Tuesday, August 17, 2010

German — warum störrisch?!

I’ve been wondering about this for some time. I haven’t researched the question, in part because I’m not quite sure where to begin, and I’m hoping some of you can help me. In a nutshell, why has declension survived into Modern German, almost unchanged after more than a thousand years? Three genders and four (sometimes five) cases make learning the language more difficult for those unaccustomed to case systems. It’s a complaint I hear a lot about such languages (not only German, but Russian, Polish, Irish, Finnish, Greek, to name a few). It’s a little easier to learn the paradigms for “dead” languages (e.g., Ancient Greek, Latin, Sanskrit), because one generally tends only to read fixed texts, not to attempt to converse in these languages. One must only recognize inflected forms; one is not normally called on to summon them up during ad hoc conversations. It’s a question of passive versus active mastery.

I know there have been a few changes to German declensional paradigms over the centuries, sure, but the rest of the Germanic family has pretty much given up on them entirely, or nearly so. In Modern English, the only really conspicuous survival of the system is in the personal pronouns. Everywhere else — articles, numbers, adjectives, and of course, nouns — they’ve been swept almost completely into the dustbin of history. This is pretty universally true of the Swedish, Danish, Norwegian, Frisian, Dutch (though Dutch held on perhaps longest of all of them). Of the Germanic family*, it’s really only Modern (High) German that has stubbornly retained its original declensional system, and the system in use today is almost the same system in Middle High German, and it’s easily recognized even in Old High German texts more than a thousand years old. Why?

Just to give you an idea of the similarities, take a look at the following table representing the definite article in Old High German, Modern German, Old English, and Modern English. As you’ll see, the ancient forms are readily recognized, and it’s very clear that the Old English forms are close cognates to those in Old High German and even Modern German. This gives German speakers some advantage over English speakers when each attempts to learn Old English. Why have these distinctions survived in German, when they have been mostly abandoned by the rest of the family? (Note: I omitted the plural forms from the OHG paradigm because these have, in fact, changed a good deal, with distinct forms for each gender collapsing into a single plural form for each case in the modern language.)

 OHG masc.neut.fem.
nom.dërdazdiu
acc.dëndazdie
dat.dëmudëmudëru
gen.dësdësdëra
...
 German masc.neut.fem.pl.
nom.derdasdiedie
acc.dendasdiedie
dat.demdemderden
gen.desdesderder
...
 OE masc.neut.fem.pl.
nom.seþætseoþa
acc.þoneþætþaþa
dat.þæmþæmþæreþæm
gen.þæsþæsþæreþæra
...
 English sing.pl.
all casesthethe

* Modern Icelandic is the other notable exception. It’s still a highly inflected language, but in this case (no pun intended ;), an insular history explains how the grammar has been preserved, virtually unchanged, since the days of Snorri Sturluson. But German is its antithesis: spoken in the middle of a busy continent, by more than one hundred million (compared to less than half a million for Icelandic). German has also been widely used as a language of science, philology, literature, and even music. One should have expected substantial erosion. Why hasn’t this occurred? Any theories?

22 comments:

  1. I don't quite understand on what basis you're assuming that German *should* have lost its declensions by now.

    I know there are some theories that languages tend to go through certain stages in the process of change (agglutinating -> flexional, flexional -> isolating and so forth), but even given such a premise I don't see why all languages should undergo this process at the same pace. English lost most of its inflection during a period of language contact (Viking and Norman settlements) where change was accelerated; this has not been the case for German.

    I'm not sure I would agree that flexions are inherently more difficult for native speakers to learn and thus tend to be lost over time. As a fairly fluent non-native speaker of German I often fail to produce the wrong article for a specific noun (i.e., masculine instead of feminine), but I generally manage to produce an article in the correct case. It really does become automatic after a time; one learns that "mit" is always followed by "der" or "dem" but never "die", one uses fixed phrases so frequently ("auf der Strasse", "jeden Tag") that the system becomes fairly internalized.

    German actually has lost a fair amount of its inflection: noun endings have almost completely disappeared, and it's a well-known complaint among "Sprachnörgler" that the dative is replacing the genitive in spoken German.

    The inflectional situation looks rather different when one considers the German dialects, where there are sometimes quite considerable grammatical differences. The diglossic situation in Germany *may* actually have contributed to the preservation of inflections in standard German, since for speakers of many regions Hochdeutsch is first learned in school, and thus the standard language would tend to be somewhat conservative.

    ReplyDelete
  2. Swedish (and the other North Germanic languages I guess) does have gender: "den" and "det", but as articles these are used mostly before adjectives: "the blue boat" = "den blåa båten", but "the boat" = "båten". "Den båt" is correct Swedish but is used rarely, and only in the beginning of the clause, I think.

    ReplyDelete
  3. Came to think of it, sometimes you just say "blåa båten" too, but I think only rarely, as a colloquialism.

    ReplyDelete
  4. Ardamir, yes, all the Germanic languages except for English still have gender in one form or another. As far as grammatical gender is concerned, English nouns have become just about completely epicene. Once in a while, mainly in archaic or poetic expressions, you’ll come across some vestige of gender in English nouns (e.g., a sailing vessel is usually referred to by the feminine pronoun), but this is pretty rare (and getting rarer). There are certainly no signs of inflectional gender agreement among adjectives, pronouns (other than personal), numbers, etc., in English. It’s really only the personal pronouns that have retained any traces of their original genders and cases.

    ReplyDelete
  5. Brenda, thanks for your comments. Allow to respond and clarify:

    I don’t quite understand on what basis you’re assuming that German *should* have lost its declensions by now.

    Inflectional systems erode for many reasons, but they do tend to. As Samuel Johnson wrote, “tongues, like governments, have a natural tendency to degeneration” (offered somewhat tongue-in-cheek). We see the loss (read: simplification) of case systems over time in many language families. Proto-Indo-European is believed to have had eight distinct cases, of which only Sanskrit attests them all. By the time of Classical Latin, you’re down to six. Primitive Germanic, five. Old High German, four, sometimes five distinct forms; Old English, three, sometimes four; the modern Germanic languages, only two or three. Why should German (among widely spoken languages) be the sole exception to this process? (For a very readable discussion of the subject of language erosion, see Chapter 3 “The Forces of Destruction” in The Unfolding of Language by the aptly named Guy Deutcher.)

    We can debate whether languages should erode, but when the entire family but German has done, shouldn’t we wonder why it hasn’t?

    English lost most of its inflection during a period of language contact (Viking and Norman settlements) where change was accelerated; this has not been the case for German.

    So this is a bit of a theory. I think that English suffered the greatest erosion of its inflectional system after the periods of contact you mention, but that’s beside the point. So, in the case of the German being spoken on the continent, you’re suggesting that contact with speakers of other languages was much less? I wouldn’t have thought so. In fact, one could have expected the insular nature of the British isles to have preserved its grammatical features longer than languages spoken on the open continent. But you might still be right. I certainly don’t know enough about the history of German. I can say that the grammatical system of its French-speaking neighbors did erode over time. Old French shows much more complex gender and case systems than Modern French (even though “Old” French is much, much later than Old High German).

    I’m not sure I would agree that flexions are inherently more difficult for native speakers to learn and thus tend to be lost over time.

    No, that isn’t what I said. Native speakers don’t generally have any difficulty. It’s foreign speakers learning the language as adults who have difficulties. When inflectional endings are lost, it’s probably not because native speakers have any problems mastering them.

    The diglossic situation in Germany *may* actually have contributed to the preservation of inflections in standard German, since for speakers of many regions Hochdeutsch is first learned in school, and thus the standard language would tend to be somewhat conservative.

    “Book” English is more conservative than spoken English, too, but this hasn’t stopped the inexorable erosion of the spoken language. I still don’t feel like I have any clear idea why spoken German has clung to its case system so well for so long.

    ReplyDelete
  6. We see the loss (read: simplification) of case systems over time in many language families.
    That's interesting. Is there a theory as to whether PIE evolved from a simpler language or devolved from something even more complicated?

    ReplyDelete
  7. Good question, N.E. Brigand. Yes, there is a theory, but when you reason backwards from a period whose languages are only reconstructed (not actually recorded), you are already making big assumptions. The further back you go, the more you have to rely on guesswork and analogy. And there’s no possible way to verify your hypotheses.

    But yes, it’s generally believed that in the ancestor tongue of Proto-Indo-European, the so-called Ur-language, what became the case inflections in PIE (as well as the person and tense inflections of its verbs) began as separate words which gradually attached to the words they modified.

    In the words of John McWhorter: “Inflections almost always begin as free words; even in the cases where they do not, they arise through sound erosion and change. To wit, inflections always emerge in grammars where they did not exist before. This means that the first language, not having existed for a long enough time for inflection to appear through grammaticalization or other gradual processes, can be assumed not to have had inflections.” (The Power of Babel, New York: Perennial, 2001, p. 301). See also pp. 20–1, 47, 123–7.

    This theory suggests that inflections arose in the Ur-language through agglutination, then eroded very gradually through a long chain: PIE to IE to Proto-Germanic to West Germanic to Anglo-Frisian to Old English to Middle English to English! That is, PIE probably developed from something simpler, but its own daughter languages from something more complex. PIE was at the zenith of grammatical complexity.

    The process occurred in pretty much the entire Germanic family (not to mention another conspicuous example, the Romance), with notable exceptions in Icelandic and German. Icelandic is explainable by isolation. But German? That’s what I want help understanding (if anyone knows).

    Any by the way, now that Modern English words are pretty much identical in form regardless of use (the consqeuences of which are much more rigid word order), the process is likely to reverse. Eventually prepositions may reattach to their nouns, forming a new inflectional system. Of course, if this occurs, it will probably take another thousand years — or more.

    ReplyDelete
  8. May I be nit-picky? »German – warum so störrisch?!« would be stylistically better and more native-like ;-)

    As to case non-erosion I’m afraid I don’t have much to say except that I’m relatively sure someone somewhere out there has already thought about it. My Prof might have an idea, but it’s term break …
    Ah, sorry, not helping! I could only repeat that genitive case has been much reduced in spoken language, and note as an aside that case matching has already become ungrammatical … I’m not sure you can make a connection to this, though.

    ReplyDelete
  9. Actually, I almost used “warum so störrisch”; I guess I should have done. With the shorter title, I thought I was being more succinct and telegraphic in style (in short, more “bloggy”). But what works for English isn’t always right for another language!

    And I’m sure you’re right that others have thought about this. They must have done. If any of you ever come across a thoughtful discussion or theory, I would love to hear about it.

    ReplyDelete
  10. Just to add a couple cents to the discussion. It should be noted that PIE was the zenith of _Indo-European_ grammatical complexity, and then only when you are talking about morphology. Grammatical complexity is difficult to put a finger on. Regarding inflectional complexity, there are hundreds of languages with more complex inflection, of course depending on how it would be quantified, if possible. See languages like Kujamaat Joola or Ojibwe for example.

    As far as maintaining old features. German has retained much of a case system. Yes this is impressive, but some Norwegian dialects still preserve the old stem classes (i-stem, a-stem, n-stem, etc) in their definite inflections. Icelandic, which has been mentioned, still maintains the stem class distinction, as well has having the case distinctions of German, and Icelandic inflects the first three numbers for case and gender.

    Going back to English, it can certainly be argued that it has the most conservative consonant inventory of any, especially West, Germanic language. Almost no consonant changes have happened. This is quite the contrary for High German (as the 2nd sound shift has happened to varying degrees). Vowels in all of the languages are a different story, and can hardly be discussed without mentioning dialects. Norwegian and Frisian (and I believe Icelandic) are the only languages that I know of that still have the [j] in class one weak verbs (e.g. *tankjan, 'to think').

    The insular argument can go somewhat far in explaining retention of old features, but sometimes it cannot. The point is, every language has something about it that is more archaic than it's relatives. It is however, I agree, extraordinary that German has the remainders of the case system that it does, considering the levelling out of case systems in all of the languages surrounding it.

    -Luke (Olme on the Plaza)

    ReplyDelete
  11. Hi, Luke.

    It should be noted that PIE was the zenith of Indo-European grammatical complexity, and then only when you are talking about morphology. […] Regarding inflectional complexity, there are hundreds of languages with more complex inflection […]

    Yes, of course. An Indo-European context was implicit, but thanks for making the point more explicit. And you’re absolutely right about languages with more complex case systems. Finnish is a convenient example. Fifteen cases, I believe …

    The point is, every language has something about it that is more archaic than its relatives. It is however, I agree, extraordinary that German has the remainders of the case system that it does, considering the levelling out of case systems in all of the languages surrounding it.

    Agreed. You make some nice points in your comments, especially noting that “everybody got their something” (to borrow from Nikka Costa). It’s easy to sit back in the comfortable cradle of one’s mother tongue and point at the (perceived) oddities other languages. :)

    ReplyDelete
  12. Yes, there are some very cool and archaic things about English! Joseph Salmons, an excellent source on Indo-European and Germanic historical linguistics, has a book coming out in the next year or two about the history of the German Language, it will be very worthwhile as well as maneagable. You might find it interesting.

    Cool musings, I like the blog!

    ReplyDelete
  13. Thanks so much! :)

    And thanks too for alerting me to the forthcoming Salmons book. I’ll definitely watch for it.

    ReplyDelete
  14. I agree that diglossia is the key to the situation: Standard German has not lost its morphology because it was used in written form only, and thus effectively frozen, from Luther's day until about 1945. The language varieties which were actually spoken during most of that time have indeed lost most of their morphology: not only noun cases, but also the loss of the preterite in favor of the present perfect.

    The same is true in Italian. Because the transition to Standard Italian as a spoken language is less than a century old and far from complete, it looks like the language has not evolved at all since Dante's time. As for English, it has separate spoken and written registers, of course, but not anything comparable to diglossia.

    ReplyDelete
  15. So, are people using the full paradigm of the definite article in spoken German, or are they using a version of it that is showing signs of wear and tear? I would have though most German speakers still used the full paradigm, more or less unchanged, but of course, I don’t know this first-hand. Anyone?

    ReplyDelete
  16. There is an argument to be made that, in terms of linguistic evolution, the most complex languages are statistically likely to be the oldest. Language seems to have been born complicated and grown, very gradually, simpler over time. In the case of English, as linguist David Crystal has said, 'if one regards simplification and loss of complexity as a form of decay, then the English language is in a very advanced state of decomposition.'

    German has held out longer than its cousin-language, I suspect, in part because of the late emergence of "Germany" as a nation. Having had no central governmental authority, no "court" at which a prestige dialect was to be formed, its linguistic insecurity index must have been high. Therefore, its defense and preservation of older forms has also been aggressive. I recall a political issue a few years ago, when a simplified German spelling was proposed -- thousands turned out in protest, led by the novelist Günter Grass!

    ReplyDelete
  17. German has held out longer than its cousin-language, I suspect, in part because […]

    This makes sense, and I know next to nothing about Germany’s political history, so I really can’t comment one way or the other. Now one might have expected some erosion since 1945, in the wake of collective shame over Germany’s over-zealous nationalism, which came to a head in the preceding three or four decades. Even “philology” became tarnished with the pro-Aryan taint of its origins in the previous century and began to fall out of fashion. But again, I overstep the bounds of my very limited knowledge on this subject.

    I recall a political issue a few years ago, when a simplified German spelling was proposed — thousands turned out in protest, led by the novelist Günter Grass!

    That’s interesting. Here, you tend the see the opposite: protests in favor of English spelling reform. And these have been going on since at least the 17th century (at some point, I’ll be blogging about this).

    ReplyDelete
  18. So, are people using the full paradigm of the definite article in spoken German, or are they using a version of it that is showing signs of wear and tear? I would have though most German speakers still used the full paradigm, more or less unchanged, but of course, I don’t know this first-hand. Anyone?

    As a native German speaker, I'd say that the full paradigm is used in spoken German, except maybe the genitive which in many dialects is not used at all.

    ReplyDelete
  19. Thanks, D.A. I find it so interesting that German preserves these complexities when all around it, they have all but disappeared. That’s, well, störrisch! And I don’t mean that in a bad way at all. :)

    ReplyDelete
  20. My point was that during the late 19th and 20th centuries, both Germany and Italy underwent a transition from a situation in which most people spoke local and regional dialects (arguably separate languages), to one where most people speak the standard language (less so in the south of both countries) with regional accents. These standard languages, having been frozen centuries ago, were still morphologically rich, and morphological decay has simply not had time to set in.

    ReplyDelete
  21. So are you suggesting that we should indeed expect the morphology of Standard German to erode, but that it has not yet had enough time to do it? Is that the hypothesis? Hmm.

    It could be. But I also find it a little counterintuitive to suppose that those speaking regional dialects in which forms had worn down considerably would find it very easy (or desirable) to master a new, much more formal dialect. They would have needed a powerful motivation (the nationalism of the 18th–20th centuries might be a sufficient explanation). This would be almost analogous to teaching today’s English speakers Old English, or Latin for that matter. (The analogy isn’t perfect, but I think you will see what I am getting at.)

    I’m not saying you’re incorrect — I don’t know enough to venture that. But it seems somewhat counterintuitive. I also don’t think the situations in German and Italian are really that comparable. (I don’t mean historically; I mean linguistically.) Italian has nothing like the morphology of German. German has retained the case system of Old High German, almost entirely intact; but Italian has kept almost nothing of the case system of Latin.

    ReplyDelete
  22. Yannick Weiler7/24/2011 4:57 PM

    "As a native German speaker, I'd say that the full paradigm is used in spoken German, except maybe the genitive which in many dialects is not used at all"

    Being a native German speaker as well, I can, unfortunately, not solve the question (I can only confirm that the full case paradigm is, at large, still in use). Rather, I would like to focus on the peculiarity hinted at above. Indeed, the genitive case is substituted by the dative case in many dialects (This has even been satirized by B. Sick (note the ?coincidential? pun), "Der Dativ ist dem Genitiv sein Tod"). You can often hear it in the context of possessions: E. g. "Das ist dem Thomas sein Haus" instead of "Das ist Thomas' Haus". Now, I wonder why this is the case (maybe an answer to this question could help with the larger issue here)?
    My suggestion, limited to utterances like the example above, would be that the genitive case is, in spoken language, not as quickly recognizable as is the dative. Thus, getting the functions of the inflected words right would take longer, which would, in turn, prolong the response.
    By the way, great blog! I have been thrilled for quiet already.

    ReplyDelete