I've read that languages with the same word order often have similarities, even if they're not related, purely because some grammatical features will force a language to use others. For instance, if a language doesn't indicate case, it'll be forced into a verb-medial order (SVO or OVS) to differentiate subject from object.

What I'm finding annoying is I can't find anything like this for verb-final languages. The only other language I know to any extent is German. I have tried to learn Japanese in the past, and more recently Korean, but I never got that far with either.

I also find it annoying that you can't just find lists of grammatical features languages can have. Everything I can find just goes into excessive detail about one specific language rather than giving a general overview of languages that fall into specific categories.

Japanese is pro-drop, and its common to leave out most noun phrases. Japanese tend to just leave it to the listener to fill in the gaps. That doesn't mean they never specify subjects and objects, they just don't bother to repeat them once they're introduced. And of course Japanese is topic-prominent, so its normal to assume that every sentence has the same topic until another is specified.

But are all SOV languages like this? Personally, I find it annoying that you have to work your way through all the noun phrases before you get to the verb, which does most of the work in determining the role of everything in the sentence really; i.e. you can't figure out what happened to the direct object until you get to the verb.

The only 'universals' I know of is they tend to be head-final and post-positional. That's it. Like I said, I don't know one myself to any degree (and yes, I know German sometimes uses SOV, but its primarily SVO, well unless maybe you count sentences with auxiliary verbs).

  • Which question are you really asking: "What grammatical features do SOV languages often share", or "are all SOV languages like this?".
    – user6726
    Feb 11, 2018 at 5:58
  • 5
    I don't think it's true that "if a language doesn't indicate case, it'll be forced into a verb-medial order (SVO or OVS) to differentiate subject from object." If a language has strict SOV or VSO word order, for example, the identity of the subject will be unambiguous even without differential marking of the subject and the object. And even if the word order is not so strict, languages can get along with some ambiguity. Consider that many languages that have grammatical case have nominative-accusative syncretism in at least some contexts (it's universal in Indo-European for neuter gender) Feb 11, 2018 at 6:06
  • 1
    The relevant universal seems to be #8 in "The Universals Archive", but it's stated a bit differently ("IF basic order for nominal arguments is verb-final (i.e. SOV or OSV), THEN there is almost always a case system") and the notes mention a number of counterexamples and express some doubt as to its validity. Feb 11, 2018 at 6:28
  • 2
    (Verb agreement also seems to be an important way subjects and objects are disambiguated in languages that don't have case markers: Universal 296 is "IF nouns do not inflect for case and verbs carry no relational marking either, THEN basic order is SVO or OSV" and it seems to be harder to find counterexamples to that one) Feb 11, 2018 at 6:39
  • To add to @sumelic's point, prepositions are yet another strategy for preserving free word order. Most Iberian Romance languages generally require a before an animate direct object. This is very analogous to Slavic requiring genitive forms for masculine accusative animate, which would otherwise be indistinguishable from nominative. Feb 13, 2018 at 7:57

1 Answer 1


One must be careful in making generalising statements, for example, languages without (morphological) case needn't be SVO or OVS, just look at Abkhaz. However there are typically some tendencies or predominant patterns. In the case of SOV, the languages tend to be agglutinative (which mostly implies rich morphology and therefore pro-drop) and non-configurational whereby SOV only applies to non-emotive (information-structurally unmarked) utterances. They also often exhibit head-marking predicates. If a SVO language has adpositions, they tend to be postpositions.

BTW German is neither SOV not SVO but V2 in the standard typology.

  • German is considered to have SOV as the "basic" clause order, with V2 being a less basic order that occurs in specific types of clauses (I forget the details, but basically finite clauses). Feb 11, 2018 at 16:15
  • German word order is complicated. A basic sentence without auxiliary verbs is v2, but if there's IS an auxiliary, then the main verb is moved to the end and only the FIRST auxiliary can stand in the second position. Multiple verbs can end up at the end of a sentence this way. In dependent clauses however, the order is always SOV. German kind of defies a specific classification.
    – user19661
    Feb 13, 2018 at 5:06
  • @IXBlackWolfXI Replace "can stand" with "must stand". A more precise description would be that German is SIOV (using X' symbols) in matrix clauses. Embedded clauses would be SOVI though there are exceptions. Nevertheless in standard typological terminology V stands for the finite verb (non-finite verbs correspond the NPs in many languages anyway and diachronically/functionally they're often nominal) so describing German as V2 makes sense (for matrix clauses) if one is familiar with the conventional terminology.
    – Atamiri
    Feb 14, 2018 at 1:25

Your Answer

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