Romanian: definite article

Discussion in 'Etymology, History of languages, and Linguistics (EHL)' started by Encolpius, May 26, 2010.

  1. Encolpius

    Encolpius Senior Member

    Praha (Prague)
    magyar (Hungarian)
    Hello, is there anybody here who knows concrete info about the interesting thing why the definite article is attached to the end of the noun? Was there any Old Slavic influence, just because this is used in Bulgarian, too?? How did it all happen? Any connection to Nordic languages? Thanks a lot.
  2. CapnPrep Senior Member

    You can find a long discussion and a few external links in this thread:
    Balkan Sprachbund

    I don't think there can be any connection to Nordic languages.
  3. Ben Jamin Senior Member

    Old Slavic (Slavonic) had no articles, either before or after the nouns, so the Bulgarian postponed article is a much newer feature.
  4. Alxmrphi Senior Member

    Reykjavík, Ísland
    UK English
    In the Balkan Sprachbund, like CapnPrep linked to, the language area has indeed developed interestingly with regard to languages sharing grammar. It is thought that the trait of Romanian having the definite article after the noun is because is it is based on the languages that surround it.

    So, a normal Romance language has its definite articles before the nou (el ombre / la donna), and the other languages in the sprachbund have articles after it, like Bulgarian zena-ta (woman-the) or Albanian mik-u (friend-the) so this mixture creates the environment where there is a lot of bi(multi)lingualism in which languages aren't always learnt completely and certain traits can pass over and you can end up with the effect of the Romance language Romanian having its definite article after the noun om-ul (man-the).

    Good point, and it ties in well with the topic of "what happens in a sprachbund", because Slavic languages generally don't have articles, but Bulgarian's adoption of the definite article can be seen as an influence of the surrounding influences of languages.
  5. Ben Jamin Senior Member

    So the question remains: "what language was the source of postponed articles?". Was it one of the existing languages, or a substrate language, or an external one? A connection to Nordic languages is not impossible, although unlikely. There were many Nordic mercenaries in the Imperial Byzantine service around the 10th century. Their language could theoretically form a nucleus of a "fashion" that spread to some Greek speakers (I'm not sure when Greek developed articles) and then spread to the neighbouring peoples. There are examples of such spreading. The guttural 'r' spread from a little area and few speakers in France (XVI century?) to neighbouring countries, reaching the west coast of Norway. The spreading is still active now.
    Any opinions on that?
  6. Alxmrphi Senior Member

    Reykjavík, Ísland
    UK English
    Yeah, I didn't know about that, but it was mentioned in a lecture I watched about 4 months ago, I was very surprised by how far it had spread, even into culture's / languages completely unrelated with French.

    As for the source language, it's hard to tell, I would (without any evidence) guess it'd be the closest related language that has the features that the language has adopted, or maybe in a situation where there is a neighbouring (but not directly neighbouring) dominant power where people learned that language, that these things are absorbed through there.

    The way languages mix when speakers are brought up to speak two languages imperfectly, or even with marriage where a child might hear native X-ese from the mother, and impaired X-ese from the father, resulting in a mix of the two, inevitably with the more difficult aspects of X-ese reduced, particularly if you can expect more paternal care. This is what linguists have proposed happened to English in the Danelaw after the Vikings invaded and intermarriage was common.

    Interesting discussion, but I'm not sure how close we are, with evidence to be able to tell (or even hypothesize) about which specific languages are likely to be the source of new language developments in a sprachbund.
  7. CapnPrep Senior Member

    A long time before any fashionable Nordic mercenaries turned up. And there is no trace of an enclitic definite article in Greek.

    My opinion is that "unlikely" is an understatement.
  8. Orlin Banned

    As far as I know, in many languages (including Bulgarian) some demonstrative pronouns evolved into definite articles, and the definite article in Bulgarian is postpositioned because the demonstratives that became articles later had a tendency to be placed after the noun. The same probably happened in Romanian.
    Last edited: May 28, 2010
  9. bibax Senior Member

    Czech (Prague)
    It was quite common in Latin to place the demonstrative pronoun ILLE after the noun, especially in a reference to a known person, for example: Socrates ille, imperator ille, homo ille. Romanian could easily adopt this Latin construction (homo ille -> omul).
    Last edited: May 28, 2010
  10. Alxmrphi Senior Member

    Reykjavík, Ísland
    UK English
    The keyword is could...

    Counting the similarities between Romanian and the other languages in its family, noticing similar developments and general trends, the fact of Romanian coming out of Proto-Romance with the opposite system of article placement compared to its sister languages is a lot less likely (in my opinion) than the influence of other languages with that same characteristic that are also present in the Sprachbund.

    My view could be easily influenced though, when I was first introduced to this, it was by watching a lecture by a linguist describing characteristics of sprachbunds and using the case of the definite article placement in Romanian as one of the core examples, his description is generally well established and known in the linguistic world, and also where I first learnt of it so I can't deny this might be why it makes so much sense to me.
  11. bibax Senior Member

    Czech (Prague)
    In Common Slavic (and Old Slavonic as well) the anaphoric pronoun j'/ja/je was used the same way like Bulgarian uses the definite article.

    Bulgarian: ženata (woman-the), but mladata žena (young-the woman)
    Common Slavic: mladaja žena (young-that woman)

    In Czech it is not unnatural (though somewhat bookish) to place the demonstrative pronoun ten/ta/to (as an anaphoric reference) after the noun or after the attributive adjective. The demonstrative ten/ta/to used anaphorically is unstressed, so it is quite natural to place it after the first stressed word. For example:

    Žena ta byla ... (The/that woman was ...)
    Mladá ta žena byla ... (The/that young woman was ...)

    Muž ten mé dcery srdci učaroval. = This man hath bewitched the bosom of my child. (A Midsummer Night's Dream)

    Bulgarian uses the definite article similarly like Czech uses the unstressed anaphoric ten/ta/to. Romanian uses the definite article differently, always postponed after the nouns.
  12. Ben Jamin Senior Member

    But can you say that j'/ja/je were articles?
    In Polish it is similar as in Chech. In a bookish language one uses the anaphoric demonstrative pronoun (may be under the influence from Latin), but this is not an definite article.
  13. robbie_SWE

    robbie_SWE Senior Member

    Trilingual: Swedish, Romanian & English

    I personally believe that the Romanian article placement is more of a "natural" development from Latin than its sister languages, where the article is separate and placed before the noun. My reasons for believing this are:
    1. I seem to recall that Latin (Classical and Vulgar) had well developed and complicated case systems. These case systems (nominative, vocative, accusative, genitive, dative and ablative) have to some extent, similarities with the Romanian postponed article. I'll give some examples.
    • Ancilla vinum portat. Sclava cară vinul. The slave woman carries the wine.
    • Dominus servi eum verberaverat. Stăpânul sclavului l-a bătut. The master of the slave had beaten him.
    • Mercator feminae stolam tradit. Comerciantul îi dă femeii o stolă.The merchant hands over the stola to the woman.
    2. Romanian and Italian are the only two Romance languages that don't add the letter "s" to form the plural (lup/lupi or lupo/lupi). I'm more inclined to believe that Italian under the influence of other Latin languages started to place their article before the noun, while Romanian due to isolation adapted the traditional Latin case system. As it was said before in another thread, the ingredients were already there; they just needed stirring!
    As I pointed out before in other threads, there seems to be a conception stating that Romanian was susceptible to all foreign elements without influencing other languages and cultures in return. People seem to forget that the Romanian influence on its neighbouring languages could have been just as contributing.
    :) Robbie
  14. Alxmrphi Senior Member

    Reykjavík, Ísland
    UK English
    Hi robbie,

    Thanks for your post, very interesting.
    I just wanted to quote from the linguistics course, here.
    It seems to be generally accepted among people that it is the case of language influence, due to the reasons listed in C, D & E.

    But your hypothesis seems interesting! *goes to research*...

    In this article, it states the University of Vienna's Linguistics department are looking into proving that Old Albanian was massive source of influence on Romanian and says:
    Last edited: May 29, 2010
  15. sokol

    sokol Senior Member

    Vienna, Austria; raised in Upper Austria
    Austrian (as opposed to Australian)
    Well, that's actually not really relevant to the question, even though it seems as if it could. :)

    The point is that the Slavic (Common Slavic!) determining suffix - the one mentioned by bibax, "j'/ja/je" - is indeed functioning like article in other languages (it marks the difference between "determined" and "not determined").
    However, only in Balkanbund Slavic languages (and of course other Balkanbund languages :)) developed a regular article which happens to be postponed.
    The development of this article is independent from Common Slavic, and happened much, much later; also, the development of this Balkanbund article definitely is not Common Slavic.

    Of course bibax has a point mentioning this Common Slavic suffix at all: after all, it is quite likely that in Balkanbund Slavic (that is Bulgarian and Macedonian) it supported the development of an article - it made it easier for those Slavs to develop the article and adapt to it, as two structural concepts met here which were similar in expressiveness.

    On the other hand it is highly doubtful whether the Balkanbund article developed from the Slavic suffix; there exist several theories about how and where and why this article emerged - but to my knowledge (which, by the way, is far from being up to date in this case ;)) there is no agreement yet among linguists which of this theories is more likely.

    Also, this Common Slavic suffix did not lead to the development of an article in other than Balkanbund Slavic languages. :)
    (There exists article in e. g. Slovene colloquial speech, but in this case it is quite clear that this was due to German influence; and the same seems to be the case for Czech: neither is in any way connected to the Balkanbund article, and thus also, of course, not to the Romanian article.)

    Further I'd like to point out that, to my knowledge (again, quite possibly not up to date ;)), Romanian also is not agreed upon by linguists as source of this article, as - to my knowledge, to annoy you yet another time with repeating it :D - there is no agreement about its origin. ;-)

    Robbie, please don't take it as criticism. :)
    I do know that Romanian, quite unfairly, often is described as only having been a "recipient" of influence. But to turn that upside down and suggest that the reverse is true - which you didn't, but you know as well as I that some extreme nationalists do :) - is just as wrong.

    The same claim - i. e., that all other languages were influenced by their language rather than the other way round - also can be heard oftentimes from Albanian nationalists, about Albanian being the "most original" if not oldest language at all.

    We've had quite some threads recently turning into this direction (that is, "language A is the most original one" etc.); I'd just thought it is necessary to point out that this is not where we want to go in this thread, right? ;)
  16. robbie_SWE

    robbie_SWE Senior Member

    Trilingual: Swedish, Romanian & English
    I completely agree with you Sokol. :D I do not suggest the other way around being more accurate. I only want to remind people that influence is most often mutual, thus making it difficult for us to decipher what came first.

    :) robbie
  17. phosphore Senior Member

    Plural -s- was originally a case ending just like -i- and it became the mark of the plural only with the loss of declension, so there is nothing there to suggest how the articles developped in the languages derived from Latin.

    Now your hypothesis seems interesting but I fail to see how we could connect cases with articles. I would write more but Eurovision is starting. :D
  18. Maroseika Moderator

    Well, obviously a kind of an article. There were 2 kinds of ajectives in Old-Slavonic: nominal (молода, нова, новъ) and pronominal (молодая, новая, новый), opposed as indefinite/definite.
  19. phosphore Senior Member

    I find it funny to write about features of Albanian in ancient times while Albanian has been attested for some six hundred years now.

    Does anybody know when were the articles first time attested in Bulgarian or Macedonian?
  20. DenisBiH

    DenisBiH Senior Member

    I haven't seen that somebody mentioned this, but Slavic "j'/ja/je" is not the only Slavic example of pronoun being attached to a word as a suffix. We recently had a discussion on the Other Slavic Languages forum regarding BCS words such as danas (today), večeras (tonight), ljetos (this/last summer) etc. (phosphore might remember) which show the demonstrative pronoun *sь being attached to the end of the word (and remaining there fossilized after it otherwise fell out of use). I've seen some Slovene and Czech examples of letos so it's not BCS only and might well be Common Slavic. While it did not function there as an article, it does in my opinion further support the notion that Slavic had a feature/tendency which could have contributed greatly to the development of articles in Bulgarian and Macedonian.
    Last edited: May 30, 2010
  21. Orlin Banned

    Koliko se sećam, u 18. ili 19. veku šta se bugarskog tiče (i mislim da je član do 19. veka bio pisan polusloženo, npr. жена-та). Ne znam o makedonskom.
    (*Regenerirao sam delimično izbrisani post.)
    Last edited: May 30, 2010
  22. sokol

    sokol Senior Member

    Vienna, Austria; raised in Upper Austria
    Austrian (as opposed to Australian)
    As for the timing of development of the enclitic article in Balkanbund languages - I am sure that this has come up before, somewhere here in EHL forum, but I couldn't find the relevant post.

    Anyway, as far as I remember somebody (not me :) I got it wrong there, of that I'm sure ;-) pointed out that there hadn't been an enclitic article in the first written Slavic documents (approx. 9th to 11th century B.C.), but it had been attested from the (14th?) 15th century onwards, in all Balkanbund languages which have it today (Greek doesn't, it's not enclitic there).
    And that in-between there was a period with hardly any written documents preserved (or something to this avail).

    Albanian, by the way, wasn't attested at all before the 15th century. So of course, yes, Albanian written documents surely had the enclitic article from the very beginning of written tradition.

    The point is (would be) what Albanian was like before the innovations of Balkanbund languages (with article included).
    However, those Viennese linguists focus only on what is called "Old Albanian" = the language from the beginning of written tradition till the 18th century.

    Of course, this development stage of Albanian indeed may give deep insights into the development of Balkanbund features - as well as does the evidence of Old Church Slavonic, and later written documents of Slavic -, but it does not indicate necessarily that those features of Albanian supposedly supporting the role of Albanian in the development of this article*) were present before the Balkanbund emerged.

    The point is that Albanian might not have had any more "supporting structures" for some features of Balkanbund languages than any other of them; as we know, there were "supporting structures" in both Slavic and Romanian for the enclitic article - and there might well have been some in Albanian (that is, before the 12th century; there surely were after, as established by this group of linguists - or so the article says).

    But this isn't good enough for me to claim any "precedence" of Albanian over the others.

    *) I'd like to give a longer quote than Alex did, the same article as quoted above by Alex:
    I must say that I don't like the way this article is written - it is a (careless) oversimplification.
    Especially "since ancient times" is misleading, as this means the late 15th century for Albanian (which isn't that ancient at all :D).
    Also I'm sure (or better: I hope sincerely ;)) that those reseachers were much more careful in their statements - meaning, I hope that not one of the participants of this project wrote this article.
    Unfortunately I am not part of this project (would like to have that job actually :).
  23. OldAvatar Senior Member

    Just a little detail.
    In Romanian, the article is -l only. u is just a connection vowel (omul)
  24. Claudiopolis Member

    I always wondered why was "l" from "ul" standardized in Romanian since most people don't use it in day to day conversation and I found it missing from many written texts from 18th and 19th centuries.
    Last edited: Jun 10, 2010
  25. OldAvatar Senior Member

    Hm... What about Vlad Dracul, for example, the father of Vlad Țepeș. Or was he Dracu? And he lived before 18th century? Oh, yes, it was 3 centuries before.
    Last edited: Jun 10, 2010
  26. Claudiopolis Member

    Of course, you are right, that's why I didn't use "all" or "most" but "many written texts".
  27. iobyo Senior Member

    Bitola, Macedonia
    In the Dobrejšovo evangelie (13th century) — злъiотъ рабъ.

    I think something similar has been observed in even earlier texts.
  28. todosmentira Member

    English - Southern English
    Isn't there evidence of a pre-latin substratum of Romanian which shows many Albanian cognates? Do you imply that Albanian suddenly popped into existence with the first attested written texts?

    Wasn't proto-Albanian (whether the language was at that time Dacian, Illyrian or something else entirely) thought to have been spoken originally further north-east due to the following features: lack of original sea/coastal vocabulary; many gothic loan-words; lack of Albanian-origin toponyms in present day 'Albanian' territory. All these, together with the cognates in Romanian are surely important regardless of the date of the first written document?
  29. sokol

    sokol Senior Member

    Vienna, Austria; raised in Upper Austria
    Austrian (as opposed to Australian)
    I referred to a quote claiming that Albanian had played a key role in the formation of Balkan sprachbund. Which it might even, but that is not the point.

    The point is that there is no proof for that, that it's just a theory - among others (claiming that other members of the Balkan sprachbund actually had "precedence" or dominating influence in its formation). The whole Balkan sprachbund discussion, unfortunately, still is very much controversial, even among linguists.

    Of course it is clear that there was a pre-Latin substratum in Romanian, and it might indeed be that those words which might (!) be of this substratum and which have been reconstructed in a way may show closer relation to the oldest stages of Albanian - but again this all is very vague and far from proven.
    (And even if proven - which might be the case for some verbs - the relationship might be on a general Indo-European basis, that is not particularly closer than any other relationship between different branches.)

    Also I never intended to imply that Albanian popped into existence out of nowhere - I don't see where you've read that in my post but as you have I apologise for for the misunderstanding.
    (Why should any language pop into existence out of nowhere anyway, the suggestion as such is ridiculous. Except of course for cases which are attested; like Esperanto, to give the most obvious one, as there's no doubt about it being an artificial language.)

    Anyway, the most important thing here, as far as this thread is concerned - with the definite article in Balkan sprachbund languages being the topic - that there is few linguistic evidence for Albanian being the origin of it, as well as there is few evidence for Slavic languages or Romanian being the source of it.
    There exist theories which favour certain features and developments in those three languages for being the origin but they actually might all be wrong, and the explanation could yet be another one. That was the point I was trying to make in my post.
  30. punctuate Senior Member

    I know neither Romanian nor linguistics, but I guessed it must be connected with the general features of adjective placement in Romanian. Whether my guess is right I don't know, because I don't know Romanian, but this short article (sections 3.3.2 and says that in Romanian, adjectives are put before the noun only in case of inversion or exclamation, and some adjectives like those naming a color can't be put before the noun even in inverted sentences. Moreover, demonstrative adjectives can be put after the noun as well, though they can be put before the noun, too. Say, in French it's different: there is a class of names that not so much modify the noun but work together with the noun on specifying the object (bonne, vrai, grand, cet). Now, a demonstrative or an article is an adjective like any other.
    I think that the approach of starting from searching a source language whenever we need to explain (that is, put in order) every detail of grammar is not constructive. Otherwise, when willing to answer why French puts the article before the noun, we would have, again, to start from looking for influences that would make this choice established as opposed to the free choice as in Latin, and that would never end. All good and well, but the key question is, why search for influences? Maybe the people decided everything for themselves? You know, when you're talking, you're inconsciously trying to assimilate to your neighbor next door, not to your "neighbor" who lives in a few villages of distance and just made trick on you while trading. And, in addition, there is a lot of depth in languages that does not let just any influence work directly; every language suggests its own algorithms of understanding. One would put a demonstrative adjective after a noun not because a neighbor does that, but because this placement has a meaning; and he would choose this meaning because that's what would be suggested by the depth of his language.
  31. irinet

    irinet Senior Member

    There are voices stating that the substrate is the trak-illiric. So, no wonder Albania could also be a source.
  32. danielstan Senior Member

    Romanian - Romania
    Definite article in Romanian
    Even from Vulgar Later the demonstrative ille has taken the function of definite article, with a "floating" position near the noun (proclitic or enclitic).
    The text Peregrinatio Aetherie (supposedly written in 380 AD) has an abundance of ille in proclitic and enclitic positions.

    Romanian and its dialects is the only Romance language with enclitic definite article (at nominative case),
    but there is an proclitic definite article at genitive/dative case for masculine antroponyms.
    E.g. Cartea lui Radu este buna. (genitive: "Radu's book is good"
    I-am dat cartea lui Radu. (dative: "I gave the book to Radu")

    Definite article for masculine singular declension - evolution from Latin to Romanian:
    Classical Latin lupus ille / lupum illum (accusative) > Vulgar Latin lupu *illu > lupulu > rom. lupul (Aromanian luplu)
    The final u in lupul is coming from the -us/-um termination of Latin lupus/lupum.

    There are many hypothesis about the enclitic position of the Romanian article. The best one I read (Al. Rosetti - Istoria limbii române, 1986):
    Romanian language uses normally the sequence:
    NOUN + ADJECTIVE (e.g. lup bun = "good wolf")
    (observation: it is correct to say "bun lup", but saying that is used to emphasis the adjective bun and has a different meaning than the normal expression lup bun)
    With this feature in mind, the expression "the good wolf" would have evolved:
    Classical Latin lupum illum bonum > VL lupu *illu bonu > lupulu bunu > rom. lupul bun

    There are linguistic facts that indicate the definite article was in proclitic position in a previous stage of Romanian (probably in late Vulgar Latin phase).
    In Romanian there some words which come from Latin words starting with a, but this a has been lost.
    lat. agnellum > rom. miel ("lamb")
    lat. Aprilis > rom. prier ("the month of April")
    lat. animalia > rom. dialectal nǎmaie (in Banat region of Romania)
    The lack of initial a could be explain by the existence of the proclitic definite article *illu/*illa which could have evolved to lu/a,
    provoking a confusion for the speaker in the above cases:
    lat. agnellum confused with "a" + gnellum
    and the form miel reconstructed from gnellum.

Share This Page