Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: VOS

From:Matt Pearson <jmpearson@...>
Date:Tuesday, April 25, 2000, 21:19
>I read somewhere that Mayan is VOS. I suppose this really means it's >ergative, doesn't it?
Well, "Mayan" is not a language, but a family of languages. Some of them are VOS, but others (I think) are VSO, or even SVO. The only two Mayan languages I know anything about are Tzotzil and Tz'utujil. Tzotzil is pretty vigorously VOS. Tz'utujil appears to have VSO basic order, but deviations from this order (SVO, OVS) appear to be common. As far as I know, all of the Mayan languages are ergative, but I believe that this is orthogonal to word order. For example, even though agreement marking on Tzotzil verbs follows an erg/abs pattern, transitive subjects and intransitive subjects are grouped together with respect to things like particle placement.
>What other VOS languages are there? Are they ergative or non-ergative?
Many of the Polynesian languages (and some other Oceanic Austronesian languages as well) appear to be VOS as well. I'm not sure whether the ones which have been claimed to be VOS are also ergative, or whether there's no correlation. Malagasy, the language I work on, is also claimed to be VOS, but I don't think so. I think it's basically VSO, but with a clause- final topic position which is often (but not necessarily) occupied by the structural subject. As far as whether Malagasy is ergative or not, that's a dicey question. With focus/pivot languages like Malagasy and Tagalog, it's notoriously difficult to tell whether the basic case-marking pattern is erg/abs or nom/acc; the languages appear to share properties of both (and neither). It's an interesting question whether there's a correlation between word order and ergativity. I know of at least one linguist who's argued for the following generalisation: "All ergative languages are verb-peripheral (i.e. verb-initial or verb-final)." So, the claim is that there are no SVO (or OVS) languages which are ergative. I can think of a possible partial counterexample (namely Georgian), but otherwise this generalisation seems to hold as far as I'm aware. However, the generalisation doesn't go the other way. That is, not all verb-peripheral languages are ergative. This is obviously true of SOV languages; I think it's also true of VSO and VOS languages. Matt.