I was looking at a sprachbund called Standard Average European, which seems to include Germanic, Romance and Slavic languages. I will not list all the features here since they can be found on Wikipedia, but I am interested in the group of features which features can be seen as dividing SAE languages into two major groupings:

One the one hand we have group A, which includes English, Norwegian, German, Spanish, French, etc. And then we have group B, including Latin, Russian, Czech, Latvian

  • definite and indefinite articles exist in group A but not in group B

  • a periphrastic perfect formed with 'have' plus a passive participle exists in group A but not in group B

  • verb-initial order in yes/no questions exists in group A but not in group B

  • subject–verb–object is generally unmarked word order in group A but not in group B

  • Group B languages generally have a much richer case system, which don't seem to syncretise prepositional and dative, and which seem to include other cases which are not found in Group A

I am sure we can find others. Do we have grounds for actually stipulating two separate sprachbunds here? Have linguists defined these and looked at this more closely?

  • 5
    Latin shouldn't be included because SAE is a grouping of modern languages.
    – TKR
    Jan 24, 2017 at 23:10
  • 2
    Essentially, group B is the synthetic group, the most inflected, with the most variation and freedom in word order. Group A is the one that's lost morphology and had to develop more periphrasis and auxiliary particles to compensate. There is an inverse relation between the amount of constituent linkage done by morphology and the amount done by word ordering. Oh, and I wouldn't call it a Sprachbund; it's just a wavicle of I-E.
    – jlawler
    Jan 25, 2017 at 1:33
  • 2
    No, no grounds for n=2. Most every language has some shared features with its historic neighbours. So there are endless tighter and looser overlapping Sprachbünde. The actual borders are arbitrary and subjective. (Czech and German share plenty, for example, and you have not listed many languages, some of which, like Macedonian break your grouping -- definite articles, auxiliary 'have', no morphological case.) Depending on what you are trying to achieve, I think a thorough statistical approach without foregone conclusions makes more sense, with multi-dimensional or even visual output. Jan 25, 2017 at 9:40
  • 2
    The features you describe for group B are the features inherited from PIE.
    – Anixx
    Jan 30, 2017 at 14:59
  • @jlawler Wavicle? Wave and particle? Most of the Indic-Iranian languages don't have any of these features.
    – Mitch
    May 7, 2018 at 13:25

1 Answer 1


First, Latin is not part of Standard Average European (SAE): the sprachbund is thought to have emerged through language contacts during the early middle ages and later, at a time when (classical) Latin was long dead. Actually, the fact that Latin does not share the SAE characteristics is one strong argument to define SAE characteristics as coming from Sprachbund, and are not retained Indo-European features retained genetically in the Romance, Germanic (and to a lesser extent) Slavic families.

Since SAE is a Spachbund, it borders are a bit fuzzy, some language being at the core of the Sprachbund (French, German), and others at the periphery (Russian, Bulgarian, Magyar). For more details on groupings of SAE languages, you can read Martin Haspelmath paper The European linguistic area: Standard Average European, in Language Typology and Language Universals, Vol. 2 pp. 1492–1510 (2001) (scanned pdf, better file on academia.edu).

In this paper, map 107.13 looks at 9 features of SAE and groups European languages depending on how many feature they share. I advise you to look at the map and read the text around it, even if you don’t have the courage to read the whole paper. Converting this map to a list, the European language ordered according to their degree of membership in SAE are:

  • 9 features (nucleus of SAE)
    • French, German
  • 8 features (next layer)
    • Dutch, Spanish, Portuguese, Sardinian, Italian, Albanian
  • 7 features
    • English, Romanian, Greek
  • 6 Features
    • Icelandic, Norwegian, Swedish, Czech
  • 5 features (the most peripheral languages of SAE, since there is no language with 3 or 4 criteria: this threshold defines the boundary of SAE)
    • Latvian, Lithuanian, Russian, Ukrainian, Polish, Hungarian, Slovenian, Serbian/Croatian, Bulgarian)
  • 2 features (Non SAE)
    • Breton, Basque, Maltese
  • 1 feature
    • Welsh, Georgian, Armenian
  • No feature at all
    • Irish, Finnish, Estonian, Nenets, Komi, Udmurt, Tatar, Lezgian Turkish

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.