavily inflected Baltic language, now known as Latvian. Where is the logic? Edited by Iversen on 18 August 2014 at 1:10pm 1 person has voted this message useful |
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
