10

I'm off to Albania tomorrow so starting to get more and more interested in the language.

It's one of the outliers on the Indo-European family tree. It's not hard to see a relationship but it's not easy to say what that relationship is.

As well as the distant genetic relationship it coexists in the Balkan sprachbund which might blur the picture a little.

Have linguists come to agree to any extent which other branch(es) are most closely related to Albanian? Greek? Balto-Slavic? If you Google you even see Romanian suggested.

hippietrail
  • 14,687
  • 7
  • 61
  • 146
  • 5
    I've seen at least three attempted reconstructions of the sequence of branchings of the IE family, each of which tried to minimize the number of independent innovations. In the oldest of these (2009 or before), Albanian's nearest cousin is Germanic, surprisingly. In the others (2012 and 2016), it's Greek or the common ancestor of Greek and Armenian. Sorry I didn't save any citation. – Anton Sherwood Nov 25 '19 at 02:49

4 Answers4

13

I'm sure you've found enough information on Wikipedia and other places that make it clear that Albanian isn't closely related to any other living language, but for those who haven't the most concise summary I've found is at Krysstal's Language Families ( http://www.krysstal.com/langfams_indoeuro.html):

"The Illyric Branch [of Indo European]. Only Albanian (called Shqip by its speakers) belongs to this branch. It has been written in the Latin script since 1909; this replaced a number of writing systems including Greek and Arabic scripts."

"There are two dialects that have been diverging for 1000 years. They are mostly mutually intelligible. Geg is spoken in the north of Albania and Kosovo (Kosova). Tosk is spoken in southern Albania and north west Greece."

"The ancient Illyric and Mesapian languages, spoken in parts of Italy, are considered by some to be an extinct member of this branch."

So we have to go back to this Illyric branch, which it is argued is part of the Paleo-Balkan group. The only other surviving member of this group is Modern Greek (http://en.wikipedia.org/wiki/Paleo-Balkans_languages). Given how completely dissimilar Modern Greek and Albanian are this answer appears rather unsatisfying - until you realise how amazing it is that Albanian managed to survive this long when any closely related languages died out so long ago!

LaurenG
  • 3,127
  • 1
  • 21
  • 34
  • 3
    Yes it doesn't feel like Greek at all. In fact I visited another Indo European outlier last year, Armenian and found it's tenuous Greek influences a bit more obvious! – hippietrail Sep 29 '11 at 14:38
  • 2
    I guess that's what a couple of thousand years of linguistic divergence will do! Still, it definitely fits into an Indo-European schema, unlike something like Basque! – LaurenG Sep 30 '11 at 02:11
  • Yes the IE link is more obvious with Albanian with its numbers and word for "is". For Armenian I only seem to recall some words similar to Greek words and I don't remember which ones. – hippietrail Sep 30 '11 at 04:21
  • 2
    Nothing or close to nothing is known of the languages from which Albanian evolved, and which are supposed close to Greek. Albanian itself is not closer to Greek than to other languages that now exist in the same Balkan area (the same area where the Albanians lived, as well as their ancestors and the ancestors of other Balkan peoples). Given that the Balkan Sprachbund trends are shared by Albanian more with Bulgarian and Romanian than with Greek, makes Greek only one of the candidates. – cipricus Nov 19 '19 at 15:41
  • 2
    Greek is not part of the " the Paleo-Balkan group". – fdb Sep 22 '23 at 17:15
5

Albanian has no surviving immediate relatives, and in fact no known close relatives at all within the IE family. It constitutes a separate IE branch all by itself.

But, taking the terms "relative" and "closeness" in a more common or general sense, one could mention its relation to Romanian.

There is a specific Wikipedia article on Albanian–Romanian linguistic relationship:

Albanian and Romanian share most Balkan features, but they also have common features which do not characterize other Balkan languages.

The Romanian philologist Bogdan Petriceicu Hasdeu who studied the pre-Latin elements of the Romanian language came to the conclusion (in 1901) that the origin of the shared vocabulary was most probably to be searched in the earliest phase of the two peoples' ethnogenesis... The Albanian linguist Eqrem Çabej was the first to emphasize the similar phonological and morphological elements of the two languages. He also drew attention to the similarities between Albanian and Romanian proverbs and the parallel development of the formation of sentences.

The Tosk dialect of Albanian, spoken in Southern Albania, in particular is held to have experienced developments parallel to early Romanian. ...The proposed interaction between Tosk and Romanian is held to have been the last stage of the crucial Albanian–Romanian period of convergence. Since these effects are marginal if present at all upon Slavic loans, they likely occurred before contact with Slavs.

When comparing the morphological elements of the four core languages of the Balkan linguistic area, scholars have concluded that Albanian and Romanian share most common features.

Albanian is not a Romance language but has a large Latin vocabulary. It also shares, along with Romanian (also Bulgarian and Macedonian), the core features of the Balkan Sprachbund.

From the Albanian point of view, Romanian is close, given no closer related languages survive. From the Romanian point of view, other languages are obviously closer relatives than Albanian: first of all the other neo-Latin languages, then maybe South-Slavic languages like Bulgarian, given the large Slavic lexicon of Romanian. On the other hand, the common Slavic influence brings Albanian and Romanian close to each other too.

Considering only the area of Eastern Romance, Albanian is close even if it is not part of the Romance family: essential features of Eastern Romance are only present in Romanian and Albanian. (At the same time these features are connected to those that define all members of the Balkan Sprachbund).

That doesn't make the two languages closely related, but makes them at least relatively close.

Here is an article with a 17-page list of common words of the two languages (of various origins: substrate, Latin, Slavic etc): Common Lexic in Romanian and Albanian. Substrate and Loanwords, by Dan Ungureanu.

Here's a snippet:

enter image description here enter image description here

Dan Alexe in Dacopatia şi alte rătăciri româneşti ("Dacopathy and other Romanian delusions" - part of that here) argues that beside the few tens of terms only common to Albanian and Romanian, Albanian has a Latin vocabulary that is not only comparable in number to the Romanian, but is also common to Romanian. He also argues that many Latin, Slavic and other words common to both languages have been transformed in the same way in both, and according to the same rules.

Funny enough (but also interestingly) he mentions many Romanian terms normally marked as having unknown origin which can be explained through Albanian — e.g. bordei (small village house with an underground chamber), from borde ("hole" in Albanian); the odd Romanian expression eu unu(l), which means in fact "as for me", but very unusual for Romanian (literally "I the one"), from Albanian üne which means "I". —The author tries also to explain the form of the first person singular of "to have" (eu am=I have) through the influence of the Albanian kam (meaning "I have"); but that form has to be related to the Slavic form of "I have" (Bulgarian, Serbian, Slovenian imam, Polish ia mam, Czech mám)—

cipricus
  • 696
  • 6
  • 16
  • 3
    Shared areal features and borrowings don't actually tell you anything about whether or not two languages are close relatives, though. One is a question of contact, the other of descent. – Cairnarvon Oct 25 '23 at 13:47
  • It's not the case that Albanian has no known relatives; it's well-established as being an Indo-European language, and as such it will be more closely related to some IE languages than others. That's exactly what the question is about. – Cairnarvon Oct 25 '23 at 21:07
  • 1
    @Cairnarvon The branch of Indo-European that Albanian belongs to has no other surviving members, so Albanian has no surviving immediate relatives (which is indeed what the answer says). It is of course more distantly related to all other Indo-European languages, but there is little consensus on which – if any – of the other branches is closest. Germanic, Greek and Italic (and possibly others) all have their proponents. – Janus Bahs Jacquet Oct 25 '23 at 21:48
  • @JanusBahsJacquet Yes, and that's what the actual answer is. This answer doesn't seem to realise what the question even is. – Cairnarvon Oct 25 '23 at 23:27
  • 1
    @Cairnarvon Have you actually read the answer? It starts off by very plainly saying that Albanian has no remaining immediate relatives (I've clarified ‘immediate’ now), but forms its own branch, answering the genetic aspect (which has also been covered in other answers). Then it goes on to cover some non-genetic ties that make Albanian closer in structure and appearance to Romanian than to probably any other Indo-European language (while clearly noting that these ties are indeed non-genetic). I cannot fathom how you can possibly conclude from this that the question has been misunderstood. – Janus Bahs Jacquet Oct 26 '23 at 00:09
  • Shared archaisms, which includes non-borrowed lexical items, are not indicative of close genetic affinity, only shared innovations can be. – Cairnarvon Oct 27 '23 at 15:26
  • 1
    This isn't going to be a very scientific argument, but there are some isolate languages that "look like" they're truly independent branches from PIE and other isolate languages that "don't look like" that. Even though Greek is an isolate, Greek "looks like" its own branch from the PIE. Even though Albanian is an isolate, Albanian "doesn't look like" an isolate. It looks like something similar happened to Albanian as the Piraha language and the Mura language family in the Amazon. – Fomalhaut Oct 27 '23 at 22:12
  • This is exactly what I'm talking about when I say it really sounds like you don't understand what the question even is. – Cairnarvon Oct 29 '23 at 13:02
  • @Cairnarvon - The OP is fully aware that Albanian is a separate IE branch and says it's an outlier. No "close genetic affinity" can be at stake here. - You are the one talking about that. - I deleted all my comments (addressing you) because they seem useless to you. You could do the same with the ones you have addressed to me. – cipricus Oct 29 '23 at 19:02
  • The question asks about Albanian's closest living relative. The fact that Albanian is "a separate IE branch" does not mean that it doesn't have one—a fact that the other answers grasp, in fact. My point is that you seem determined not to understand what it means for Albanian to be "a separate IE branch", and in particular that it doesn't mean that there is no other branch that is genetically closer to it than the others, and that that is in fact exactly what the question is asking. – Cairnarvon Oct 29 '23 at 20:13
2

From what I've read, the clear consensus among linguists is that Albanian is a descendant of the Paleo-Balkan language, which has already been mentioned. What hasn't been mentioned is the ancient Dacian language. The Pre-Roman Romanians seem to have been Dacians. There also seems to be a strong relationship between Dacian and Albanian. Romanian seems to be a Latinization of a Dacian substrate. Dacian, Thracian (or Daco-Thracian as a single language group), Illyrian and Albanian have a close relationship that linguists haven't been able to agree on though. Greek, Macedonian (or Macedonian as an archaic northern branch of Greek), Armenian and Phrygian would be other, more distantly related branches of the Paleo-Balkan branches of the Indo-European languages. The relationship of Greek and Armenian with Indo-Iranic languages is another complicating factor.

Hope this helps.

DarkWolf
  • 29
  • 2
  • 2
    a strong relationship between Dacian and Albanian - we have to note that there is no known Dacian language as such. The relation of "Dacian" to Albanian only makes sense in an effort to re-construct or hypothesize Dacian lexicon based on the Albanian one, and especially on the one common with Romanian. We more or less know who the Dacians were but we don't know their language. That language (or languages) would be just one (or more) of the many that existed in Dacia and the Balkans that didn't survive (including ancient Macedonian). Albanian is the only remnant. – cipricus Oct 25 '23 at 22:03
  • 1
    If a word is present only in Romanian and Albanian it is sometimes considered "Dacian", although "of Balkan substratum" is more adequate. The number of such words is small, and getting smaller as some of them are identified in Western Romance that were initially considered Dacian-Albanian only (barză or balegă) but are in fact present in Italy, Switzerland and France, that is in Celtic areas. Maybe Romans brought those words or the Balkan languages were closer to the Celts than expected. See this. – cipricus Oct 25 '23 at 22:17
1

I have seen a lot of dendrograms of the Indogermanic language family and it seems that Albanian is rather randomly grouped with respect to other branches, sometimes it is closest to Hellenic/Greek, sometimes to Armenian. What I find rather striking is a grouping together with Germanic as in this diagram

enter image description here A chart of the history of the evolution of Proto-Indo-European, with approximate/hypothetical dates. (Image taken from this answer)

and I find this grouping particularly appealing, since Germanic and Albanian share some innovations such as the word for "and" based on a word meaning "before" (like Latin ante) in other branches.

Sir Cornflakes
  • 30,154
  • 3
  • 65
  • 128
  • 3
    iirc (I think from the dictionary of selected synonyms in the principal indo-european languages) Albanian shares more lexical isoglosses with Germanic than any other branch (followed by Baltic). As statistical phylogeny tends to rely primarily on lexical isoglosses it makes sense that this chart would place Albanian closer to Germanic than any other family – Tristan Sep 27 '22 at 09:53
  • 1
    connections to Greek are also not as arbitrary as they might seem. Phonologically they share some features: palatalisation of labiovelars (although this is certainly a secondary areal effect, as it is not present in all Greek varieties), s > h (only intervocalically in Albanian), syllabic nasals > a (in Greek only short syllabic nasals), and loss of y & w between vowels. From what I remember reading Orel's book on Proto-Albanian it's verbal morphology (at early stages) also seems quite similar to that of Greek (although also to Proto-Balto-Slavic) – Tristan Sep 27 '22 at 10:00