10

I was just answering a question about the origins of English and Latin and wanted to talk about their common ancestors but ran into a surprising problem.

So we know the majority of languages in Europe and the Indian subcontinent are related under the name "Indo European".

If you just want to look at the languages in the subcontinent there is the name "Indic".

But when you want to look at the other half, I can't seem to find a name.

English is a member of the Germanic family. Latin is a member of the Italic family.

Both are on close branches on the European side of the Indo European tree but "European" is not the name of a language family as far as I am aware. When we talk about European languages the terms covers an area including unrelated languages such as Basque, Finnish, and Hungarian.

So if "European" is not a name of a language family and Indo-European is far too vast, is there a term I'm overlooking?

hippietrail
  • 14,687
  • 7
  • 61
  • 146
  • 4
    Very roughly speaking, PIE breaks down into Anatolian, Centum, and Satem. Anatolian died out early. Satem is Balto-Slavic and Indo-Iranian. Centum is Germanic, Italo-Celtic, and Greek. So Centum would be the next common ancestor of Germanic and Italo-Celtic. The big difference between them is an early palatalization of k in Satem that didn't happen in Centum. Both words mean "100"; Satem is Iranian and Centum is Latin, both from PIE kmtom. This is extremely simplified; there's lots of loose ends. – jlawler Jun 12 '13 at 02:44
  • @jlawler: That sounds like the answer I'm looking for then. I saw Centum and Satem on some diagrams but those labels weren't applied in the same manner used for language families so I wasn't sure. – hippietrail Jun 12 '13 at 03:12
  • 10
    @jlawler centum and satem are not taxonomic divisions. – Anixx Jun 12 '13 at 07:20
  • 1
    There is no exact taxonomy with reconstructed languages. There are only hypotheses with greater or lesser likelihoods, in the opinions of experts. Who do not agree, generally, on details. – jlawler Jun 12 '13 at 14:37

4 Answers4

16

I have never heard of such a term. And it may even be impossible, depending on which hypothesis you subscribe to. According to the below reconstruction, the last common ancestor of the Italic and Germanic branches existed some 5200 years ago, when the Italo-Celtic super-branch split off. That was after the Anatolian and Tocharian branches had split off from the rest, but before the Germano-Albanian super-branch and the Indo-Iranian branch diverged. In other words, the Indo-European languages mainly located in Europe share no common ancestor that excludes the Indo-Iranian branch, and neither do the Germanic and Italic branches.

But other reconstructions may have different ramifications. [Update: You can see what a mess Proto-Indo-European migration was: it all depends on dating the big yellow arrows in the map at the bottom.] Perhaps some reconstruction exists that groups the Italic and Germanic branches closer together than either branch to the Indo-Iranian (super-)branch. Note also that the Greek, Albanian, and Balto-Slavic (super-)branches are rarely grouped closer to both Italic and Germanic than to Indo-Iranian. I therefore do not expect any reconstruction to have a "European" super-branch that excludes Indo-Iranian, even if a super-branch may be reconstructed including Italic and Germanic but excluding Indo-Iranian. To put this into perspective, there is no "non-Germanic" super-branch including all Indo-European branches except Germanic either.

enter image description here

A chart of the history of the evolution of Proto-Indo-European, with approximate/hypothetical dates.

http://indo-european-migrations.scienceontheweb.net/map_of_indo_european_migrations.html

Source.

Cerberus
  • 7,976
  • 33
  • 49
  • This was actually the diagram I mentioned seeing the words Centum and Satem in. But I mustn't've understood it deeply enough because it disproves some of my intuitions, which I bet other people must also have. Thanks for a great answer! – hippietrail Jun 12 '13 at 03:54
  • 1
    @hippietrail: Right, that's what I thought. Note also that the centum languages are not necessarily (and not in fact) all more closely related to each other than to the satem languages: it's just that a certain super-branch came up with a new innovation. The only thing that's special about that is that all the other languages kept the typical centum features (even though they may have other, starker differences among them). Lastly, note that centum itself was an older innovation, which probably happened before Proto-Indo-European branched out (which is why all but the satem languages have it). – Cerberus Jun 12 '13 at 04:27
  • 1
    @Cerberus no, there was never a satem super-branch. The satemization was a continual process affecting diverse branches. For example, Balto-Slavic is close to Germanic but Germanic is centum while Balto-Slavic is satem (and Slavic is more satem than Baltic). – Anixx Jun 12 '13 at 07:24
  • @Anixx: I made my comment with reference to the graph I posted above. – Cerberus Jun 12 '13 at 11:58
  • Under this model (and I think most current models), the Centum languages are what (biological) taxonomists call paraphyletic: they're what's left when you take out one or more subnodes. – Colin Fine Jun 13 '13 at 23:09
  • @ColinFine: Right, exactly. And the only reason why this paraphyletic group is interesting because it has certain properties in common. The mere fact that it is non-satem doesn't make it interesting: it is rather the fact that centum didn't change into something else in that entire group (like satem, but also censum or hentum). If it had changed into, say, censum in the Italo-Celtic sub-branch, then we couldn't call all non-satem languages "centum" any more. – Cerberus Jun 13 '13 at 23:26
  • 2
    @Cerberus: that more or less characterises paraphyly. You could call (say) birds-and-crocodilians a paraphyletic group, but nobody would bother to do so because they don't have any salient characteristics other than those shared by the stem group. But reptiles were treated as a group for a long period, but now that it has been established that birds belong in the same clade, reptiles can be seen as a paraphyletic class. A similar thing has happened to centum languages. – Colin Fine Jun 14 '13 at 13:32
  • @ColinFine: Right. The only thing that might not be common to all paraphyletic groups is the salient features of the centum languages are all supposed stem from a single ancestor, i.e. there was no convergent evolution. This may not always be the case with biological paraphyletic groups. – Cerberus Jun 15 '13 at 02:12
  • @Cerberus: I don't follow. If there is a common ancestor to centum languages (and no others) then that ancestor is a valid clade (in biological terminology) and they are not paraphyletic. But I don't believe that current scholarship supports that view. Convergent evolution can make it harder to see the real relationships, in linguistics as in biology; but the factor where the parallel between the domains does break down is major borrowing (or language blending). – Colin Fine Jun 16 '13 at 16:31
  • @ColinFine: Right, borrowing is quite unusual, I believe (except from/to miscroscopic organisms). I read somewhere that the centum features developed only once and in a common ancestor of all known IE languages (perhaps some IE branch split off before centum developed, but then it must have disappeared without a trace), so that would probably have to be late PIE. The whole group would then be a clade; but the whole group minus satem (satem features emerged later in a super-branch that used to have centum features) would be paraphyletic, I would say. – Cerberus Jun 16 '13 at 23:17
  • But why is Armanian in the same branch as Germanic in the chart? – Ergative Man Dec 14 '19 at 18:34
  • 1
    @ᚹᛟᛞᚨᚾᚨᛉ: I'm not sure I understand. In the chart, Armenian and Greek share a branch. – Cerberus Dec 15 '19 at 20:31
  • Oops, @Cerberus, I meant Albanian hahahah – Ergative Man Dec 16 '19 at 03:12
  • Contrasting with this tree, I've seen two that, while they disagree in much else, both make Germanic the sibling of Italo-Celtic, and their parent the sibling of Balto-Slavic. (Sadly I do not know where I found them.) — Satem strikes me as a non-issue, given the similar shift in Romance; if we know it has happened at least twice, why not three times? – Anton Sherwood Aug 18 '20 at 04:03
  • One of the trees mentioned in my last comment is in https://www.linguisticsociety.org/sites/default/files/news/ChangEtAlPreprint.pdf – Anton Sherwood May 16 '21 at 02:09
2

No. But some researchers hypothesize common ancentor of Germanic and Balto-Slavic. Others hypothesize a common ancestor of Italic and Celtic.

So, Latin and Germanic according to these theorires belong to different sub-branches of Indo-European.

There is nothing in common between Latin and English besides they are both Indo-European.

Note though. That there is something that links Germanic and Italic: the both are centum languages. But this is not a genetic trait, rather satemization is an areal feature that spread over some region over different branches after they already split.

Being centum is the default mode for all Indo-European languages because PIE was centum. That said, Germanic is typologically closer to Balto-Slavic than to Italic, for instance, they share endings of plural Dative and Instrumental cases: the both use -mos and -mis, while the rest of IE use the derivative of -bhi̯os and -bhis (in Latin -bus), also they share the word for thousand: tua̯scmtom, not attested anywhere else in IE (other branches used smĝheslom for thousand), but Germanic is centum while Baltic and Slavic are satem.

Anixx
  • 6,643
  • 1
  • 26
  • 38
  • 3
    But being centum isn't a "default mode", because proto-indo-european wasn't centum nor satem, since the latter merged the labiovelars with the plain velars and the other merged the plain velars with the palatovelars. In the case of satem languages, the palatovelars were assibilated. – Ergative Man Dec 14 '19 at 18:49
  • @ᚹᛟᛞᚨᚾᚨᛉ PIE was centum. It is the default. We know this because the periphery is centum while the middle is satem. – Anixx Dec 14 '19 at 20:34
  • 1
    But it doesn't make sense, how can PIE be centum if it had both plain velars and palatovelars? – Ergative Man Dec 16 '19 at 15:38
  • 3
    No, PIE was not centum, nor was it satem. The centum/satem divide does not apply to PIE at all, and satem did not evolve out of centum – it couldn’t have, in any possible way. The fact that satem languages are located centrally (sort of – Indic is only ‘central’ inasmuch as Tocharian is slightly further east) does not in any way prove, or even indicate, that centum was the ‘default’. – Janus Bahs Jacquet May 14 '21 at 06:10
  • @Anixx And what about the languages that are neither centum nor satem, showing distinct reflexes for all three "velar" series? – Draconis May 14 '21 at 16:38
-2

Germanic arose fairly early in southern Scandinavia (c. 2500 BC) as a patois used between indigenous Vasconics, Para-Italic migrants such as the Ligures and Venedi, and Paleo-Baltic invaders from the east. These pre-Germanic peoples solidified into the Proto-Germanics by 2500 BC. In the near future, this will probably be expressed by: Celto-Italic splits into Celtic and Italo-Germanic, followed by: Italo-Germanic splits into Italic and Germanic.

Mark Lanzarotta
  • 5
  • 2
  • 6
    Any source? I know of no scholar that would group Italic and Germanic, and especially count Germanic closer to Italic than Celtic. Germanic does not have the sound changes, commonly shared by Celtic and Italic. – Anixx Jun 22 '15 at 03:31
  • In prehistoric Scandinavia, the Vasconic indigenes (some of whom were Uralicized to become the Sami) most likely worshipped the Jotuns. These were followed by the Ligures from the North Sea coast, who worshipped their God Lugh (Loki). The Venedes came next from the area of Mecklenburg, bringing their pantheon of the Vanir. But it was the Baltic invaders who brought the Aesir gods, who were really the Asuras borrowed from their neighbors the Scythians (who were Iranians). – Mark Lanzarotta Jun 22 '15 at 02:56
  • The satem Balts and Iranians are actually related. Germanic has Baltic features due to an early invasion of satem Balts into an essentially centum Scandinavia. Slavic later arose when Balts combined with Venedes and Daco-Thracians in Silesia (Lusatian culture). Some of these Proto-Slavs moved east to the Pripet Marshes, mixed with Iranian Scythians, and became the Eastern Slavs. The date of the Slavic genesis was about 1450 BC. If you check some linguistic articles and charts on proto Indo-European, you will find that a connection exists between Germanic and Celto-Italic, AS WELL AS a strong c – Mark Lanzarotta Jun 22 '15 at 03:07
  • The satem feature arose when a group of Proto-Indo Europeans briefly conquered and occupied the Uralic heartland, the Volga-Kama confluence, around 3000 BC. Satemization comes from Uralic influence. The Uralics eventually drove them out, and they spread to other parts of Eurasia, introducing their new linguistic feature. The three main branches of Satem were Old Baltic, Old Balkan, and Proto-Iranian. – Mark Lanzarotta Jun 22 '15 at 03:21
  • @MarkLanzarotta Please edit your comment and include all the other paragraphs right there. Multiple answers are OK if you're adding a completely independent answer, but splitting your answer across multiple posts makes them all unreadable. – prash Jun 23 '15 at 03:16
-3

It will be "North-West Indo European", a branch of IE. See Wikipedia if you don't believe. Speakers of this branch have the paternal haplogroup R1b.

Mahmud
  • 11
  • 3
    North-West Indo-European is not an IE branch. It’s a cultural region whose languages have various areal and substrate features in common, but it is not a taxonomic entity. Also, Italic is not normally considered part of NWIE. – Janus Bahs Jacquet May 14 '21 at 06:18