"where?" = relativizer

claude-hagege claude-hagege at WANADOO.FR
Thu Oct 22 13:02:45 UTC 2009

Dear all, 

    Peter Arkadiev's query is of course relevant only for languages that have a bona fide relativizer (= a form blending (morphologically, semantically or both) a subordinator + a substitute of the relativized (mostly nominal) element). Such languages amount to no more, probably, than 25%-30% of all languages (among which European languages), whereas 22% use a nominalizing or participial strategy, 15% a subordination strategy (by means of 1) a special (sometimes optional) morpheme, or 2) a verbal affix or inflectional element, or 3) a special personal paradigm (as in Salishan languages) ), and 18%  use a genitive morpheme, as in Mandarin Chinese, or simply a fronted relative clause with no marker, as in Japanese (I blush recalling that this is treated in detail in La structure des langues, Paris, PUF 2001: 61-62). 
    The formal homology between "where?" and "who?" is not uncommon, as has been stressed, yesterday and today, by several participants to this forum, but it is not quite widespread either. To the languages already mentioned, I would add Lao and its very closely related neighbour Thai, which have thi [falling tone](nai [midtone]) for "where?" and thi [falling tone] for "who, which". 
    Not infrequently, a formal identity, rather than involving "where?" and "who, which", involves in fact the relativizer and an interrogative pronoun, as in Lithuanian Yiddish vos. Or the relativizer and the interrogative pronoun are distinguished by the fact that one is derived from the other: Georgian  has vin "who?" and ra "what?", while it has vints and rats for the relative pronouns "who" and "which" respectively (although Georgian more commonly uses another relative pronoun, romeli (+ case suffixes) ).             
    Furthermore, the relative pronoun may share the same root as (rather than "where?") a relative locative adverb , as in such Indo-Aryan languages as Hindi: relativizer = jo, and "in which" = jahân. 

All best 

Claude Hagège, Collège de France
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20091022/c10b3865/attachment.htm>

More information about the Lingtyp mailing list