[Lingtyp] languages lacking a verb for 'give'

David Gil gil at shh.mpg.de
Thu Jan 27 17:22:28 UTC 2022

```Matthew, Daniel, Russell and all,

I'm with Daniel on this one; in my 2017 paper I discuss precisely this
issue, namely, how do we define the meanings of the relevant forms, and
whether one of them has a meaning that can appropriately be
characterized as "give".

Abstracting away from word order, and assuming minimal morphology, what
we have is a situation of the abstract form

A X P Y R

where A = Agent, P = Patient, R = Recipient, a translation into our
contact language along the lines of A gave P to R, and an analytical
question:  What are the meanings of X and Y (and should one of them be
assigned the meaning "give")? Typically, both X and Y are poly- or
macro-functional, and either of the two can occur without the other,
resulting in constructions whose translational equivalents into our
contact language do not involve "give". X and Y don't wear their glosses
on their sleeves, as it were; the answer to our analytical question will
depend on an in-depth language-specific analysis of the various
functions of each of the two elements, X and Y.

In Roon (SHWNG, Austronesian), there are two "give" constructions.  For
the one illustrated in (3) of my 2017 paper, I argue that the basic
meaning of X (/ve/) is "do", while that of Y (/fa/) is an oblique
marker.  For the one illustrated in (10), X = Y (both have the form
/ve/) and here too I argue that both mean "do".  Under an alternative
more splitting analysis, the second element might be characterized as a
"different" /ve/ associated with an allative function; however, the
first /ve/ would still be vague between a variety of different
functions, and, as I argue there, is most appropriately analyzed as
meaning "do".

The second part of the 2017 paper looks at a number of other languages
of the region; while the patterns of poly-/macro-functionality differ
from one language to another, in many of the cases, neither of the two
elements, X and Y, would seem to warrant the assignation of a meaning
"give".

David

(3)/MusaivyepipifaRiksoni/

Musa:pers<3sg.anim>*ve*moneyoblRikson:pers

'Musa gave money to Rikson.'

(10)/MusaivyepipiveRiksoni/

Musa:pers<3sg.anim>vemoney*ve*Rikson:pers

'Musa gave money to Rikson.'

On 27/01/2022 18:16, Russell Barlow wrote:
> Daniel,
>
> I'm not sure I follow. Presumably we'd be relying on the translations
> of the *arguments*, not of the verbs, when figuring out the semantic
> roles of each verbal object. So, in examples of the sort that you,
> Eline, and I (maybe others) have provided, we see something like:
>
> "boy take apple, give girl"
>
> ... to mean something like "the boy gives the girl an apple". I share
> your unease about considering the second verb in such cases to be
> "give" in the English sense. But I don't think there's any issue in
> figuring out which NP is semantically the theme and which NP is
> semantically the recipient. We could ignore the glosses of the verbs,
> and the semantic roles of the participants would still be clear:
>
> verb1 apple, verb2 girl
>
> Provided we know that "apple" and "girl" are both the objects of the
> verbs they follow, then we could say that the object of verb1 is a
> Theme, and the object of verb2 is a Recipient. I think what Matthew
> and I are both interested in finding is something like:
>
> boy verb1 apple, girl verb2 apple
>
> ... something like "the boy proffered the apple; the girl took the
> apple".
>
> Best,
> Russell
>
> Russell Barlow
> Postdoctoral Researcher
> Max Planck Institute for Evolutionary Anthropology
> Department of Linguistic and Cultural Evolution
> russell_barlow at eva.mpg.de
>
>> On 01/27/2022 4:41 PM Daniel Ross <djross3 at gmail.com> wrote:
>>
>>
>> Matthew,
>>
>> Relying on translation equivalents in this case is not clear. If the
>> verb "give" exclusively appears in SVCs (as is claimed for some
>> languages), then it's only half of the lexical meaning of English
>> /give/. We could translate it as something else, e.g. some active
>> equivalent of 'receive' (several verbs like 'supply (the army)' or
>> 'load (the truck)' can be used in this way, although they're flexible
>> including ditransitive usage like 'give' at least with prepositional
>> arguments).
>>
>> Russell, I have the same uncertainty about your question: how do we
>> know what a "Theme" argument is, without relying on translation? In
>> many languages with SVCs of this type, there is no case marking (in
>> fact, SVCs are said by some to function as case markers), so I don't
>> know what other evidence there would be aside from the translation of
>> the verb itself, which only in the construction as a whole means 'give'.
>>
>> I assume that the etymology of the verbs in these constructions is
>> not 'give': that is, it's not the case that an original, full lexical
>> verb 'give' taking three arguments was reduced to taking two
>> arguments and expanded into this construction, but that some other
>> verb grammaticalized into that function. There's been a lot written
>> about these kinds of usage, but I'm not sure about the best sources
>> to recommend for that specific etymological question. I do think it
>> would be relevant to the original question, though.
>>
>> Daniel
>>
>>
>> On Thu, Jan 27, 2022 at 7:03 AM Matthew Dryer <dryer at buffalo.edu> wrote:
>>
>>     Daniel,
>>
>>     This does not seem to be what my colleague is looking for since
>>     the second verb still arguably means ‘give’.
>>
>>     Matthew
>>
>>     *From: *Daniel Ross <djross3 at gmail.com>
>>     *Date: *Wednesday, January 26, 2022 at 11:27 PM
>>     *To: *Matthew Dryer <dryer at buffalo.edu>
>>     *Cc: *"lingtyp at listserv.linguistlist.org"
>>     <lingtyp at listserv.linguistlist.org>
>>     *Subject: *Re: [Lingtyp] languages lacking a verb for 'give'
>>
>>     Dear Matthew,
>>
>>     This is a common pattern for languages with serial verb
>>     constructions, along the lines of "take book give him", etc.
>>     There has been a lot written about the lack of argument structure
>>     in these languages (some claiming that three arguments are not
>>     possible in some languages), and that SVCs can supplement that
>>     argument structure (and possibly a small inventory of verbs,
>>     according to some sources). I'm not as confident in some of the
>>     is widespread among many of these languages (I know I've seen
>>     explicit claims for West Africa and creoles, and probably
>>     elsewhere). At the same time, it is not clear that these
>>     languages, strictly speaking, lack a lexical verb "give", since
>>     one of the verbs in this construction can be translated as such,
>>     although it is used with another verb (often 'take') to
>>     supplement it for the full argument structure. Other patterns are
>>     found too, and probably various other lexical verbs are used in a
>>     function like 'give', so it becomes a question of lexical
>>     translation. (This more generally is related to patterns of verbs
>>     in SVCs developing into prepositions.)
>>
>>     I'm sorry I don't immediately have any specific
>>     languages/references in mind, but let me know if you'd like me to
>>     try to find some. I know that Sebba 1987 discusses this in some
>>     detail, and here's one example:
>>
>>     ɔde sekaŋ no mãã me
>>     he-take knife the give-PAST me
>>     'S/he gave me the knife' [originally from Christaller 1875: 118]
>>
>>     Sebba, Mark. 1987. The syntax of serial verbs: an investigation
>>     into serialisation in Sranan and other languages. Amsterdam: John
>>     Benjamins. https://doi.org/10.1075/cll.2
>>
>>     (Tangential note: SVCs like this are generally considered
>>     /monoclausal/, by a variety of metrics, so I wouldn't call this
>>     "two analytic clauses", although the effect is the same. My
>>     dissertation thoroughly reviews the issue of monoclausality:
>>     https://doi.org/10.5281/zenodo.5546425
>>     -- but I don't discuss this specific question about 'give'.)
>>
>>     Finally, one extra comment, which is probably not what your
>>     colleague is after, is that there are some languages where the
>>     lexical verb 'give' is (at least in some cases) a zero root or
>>     null morpheme, i.e. indicated by lack of phonological content
>>     plus other inflectional morphology. This is discussed for some
>>     PNG languages here:
>>
>>
>>     Daniel
>>
>>     On Wed, Jan 26, 2022 at 7:43 PM Matthew Dryer <dryer at buffalo.edu>
>>     wrote:
>>
>>         I am sending this query on behalf of a colleague.
>>
>>         He wants to know whether anyone knows of a language that
>>         lacks a "give" type verb and would express something like "I
>>         gave him the book" instead as something like "I presented the
>>         book (to him) and he took it". That is, is there a language
>>         that can only express a give-type concept with two more
>>         analytic clauses?
>>
>>         Matthew Dryer
>>
>>         _______________________________________________
>>         Lingtyp mailing list
>>         Lingtyp at listserv.linguistlist.org
>>         http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>>
>> _______________________________________________
>> Lingtyp mailing list
>> Lingtyp at listserv.linguistlist.org
>> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp

--
David Gil

Senior Scientist (Associate)
Department of Linguistic and Cultural Evolution
Max Planck Institute for Evolutionary Anthropology
Deutscher Platz 6, Leipzig, 04103, Germany

Email:gil at shh.mpg.de
Mobile Phone (Israel): +972-526117713
Mobile Phone (Indonesia): +62-81344082091
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20220127/9ab0659f/attachment.htm>
```