LINGUIST List 11.1166

Mon May 22 2000

Sum: "Give" and Person Suppletion

Editor for this issue: Scott Fults <>


  1. Bernard Comrie, 'give' and person suppletion

Message 1: 'give' and person suppletion

Date: Sun, 21 May 2000 13:02:24 +0100
From: Bernard Comrie <>
Subject: 'give' and person suppletion

About a month ago I posted a query asking for information on person
suppletion for the recipient of 'give'. The following are the results
of replies received so far, supplemented by some of my own further
research, arranged by language (and including some material from my
original posting). In some cases I have only an indication that the
language is to be checked further.

Ao (Kuki-Naga-Chin), more specifically the Mongsen dialect: The verb
'give' has distinct forms, in the imperative only, determined by the
grammatical person of the recipient: _kh-ang33_ 'give (to first
person)' versus _khi-ang33_ 'give to third person'. (Information from
Alec Coupe from research in progress, who points out that no forms
with second person recipient, i.e. 'give to yourself' are attested,
and that there may be occasional exceptions.)

Ik (Kuliak).

Ipili (New Guinea Highlands) distinguishes _gi_ 'give me/you' and
_mai_ 'give to 3rd person'; the language's morphological inflection
collapses second and third persons in the non-singular, but the
suppletion of 'give' remains. (Information from Frances Ingemann.)

Japanese _yaru/ageru_ are used when the gift moves away from the
speaker, the verbs _kureru/kudasaru_ when the gift moves towards the
speaker. (The difference between the two verbs in each pair concerns
the relative social status of giver and recipient.) For instance, if
someone gives something to me, I will use _kureru/kudasaru_. If the
mayor of my town gives something to my father, then I will use
_kureru/kudasaru_, as the gift is coming towards me. But if the mayor
of my town gives something to the mayor of another town, I will use
_ageru/yaru_, since the gift is going away from me. (Extensive

Malayalam (Dravidian) distinguishes _koTukkuka_ 'give (to third
person)' from _taruka/tarika_ 'give (to first or second
person)'--these seem just to be distinct roots. (Data from R.E. Asher
and T.C. Kumari, Malayalam; London, 1997: Routledge, p.348.)

Manambu, a Papuan language from Ndu family (closely related to Iatmul,
Wosera, Boiken etc). The same stem is used, but the forms are quite
idiosyncratic; 'give' to 1st or 2nd person is _kwatay/kwatiy_, and to
3rd person is _kwiy_. (Information from Alexandra Aikhenvald.)

Maori, where the verb _ho-_ 'give' obligatorily requires a directional
particle, to give either _homai_ 'give to the speaker' or _hoatu_
'give away from the speaker'. Note that the particles seem to indicate
direction towards or away from the speaker, not necessarily that the
speaker is/is not the recipient (which thus makes this the closest
typological parallel to Japanese). (Cf. Winifred Bauer, Maori; London,
1993: Routledge, pp. 470-476.)

Nandi (Nilotic) has _ke:-ko:n_ 'to give' with first and second person
recipients (which are marked with suffixes); when used alone it is
conjugated as though it were derived with the Ventive ("motion
towards") suffix _-u_. _ki:-ka:-ci_ 'to give to' is used with 3rd
person recipients and is conjugated as though it contained the dative
suffix _-ci_. (Cf. Creider and Creider 1989.)

Nile Nubian languages (according to Greenberg 1993).

Old Basque. (Ongoing research by Gontzal Aldai.)

Rossell (aka Yele, "East Papuan") shows suppletion between 1/2 and 3
person forms. (Ongoing research by Stephen Levinson.)

Saliba (Oceanic) distinguishes _le_ 'give (to first or second person)'
from _mose-i_ 'give (to third person)'. The roots are
distinct. Interestingly, there is a further difference (not found in
the Malayalam, Tsez, or Japanese cases), namely the two roots have
different argument structures: _le_ is syntactically a monotransitive
verb having giver and gift as arguments; _mose-i_, which includes an
applicative suffix, is a ditransitive verb, with three more specific
syntactic frame possibilities. For details, see Anna Margetts,
Valence and Transitivity in Saliba; Nijmegen, 1999: MPI Series in
Psycholinguistics, pp. 300-308.

Tsez (NE Caucasian) distinguishes _teL_ 'give (to third person)' from
_neL_ 'give (to first or second person)' (where _L_ represents a
voiceless lateral affricate). Etymologically at least, the initial
consonants seem to be deictic prefixes, but this is not a productive
pattern in Tsez. Simple and derived transitive verbs have different
imperative formations in Tsez, and different varieties of Tsez attest
both the expected simple-verb imperative _teL-o/neL-o_ and the
expected derived-verb imperative _teL/neL_.

Waskia, a Madang language, has /asi-/ 'give to 1sg', /kisi-/ 'give to
2sg'; /tuiy- ~ tuw-/ 'give to 3sg/; /idi- 'give to pl'. For the 1sg,
2sg and pl verbs one can see the outlines of an etymology consisting
of a monosyllabic stem and a clitic pronoun, but this is not so clear
that one can reconstruct earlier forms. Such suppletion seems to be
found in a number of other Madang languages. (Information from Malcolm
Ross. Note that in Amele, following John R. Roberts, Amele; Beckenham,
1987: Croom Helm, the verb 'give' seems to be best analyzed as a zero
root taking indirect object affixes, the resulting forms being
morphologically transparent.)

Some more general conclusions can be drawn:

The phenomenon seems to be sporadic across many parts of the world,
and while there may be some areal concentrations (e.g. in New Guinea),
it is not unusual to find languages having it while closely related
languages don't (e.g. Malayalam vs. Tamil; Tsez vs. Hunzib).

Suppletion for first/second person versus third person seems
particularly widespread, but at least in New Guinea richer suppletion
is found, perhaps reflecting an origin in frozen pronominal affixes
and a zero or minimal stem. Japanese and Maori have a deictic
opposition that is not literally to be interpreted as grammatical
person of the recipient, and that distinguishes towards versus away
from speaker.

While some forms clearly involve suppletion, others have formal
relations that either synchronically (Maori directional particles) or
diachronically (e.g. Tsez and some Papuan languages) involve

The phenomenon is found even in languages that have no inflectional
person agreement with the recipient (e.g. Tsez), or indeed no person
agreement at all (e.g. Malayalam).

Incidentally, I plan to continue research into this topic.

I am grateful to the following who responded to my query: Alexandra Y.
Aikhenvald, Gontzal Aldai, Alec Coupe, Chet Creider, Frances Ingemann,
Petek Kurtboke, Jaakko Leino, Stephen Levinson, Sebastian Loebner,
John Newman, Barbara Pizziconi, Malcolm Ross.

Additional references suggested to me are:

Creider, Chet A., and Jane Tapsubei Creider. 1989. A Grammar of Nandi.
Hamburg: Helmut Buske.

Greenberg, Joseph H. 1993. 'The second person is rightly so called' In
Mushira Eid and Gregory Iverson (eds.), Principles and Prediction: The
Analysis of Natural Language. Papers in Honor of Gerald Sanders, 9-23.
Amsterdam: John Benjamins.

Newman, John. 1996. Give: a Cognitive Linguistic Study. Cognitive
Linguistics Research 7. Berlin: Mouton de Gruyter.

Prof. Dr. Bernard Comrie Director, Department of Linguistics

Max Planck Institute for Evolutionary Anthropology
Inselstrasse 22 tel +49 341 99 52 301
D-04103 Leipzig NEW 01/00 tel secretary +49 341 99 52 315
Germany fax +49 341 99 52 119

Home page:

A copy of all incoming e-mail is fowarded to my secretary. If
you do not wish your message to be read other than by me, please
put "private" in the subject box.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue