Typology & Change

[Ketumak]

We classify languages according to criteria like word order, morphosyntactic alignment type or marking strategy, so I was wondering, can we say anything about what happens when these change? Can we make general or general-ish statements like "A VSO language is most likely to turn into an SVO language" or "An isolating language is more likely to move towards head marking than dependent marking" (or vice versa)?

I've long been curious about questions like that, though I've never come across anything that attempts to answer them at that kind of general level.

[Grunnen]

I do think I read an article once on this topic. I don't remember too well... What I did find after some googling was a book chapter named: On the reconstruction of 'proto-world' word order. that you can find on google scholar. They claim that OV -> VO is a "more common and natural" change than the reverse. It seems like they cite something to back this up, but the google preview leaves me guessing.

[vokzhen]

I have seen a supposed map of how languages change word order, in terms of which S-O-V patterns turn into which types, though I don't have time at the moment to search for it. I also remember seeing a more specific map of which features change in which order, I believe on here, but was unfortunately unable to re-find it when I went searching a few weeks ago. So they do exist, somewhere.

I have nothing but intuition to back this up, but going from isolating to dependent or head marking might be interdependent on word order or word order change. Given than it's apparently much more common to have case suffixes, it seems to follow dependent-marking languages are more often likely to start out SOV, grammaticalizing postpositions as case markers. Of course, there's some big problems with that conclusion that would need to be worked out (most isolating languages are SVO, and while most dependent-marking languages are SOV so are most of the languages with highly synthetic verbs).

And in that vein, the universals archive says that a higher number of V1 languages are ergative than for other word orders, so there may be some interdependency there. Though with a much smaller sample of languages than with SOV or SVO, and that a huge number of them are either genetically related (Austronesian) or under strong areal influences (Mesoamerica and the Northwest Coast), it also seems like it might just be historical accident. V1 and case-marking is similar, I believe, with a minority of V1 languages having case, and extensive case systems are astoundingly rare, while a large number have polypersonal agreement and many are polysynthetic.

[Ketumak]

Thanks, both, it's occurred to me I've not got back on this. I found grunnen's "Proto-world" article on Google Scholar. It's called "On the reconstruction of "Proto-World" word order" and is by Frederick J. Newmayer. Looks interesting. It's here.

I may also have found vokzhen's map. Chapter 81 of WALS has a map of word orders in ancient Europe, N. Africa and S.W. Asia.

I also turned up this article on Languages of the World. It's about word order aquisition in young children and history of word orders in languages. Whilst it's all of interest, the bit that caught my eye was part down and is worth quoting:

Note that the more frequently encountered types of word order changes correspond also to the more frequently encountered types of mixed word order languages (see Table 1 in G-M&R’s paper): SOV/SVO, SVO/VSO, VSO/VOS, SVO/VOS and SOV/OVS. According to G-M&R’s data, 46 languages with mixed word order patterns (37% of the total mixed word order languages) have both SOV and SVO (according to WALS figures, these numbers are 29 languages, 43% of all mixed word order languages). The next most common combination is either the combination of SVO and VSO (according to G-M&R, 24 languages or 19%; according to WALS, 13 languages or 19%) or the combination of VSO and VOS (according to G-M&R, 17 languages or 14%; according to WALS, 14 languages or 21%). Next comes the combination of SVO and VOS (according to G-M&R, 11 languages or 9%; according to WALS, 8 languages or 12%), followed by the combination of SOV and OVS (according to G-M&R, 9 languages or 7%; according to WALS, 3 languages or 4%).

This would figure if the mixed order languages represent the transitional stage.

[vokzhen]

Sorry, it wasn't an actual geographic map. I did refind it here, though looking at it again, a lot of their assertions are rather... suspect (taking Amerindian seriously?). I don't know if that means their map of word order changes should be taken with a grain of salt or if it's okay.

Unfortunately, have yet to refind the implicational map of different changes on the way to changing order between S-O-V, such as whether adposition order tends to be an early or late change, shifts in adjective ordering, etc.

[Boşkoventi]

This reminded me of something posted a few years ago, on pretty much the same question.

I have just finished reading Ian Roberts, Diachronic Syntax, and he gives an example of a VSO >SOV change (citing Greenberg), namely in Ethiopian Semitic. Ge'ez, the liturgical language of the Ethiopian Christians, is a representative of the state before the change: it is as VSO as its sisters Hebrew and Arabic. Then, the Ethiopian Semitic languages shifted to verb-final order under the influence of neigbouring non-Semitic SOV languages; later, other word-order parameters changed, one after another, to harmonize with the SOV order. Several languages represent the stages of the change:

Ge'ez: VSO, NA, NGen, Prep
Tigre: SOV, NA, NGen, Prep
Tigrinya: SOV, AN, NGen, Prep
Amharic (14th cent.): SOV, AN, GenN, Prep
Harari: SOV, AN, GenN, Postp

(Tigre and Tigrinya are spoken more northward of Amharic, Harari to the south; the change propagated from south to north.)