Can WALS do this?

[Terra]

I want to compare two different maps, (http://wals.info/feature/85A and http://wals.info/feature/87A , specifically), to see if there's a correlation between their values. Does a language having prepositions mean that it'll probably be noun-adjective instead of adjective-noun? I mean, I could do it by hand, but it'd take so long...

[Aurora Rossa]

I believe you use the feature combine option to get this: http://wals.info/feature/combined/85A/87A

[Terra]

That's exactly what I want! Thanks.

I feel bad about making this in L&L instead of Ephemera now...

[Ser]

Well, let's see... Out of the 466 languages identified as having prepositions, 351 languages or 75.32% are identified as noun-adjective. There's quite a strong correlation.

[spats]

Well, let's see... Out of the 466 languages identified as having prepositions, 351 languages or 75.32% are identified as noun-adjective. There's quite a strong correlation.

Not only that, but among languages that have a specific order of noun and adjective and have strictly pre- or post-positions, the distribution is extremely areal.

[Aurora Rossa]

That reminds me of some questions of linguistic typology that I have been wondering, but have not seen addressed on the WALS site. I have been trying to figure out what proportion of languages mark person on the verb with prefixes compared to those that mark verbal person with suffixes. My general impression so far, although based on casual surveys of well-known languages, suggests that more languages mark person on verbs with suffixes that prefixes. Suffixed marking occurs everywhere from Indo-European to Eskimo-Aleut while prefixed person marking seems much less common, though I see it in Nahuatl and many other American languages.

I am also wondering whether the position of person marking correlates with word order and direction of branching. My intuition suggests that left-branching SOV languages would strongly prefer pronominal prefixes on verbs because the markers would have evolved from pronouns that became clitics and attached to the following verb. It would seem odd for the pronouns to jump to the other end of the verb stem and become suffixes when the language otherwise has well-established left branching syntax. Many of the SOV languages I have seen do use person suffixes rather than prefixes, though, so it seems unclear whether my intuition holds any water.

[Shihali]

Again, no research, but Bantu languages use prefixes for marking person/gender on the verb, and they're heavily right-branching and tend to SVO. Not sure how much evidence Bantu is collectively.

[cromulant]

I have been trying to figure out what proportion of languages mark person on the verb with prefixes compared to those that mark verbal person with suffixes.

The order of the A and P relative to each other is not independent of whether the two are prefixes, suffixes or occur on opposite sides of the verb. AP order is particularly favoured when the two person markers are on opposite sides of the verb; of the 84 languages which have the two person markers on opposite sides of the verb and display a unique order of the A and P, in 75% the A precedes the P. This tends to be the case in SVO and some verb-initial languages. This explains the frequency of AP order in Southeast Asia, the Pacific and Africa. Among the 36 languages in which the A and P are both prefixes, AP and PA orders are very evenly distributed: 19 languages have AP order and 17 PA. This is also more or less the case in the 34 languages in which the A and P are both suffixes: 15 have AP order and 19 have PA order. Fused markers occur somewhat more frequently in prefixal position (12 languages) than in suffixal position (8 languages). This also holds for the languages in the sample manifesting alternative orderings of the A and P; 13 (65%) are prefixing.

[spats]

Regarding word order and suffixing vs. prefixing (specifically in SOV languages):

I suspect it's a case of how the language ended up in that word order to begin with. Word order is always changing around. Just in IE, you have a proto-language that was SOV having daughter languages that are everything from VSO to SVO to SOV. But most of these languages that kept verb conjugation kept personal agreement suffixes (we'll ignore French for now, because French is weird).

In other words, the real question is not, "Is there a correspondence between word order and prefixing/suffixing agreement?", but rather, "What the heck was the word order when the languages started requiring agreement?"

[chris_notts]

In other words, the real question is not, "Is there a correspondence between word order and prefixing/suffixing agreement?", but rather, "What the heck was the word order when the languages started requiring agreement?"

Although it's not even as straightforward as that, because there are a number of examples where the choice of prefixing vs suffixing doesn't align with what we know about historical word order. E.g, where the entire family is verb-final, but some languages have developed agreement suffixes rather than prefixes. Since agreement controllers (especially S/A) tend to be topical and therefore not urgent to state (=can be said last), and since suffixes tend to be preferred anyway cross-linguistically, and since in many verb-final languages the pre-verbal position is the focus position, this might not be surprising, but the point is that if the most common word order dictated whether you got prefixes or suffixes then verb-final languages ought to have more agreement prefixes than they actually tend to.

Today's morphology is not necessarily yesterday's syntax.

[Terra]

Another question: Is there anything on WALS that deals with compounding? Looking at the features page, searching "compound" turns up nothing. What are the patterns of compounding? Are certain types more common than others? (noun+noun vs adj+noun vs verb+noun, etc) Do the patterns usually follow the same order as adjectives and phrases? How do compounds deal with suffixes, if they language is strongly head-first?

[cromulant]

Another question: Is there anything on WALS that deals with compounding?

No.

[Rin]

Something I noticed in Japanese, is that, although it's SOV, pronouns sometimes fall at the end of the sentence as a sort of afterthought:

Dare ga watashi no ringo tabeta ka? (Who ate my apple?)

Aa! Tabeta, watashi. Suman. (Oh! Ate it, I. Sorry.)

Anyway, it seems like a plausible way in which originally separate pronouns might have fused to the ends of verbs to form suffixes in otherwise SOV languages.

[Torco]

it's hard to establish correlation only by comparison of proportions, but yeah, it's likely significant. I've been trying to turn the WALS database into a spreadsheet that's statistically workable, perhaps in the interest of finding underlying factors that account for a large part of the variation of languages, or to check cluster analysis versus language families, or stuff like that... but I've found I'm too lazy to do so of late.

[Richard W]

Although it's not even as straightforward as that, because there are a number of examples where the choice of prefixing vs suffixing doesn't align with what we know about historical word order. E.g, where the entire family is verb-final, but some languages have developed agreement suffixes rather than prefixes.

Better than that, VSO Middle Egyptian with subject suffixes gave rise to Coptic with subject prefixes. What happened here is that the relevant part of the sequence was aux-verb+subject main-verb, and that, in the simple cases, wore down to subject+main-verb. (Some inflectional prefixes do remain before the subject in some tenses and verbs.)

[brandrinn]

Something I noticed in Japanese, is that, although it's SOV, pronouns sometimes fall at the end of the sentence as a sort of afterthought:

Dare ga watashi no ringo tabeta ka? (Who ate my apple?)

Aa! Tabeta, watashi. Suman. (Oh! Ate it, I. Sorry.)

Anyway, it seems like a plausible way in which originally separate pronouns might have fused to the ends of verbs to form suffixes in otherwise SOV languages.

That's called clefting, and it only works because it's contrary to the "standard" order. As for whether it can lead to verb suffixes, I don't know.