Why do synthetic languages become analytic?
Because analytic languages are simpler, and the languages (or rather they’re speakers) tend to simplify, even if it appears otherwise.
Which raises the question of whether all languages are moving from complexity (morphologically) to simplicity, from synthetic to analytic. Will we all end up speaking purely analytic languages devoid of cases, etc.? Surely there must be languages that become more synthetic over time? This also raises the question of why languages, at least in the Indo-European tradition, started in a state of great complexity in the first place if it would have been easier to be analytic. Perhaps long ago, before our records, indo-european languages reversed this process by becoming more synthetic over time and then started reversing the trend circa the birth of Latin, Greek, etc.
PS Does anyone object if I move this thread to the General Forum?
The simplification of languages over time is a fact, as well as that the analytical languages are simpler compared to analytic languages. But the route between these 2 is not straight and languages rarely move step by step in the analytic direction. Analytic-synthetic is also not the only way of dividing languages. Polish didn’t become more analytic when animate-inanimate distinction became non-productive, but it certainly simplified the language and it is going to simplify it further as the declension paradigms will probably merge or become more uniform.
Well, simpler for whom?
From what I’ve read, it would seem that languages tend to move from synthesizing to analyzing when many adult learners are speaking the language. To wit: English and Persian. Both were at one time fairly typical IE languages with case and gender on their nouns. Both have had long histories of adult learners. Persian as a court language and lingua franca throughout central Asia. English as the language of the conquerers imposed on Celtic-speaking natives and as the language of the conquered spoken by the Norse conquerers. (Or at least that’s what I drew from my reading of McWhorter’s Our Magnificent Bastard Tongue and Ostler’s The Last Lingua Franca.)
Now of the two classical languages we focus on here, which one had lots of non-native adult learners? Which one didn’t so much? Which one ditched case? Which one (mostly) hasn’t?
Native speakers obviously.
Grochojad: Explain what you mean by “simple” and “complex” here. It’s true that analytical languages are simpler morphologically than synthetic ones, but at the same time there’s an increase in complexity in other forms of the language. If you’re not learning case endings, you still have to learn the ways your language chooses to express this relationship instead (i.e., things like syntax and prepositions). Loss of inflection is not “decay”, nor is analytic equivalent to “easier to learn”. They’re just different ways of dividing up the work.
According to some theories of language change the synthetic/analytic shift is only part of a larger cycle which looks like this:
fusional (i.e. synthetic) → isolating (analytic)
isolating → agglutinative
agglutinative → fusional
This blog post has a good summary of the topic, as well as a few remarks on possible causes for such changes.
http://languages-of-the-world.blogspot.com/2011/10/more-on-word-order-morphological-types.html
As I understand it, it has to do with how independent the bits of meaning in a sentence are. If separate words are consistently used together so that they function as one word (including stress etc), you start to get agglutination (think of “gunna” for “going to” in spoken English). On the other hand, things like syntax and stress can mean that inflectional endings become less and less important for determining meaning, and so they start to disappear and be replaced with more moveable units like prepositions and auxiliary verbs.
Nonsense. Any given language is “simple” to its native speakers, no matter how convoluted the morphology and syntax may seem to outsiders.
Nonsense. Native speakers usually know they language well, but not perfectly, their mistakes are what largely constitutes language change. And they are most often mistaken where there is some kind of difficulty.
But it is possible to recognize a foreign language as simpler than one’s own.
Conflation of distinctions is a kind of simplification. If someone were to make no distinction between see and look and perhaps use only one of them, his English would be simpler from the perspective of someone whose English was therefore not for him the simplest kind.
Exactly. And therefore languages and chronolects of less democratic times are superior.
No, no, no and no.
This is a belief which classicists seem to be particularly prone to (partly because for a long time inflected languages were seen as superior to analytic languages). There are no serious linguists today who would support this position, though.
Languages are always changing. This is perfectly normal, and has very little to do with people “simplifying” because of “mistakes”. At any given time there is always variation in a language. After a certain amount of time it may happen that a construction which was previously used only by a minority becomes predominant. Changes aren’t always in the direction of simplification, either. For example, there are a number of formerly regular English verbs which have become irregular.
Please look at the link I posted above. These two articles also discuss some of the reasons for language change:
http://www.lsadc.org/info/ling-fields-change.cfm
http://www.uni-due.de/SHE/HE_NatureOfChange.htm#T_Nature%20of%20language%20change
edit: I’d also like to challenge the assumption that “simplification” – for example, regularization of forms, so that all verbs follow a particular pattern – necessarily equates to loss in meaning.
What meaning is lost when I say that I write “with a pen”, using a preposition instead of an ablative of instrument?? It could be argued, in fact, that the preposition is actually more precise than the case marker because “with” has a more limited range of meanings (accompaniment, instrument) than the ablative (agent, manner, time, accompaniment, instrument, location etc).
Appealing to the authority of »serious linguists« doesn’t prove anything.
No, but actually reading some of their arguments might.
All I’m saying is that there’s another way to look at this, one which is supported by what scientists today know about how language works. I’ve given some reasons why I disagree with some of the specific claims you are making. I’ve offered counterexamples and alternative explanations. I would be interested in knowing on what basis you feel the argments are insufficient.
Ha. I deserved the nonsense right back at me. Here is my understanding of the linguistic score of moving from synthetic to analytic.
Over time, adult learners have stripped complex morphology out of English. Case, gender, noun/adjective agreement in English all went away due to many years of adult learners—Norse and Celt—certainly no later than the 13th century. Given the number of Slavic-language speakers learning English as adults in the area I live in, the article could find itself next. (I live in Chicago which has a large and long-standing Polish immigrant community). 
Adult learners (probably) did the same to Latin. Case, gone. Remember, Latin was not spoken across the Italian peninsula in 753 BC so someone had to learn it to spread it across all of western Europe—that would be the adult learners.
This isn’t to say that these sorts of pressures are strictly destructive, but since we’re talking about morphology they have been in the case of case. I don’t know enough about the source of all the verbal reanalyses going on in the Romance verb, e.g. future tense moving from amābō to amare habeo to (Spanish) amaré, so I won’t comment.
Native speakers are responsible for different sorts of changes. And I do not deny that these changes exist. If I were to deny their existence, I’d be bounced form my grad program. Some of changes driven by native speakers are phonetic in nature—Northern Cities Shift (21st century America). Some are grammatical in nature—turning “going to” from a verb of motion into “gonna” a future marker (15th to 17th century England). Some may even be morphological in nature—the decline of the English subjunctive. But native speakers tend not to make wholesale remodels to the language, as in the previous paragraphs.
Language change is devilishly complicated business, as far as I can tell. But if I were to simplify the way these changes come about, I’d lay it out this way: Natives drive changes that while incremental, add up. See Greek (or that’s what I understand). Adult learners drive wholesale changes to languages. See Latin and English.
One of my reasons for choosing a linguistics degree over classics. And I do love the classics.
I suspect that these views are the result of an ideology.
And the language of the linguist is not the poet’s.
Ideology?? I beg your pardon?
What does ideology have to do with it? Linguistics proposes a model for how language works, yes. “Belief” has little to do with it, however.
I don’t think that linguistics and poetics are incompatible. We may have to agree to disagree on that, however, as you seem to be pretty convinced that the scientific investigation of language can bring no insight into how language works.
I do wonder a little bit why you ask a question about why languages change but are apparently not willing to even consider or try to understand the responses you get. If what I’ve written isn’t clear, I’m happy to explain, but simply dismissing or ignoring it doesn’t take the discussion anywhere. If you feel like you already have refuted my arguments and I’ve missed it, I apologize. You’ll obviously have to explain yourself more because I can’t find any substantiation for the belief that synthetic languages are automatically capable of making more nuanced distinctions than analytic ones.
I was going to post about how the English system of helping verbs is in some respects actually a much more elegant and efficient way to express tense and aspect distinctions than Latin or Greek with their endings. I don’t think I’ll bother, however, since I suspect I would be wasting my time.
Linguistic models are based on assumptions that are believed to be correct, but might not be,—assumptions which escape testing, especially if commonly held, or are untestable. In hindsight, scientists/academics, believing themselves scientifically objective, may be shown to have uncritically adopted cultural, historical, political, social (ideological) perspectives within their research paradigm.
A person might accuse another of ideological assumptions to deflect attention from their own.
Pendent paradeigmata linguistica de praesuppositionibus quae ab eis tenentibus rectae creduntur forsit perperám, nec spectantur quidem (nec sic fieri possunt, praesertim si communes). In retrospecto, possibile est ut ita ostendatur, scilicet naturales/academicos, objectivos se credentes, in exquaerendo intra paradeigma proprium, sententiis ideologicis (historicis, politicis, socialibus cultûsque) modo subobscuro utos esse.
Aliqui alium de praesuppositionum ideologicarum usu accuset ut usum suum dissimulet.
Spiphany,
if I somehow offended you, then I sincerely beg your pardon, although I am surprised that the utterance of a suspicion should be offensive. I’ll look at your articles when I have time. Is there some locus classicus of your view?
Adriane,
Before the last sentence I thought you were on my side. For clarification: I don’t consider myself to be objective (I would have thought the Heidegger-quotation had made that clear).