ASS UE Possession in Hocąk (Winnebago)

ASS UE  Possession in Hocąk (Winnebago)
Arbeitspapiere des Seminars für Sprachwissenschaft der Universität Erfurt
Nr. 8
Possession in Hocąk (Winnebago)
Problems for a prototype approach
Johannes Helmbrecht
November 2003
ISSN 1612-0612
Seminar für Sprachwissenschaft
der Universität
Arbeitspapiere des Seminars für Sprachwissenschaft der Universität Erfurt
Seminar für Sprachwissenschaft
Philosophische Fakultät
D - 99105 Erfurt
Prof. Dr. Christian Lehmann
© bei den Autoren
ISSN 1612-0612
Introduction ........................................................................................................................4
Seiler's prototype approach to possession ..................................................................6
Typological evidence and predictions ........................................................................8
A brief typology of possessive constructions .............................................................9
Goals and method of the survey ...............................................................................11
Attributive possession in Hocank .....................................................................................12
Nominalized possessive clauses .......................................................................................15
With a lexical human possessor................................................................................15
With a pronominal possessor....................................................................................18
Predicative Possession in Hocank ....................................................................................23
Indirect participation.........................................................................................................25
Benefactive applicative.............................................................................................26
A non-actor is possessor of undergoer .....................................................................28
Reflexive possession.................................................................................................31
Summary and conclusions ................................................................................................34
References ........................................................................................................................37
1. Introduction
Possession is a functional domain of language. Probably all languages have conventionalized
means to express possessive relations between a person, called the possessor, and his or her
belongings, the possessum. Possession subsumes the specific relations a possessor may have
to his possessum within his or her bio-cultural sphere. It is the relation of a human being to his
kinsmen, body parts, his material belongings, and his cultural and intellectual products (cf.
Seiler 1988:82). It is difficult to find one general and unifying notion of possession that
allows subsuming all possessive constructions language-internally as well as crosslinguistically. The reason is that the properties of the possessor and/ or the possessum may
vary significantly resulting in very different types of possession.
The possessor is most frequently a highly individuated human being occupying a high
position on the empathy hierarchy, highly topical (pronominal and nominal) third persons,
and speech act participants such as first and second person (singular) pronouns. Sometimes,
however, languages also allow entities that are very low on the empathy hierarchy - e.g.
inanimate objects - in the same possessive constructions functioning as possessors. These are
then part-whole relationships such as English
(a) the leaves of the tree
(b) the wheels of the car
that represent a type of possession that is not really compatible with the standard ontological
notion of possession as ownership as it is prevailing in western societies and cultures.
The same can be observed with regard to different properties of the possessum. A
kinsman such as father cannot be possessed in the same way as a head, a bicycle or an
emotional property or state such as sadness, cf. the examples in E 2a-d.
(a) my father
(b) my head
(c) my sadness
(d) my bicycle
Kinship relations, and particularly ascending kin relations are given from birth, the possessor
has no influence or control over it. The same holds for the possession of body parts and to a
lesser extent with regard to feelings and other mental states that are rather experienced than
controlled by the possessor. Otherwise, control is an almost defining property of possession in
the sense of ownership. The possessive construction in E 2d comes close to this concept of
ownership in a legal sense which includes full control of the possessor over the possessum.
But even here differences are possible. The possession of the bicycle for instance can be only
temporary (a kind of borrowed ownership) or may be permanent. These different types of
possession are treated equally in English employing the same construction with a possessive
pronoun (possessor) and a noun (possessum). Other languages have different construction for
these types of possession. It will be shown below that Hoca\k has three different constructions
for E 2a-d.
Cognitive approaches to possession tried to deal with this variety of possessive
relations in two different ways. One way is to assume that possession is a cluster of different
cognitive concepts including ownership, kinship and part-whole relations etc. (cf. Langacker
1993), the other way is to assume a common prototype notion explaining the different types
of possessive relations in terms of closeness and deviation from this prototype. The later
approach is more ambitious, because the various concepts of possession are not merely listed
as such, but they are also integrated in a theory of this functional domain. This approach was
developed by Seiler (1983, 1988; see also Lehmann 1998). Another advantage of the
prototype approach à la Seiler is that it makes predictions about markedness properties of
possessive constructions in specific languages that are testable. Since this is the most
elaborate typological approach of possession up to date it is chosen here to serve as the
theoretical background for the description of possession in Hoca\k.
The object language of this survey is Hoca\k a language of the Mississippi Valley
branch of the Siouan family still spoken by around 200 members of the Hoca\k tribe in
Wisconsin. Another denomination of this tribe is Winnebago, a name traditionally used in the
relevant anthropological and linguistic literature. Hoca\k is the self-denomination of the
Wisconsin tribe therefore this name will be used throughout this paper
The goals of the present survey are twofold. First of all, a detailed account of the
expressions of possession in Hoca\k will be given. This survey will later on serve as a part of
the grammatical description of Hoca\k. The second goal is to evaluate the predictions of the
prototype approach with regard to the relevant data from Hoca\k. It will be shown that the
Hoca\k case causes several problems with regard to this approach. Solutions and alternative
views that fit better the Hoca\k data will be sought at the end of this survey (cf. §6). In the
following subsections of this chapter (§1.1-§1.2), a brief summary of Seiler»s prototype
approach will be given together with the typological predictions in terms of markedness
relations that follow from that approach. Subsection (§1.3) gives a summary of different types
of possessive constructions found in the languages showing that there are different structural
levels or domains where possession is expressed. Attributive possession is mainly expressed
within the limits of the noun phrase. Predicative possession is expressed on the clause level.
Two cases can be distinguished here. The possessor and possessum are participants of the
clause and the possessive relation between them is predicated by the main verb of the clause,
or the possessive relation between the participants of the clause is of secondary character
treating the possessor as an indirect participant of the clause. The latter includes what is
termed external possession marking (cf. Payne & Barshi 1999). This typology provides the
structure for the presentation of the possessive constructions in Hoca\k in the subsequent
chapters §§2-5. In subsection §1.4, the goals of the survey are made more explicit and some
remarks on the source of the data and the method are made.
Seiler's prototype approach to possession
Possession is a binary and asymmetric relation between two entities, the possessor and the
possessum. The possessum is a part of the bio-cultural sphere of the possessor and belongs to
the possessor. The prototype notion of possession is solely dependent on the prototypical
properties of the possessor and the possessum. There is no separate relator between them that
creates the relation or designates the nature of the relation. The relevant parameter for the
prototypicality of the possessor is the empathy hierarchy (cf. Lehmann 1998:4), i.e. the
prototypical possessor is EGO (speech act participants in general), less prototypical
possessors are human beings and inanimate objects further below the empathy hierarchy. The
prototypical possessum are relational nouns that have an inherent semantic argument position
for the possessor. Nouns such as kinship terms, body-part terms, and local nouns imply
inherently a relation to a possessor. They establish such a relation to begin with. The noun
father contains the relation «father of someone», the noun head implies the relation «head of
someone», and the noun top implies the local relation «top of something». The prototype
notion of possession is summarized in Figure 1.
Figure 1: Prototypical possessive relation
Deviation from
SAP-----local nouns
Absolute nouns, i.e. non-relational noun, are not able to establish a possessive relation
by themselves, there is no inherent necessity in animate and inanimate entities such as bird,
tree, stone, or car to be possessed by someone. The relation between a possessor and a
possessum in these cases has to be established by additional means, grammatical forms that
indicate (or lexical forms that predicate) the possessive relation. It is particularly the
movement from a prototypical possessum (relational nouns) to a non-relational, absolute noun
as possessum, where the property of control comes into play. If the possessor is high on the
empathy hierarchy, and the possessum deviant from the prototype, i.e. a non-relational noun,
then a high degree of control between possessor and possessum can be observed. In short, if
the possessor (e.g. EGO) owns a house, a car, a bag, a book, etc. he or she controls this
possessive relation, because he or she has control over the possessum. Control means that the
possessor is responsible for and has the ability to obtain such objects, to keep such object, to
use and manipulate such objects, and to give away such objects, i. e. to terminate this
possessive relationship (see also Hagège 1993:93ff; Lehmann 1998:6). Hence, established
possession subsumes our general notion of ownership.
Both parameters, the empathy hierarchy for possessors, and the distinction between
relational vs. non-relational nouns for possessa, are countercurrent to some degree. The higher
the possessor is on the empathy hierarchy, the more he or she is assigned control. But if the
possessum is high on the relationality scale of nouns, the more this control is reduced. The
possession of kinship nouns (in particular ascending blood relations) and body-part nouns (in
particular organs and essential parts for surviving) cannot be controlled because these nouns
determine (inherently) the nature of the relation (cf. Lehmann 1998:7).
The distinction between inherent and established possession (cf. Seiler 1983:5) is
reminiscent to and largely coincides with the traditional distinction between inalienable vs.
alienable nouns1. Alienable nouns usually constitute the open class of this distinction.
Alienable nouns comprise possessed animate and inanimate objects that the possessor can
control in the above sense. Inalienable nouns are usually the smaller mostly closed class of
nouns that comprise kinship terms, body part terms, and local nouns. A finer semantic
distinction that occurs here with regard to kinship relations is the distinction between
necessary vs. optional relationship, e.g. ascending (mother) vs. descending (son) kinship
relations. Body parts may be extended to physical and mental states such as strength and fear
(cf. Lichtenberk 1985:105), or to parts of other items such as branch and handle. In addition,
nominalizations with a verbal noun as possessum such as «his singing» may belong to the
group of inalienable possessive constructions.
However, Seiler made it clear that the distinction between alienable and inalienable
nouns cannot be a categorical distinction in a strict sense, because nouns in a specific
language may sometimes be used in both constructions of alienable and inalienable
possession, and because there is a broad variation among languages with regard to which
nouns count as alienable and which ones as inalienable (cf. Seiler 1988:80; see also Chappell
& McGregor 1996).
Typological evidence and predictions
Empirical evidence for the prototype approach to possession is found in markedness relations
between different possessive constructions language internally as well as cross-linguistically.
There are two central claims with regard to markedness that are associated with this approach.
The first one is that the closer a possessive construction is semantically to the prototypical
possession in the above sense, the less it is marked. And the other way round: the more a
possessive construction in a language deviates from the prototype notion of possession the
more marked it is. This means that inherent possession correlates with un-markedness, and
established possession with markedness. This claim is supposed to be valid crosslinguistically, i.e. in comparison of possessive constructions in different languages, as well as
language internally, comparing alternative possessive constructions in a single language. If
This distinction was first introduced by Lévy-Bruhl (1914) with respect to the expression of possession in
Austronesian languages.
there are two alternative possessive constructions in a language, one closer to the prototype,
i.e. reserved for more inherent possessive relations than the other, the one for inherent
possession is less marked (structurally) than the other. The reason is that possessive
constructions with relational nouns as possessa do not need to indicate the nature of the
relation by additional grammatical forms, since this relation is inherent in the lexical items.
The second claim is related to the first. The prototypical possessive relation is not
marked at all. The construction that corresponds best to the prototypical possession from an
iconic point of view is the juxtaposition of two nominals with a personal pronoun (speech act
participant) as the possessor and a relational noun as possessum. The fact that many languages
utilize indeed this type of construction for the expression of inalienable possession counts as
evidence for the assumption of the prototype in Figure 1 rather than another prototype notion
such as the notion of ownership.
The predictions of the Seilerian prototype approach for possessive constructions in
Hoca\k are the following.
a) Constructions expressing inherent possession, i.e. with inalienable nouns as possessa
are less marked than alienable possessive construction (established possession). This
holds for different types of possessive constructions (different techniques) such as
juxtaposition, constructions with connectives, case marking, and possessive verbs, and
for constructions on different syntactic levels such as the noun phrase and the clause.
b) The second prediction associated with the first is that if possession is expressed by
juxtaposition, it is reserved (not necessarily exclusively) for inalienable possession.
There is an iconic relationship between the semantic type of possession and the
markedness of the constructions
A brief typology of possessive constructions
There are two functional and structural domains where possession is expressed. One is
reference within the structural limits of the noun phrase; the other is predication with its
structural counterpart of the clause. Within the latter two possibilities have to be
distinguished, possession may be predicated by means of a possessive verb with the possessor
as a directly involved participant of the clause, or the possessor is only indirectly involved,
while the possessum is directly affected by the situation predicated by the verb. All three
types of possessive constructions will be illustrated briefly.
There are several types of attributive possession that can be observed in European
languages. The most important are summarized in Figure 2 below. One of the most important
differences between the last two types, genitive attribute and prepositional attribute, is that the
former has the possessor as the attribute of the head noun (possessum), while in the latter the
possessum is an attribute to the head noun (possessor). All types given in Figure 2 may be
used either for inherent or for established possession.
Figure 2: Attributive possession
Construction type
Noun1 √ Noun2
Possessive Pronoun √ Noun
Genitive attribute
Prepositional attribute
Noun1-GEN √ Noun2
Possessor √ Possessum
Noun2 √ PREP Noun1
Possessum √ Possessor
Noun2 √ PREP Noun1
Possessum √ Possessor
das Regierungsauto
«the car of the government»
mein Auto
«my car»
Peter»s car
the car of the government
the man with the red hat
With regard to predicative possession there are in principal two types of possessive
verbs. One predicates the possession to the possessor with the possessor as the subject and the
possessum as the object of the clause. The other ascribes the possessum to the possessor
shifting the possessor in a less central syntactic position as direct object, oblique object, local
adjunct, etc. (cf. Heine 1997:29), cf. Figure 3.
Figure 3: Predicative possession
Construction type
Subject have Direct Object
Possessor √ Possessum
Subject belong to Direct Object/
Local Adjunct
Possessum √ Possessor
He/ John/ father has a car.
The car belongs to him/ John/
The types of predicative possession summarized in Figure 3 contain a verb of
possession that requires possessor and possessum as arguments of the clause, i.e. they are
direct participants of the event. These verbs explicitly state the possessive relation between
the two. These constructions are therefore prototypically used for established possession.
There are, however, other ways to express possession on the clause level. There may
be a kind of secondary possessive relation between two participants of the clause in the sense
that the possessor is indirectly affected by the event because of the possessive relation to the
directly affected participant. These constructions are termed external possession, or possessor
promotion, because the possessor is promoted from a non-argument status (e.g. genitive
attribute) to indirect (dative) or direct object (accusative) function. There are two different
constructions in German that can illustrate this, cf. E 3a-b.
PeterNOM trägt MaryDAT den KofferACC
«Peter carries the suitcase for Mary»
Peters HundNOM hat MaryACC
in die HandLOK gebissen
«Peter»s dog bit Mary in the/her hand.»
The syntactic difference between the constructions in E 3 is that in E 3a, the dative possessor
is optional while the accusative possessor in E 3b is obligatory. In the latter case the
possessum is an optional local adjunct modifying the core predication. It is hypothesized that
both types of constructions are limited to possessa with relational nouns, i.e. inherent
possession (cf. Seiler 1983:41-49; 1988:91; see also Lehmann 1998:10).
Goals and method of the survey
The goals of the present survey are twofold. First of all, a detailed overview will be given on
the expression of possession Hoca\k that will serve as a chapter of the functional part of a
grammar of Hoca\k. Secondly, the predictions of the prototype approach will be tested against
the Hoca\k data. It will be shown that some of the markedness relations to be expected with
regard to inherent and established possession do not hold for Hoca\k.
The data for the survey are taken from already published (e.g. Lipkind 1945, White
Eagle 1988) or unpublished (e.g. Susman 1943, Zeps 1994) sources as well as from my own
field notes.
2. Attributive possession in Hoca\
Hoca\k has no possessive pronouns comparable to English my, your, his, her, etc., no nominal
case marking in general, and no genitive case marker in particular. In addition, there are no
connectives or linkers, i.e. grammatical forms that indicate a possessive relation between two
nominals. The only way to express a possessive relation between two nominals in a noun
phrase in a narrow sense is juxtaposition. Such a construction has a rigidly fixed order of
elements. The determinans, i.e. the modifying noun always precedes the determinatum, i.e.
the head noun of the construction (cf. upper part of Figure 4). The position of the determinans
can only be filled with a noun. Elements of other syntactic categories have to follow the head
noun (cf. Helmbrecht 2002). However, the juxtaposition of two nominals in Hoca\k does not
only express possessive relations, but also part-whole relations, spatial relations, and
specification of a particular nominal concept. The various functions of this construction are
summarized in Figure 4 (lower part).
Figure 4: Juxtaposition of two nominals and their functions in Hoca\k2
[Noun1 √ Noun2 √ DET]
[Determinans √ Determinatum]
[Modifying Noun √ Head Noun √ Determiner]
Possessor √ Possessum
Part-Whole Relation
Spatial Relation
Classification, Specification of a Nominal Concept
The expression of a possessive relation by means of juxtaposition is restricted to cases
where the possessor and the possessum are lexical nominals. If the possessor is a pronoun (a
pronominal affix) of no matter what person category, another construction has to be used
The following abbreviations for the grammatical glosses are used: DET = determiner, DEF = definite article,
INDEF = indefinite article, PN = proper name, DECL = declarative, 1, 2, 3, = first, second, and third person,
SG = singular, PL = plural, INCL = inclusive, EXCL = exclusive, A = actor, U = undergoer, BEN = benefactive,
TR = transitive, INTR = intransitive, APPL.SUPESS = locative applicative superessive, REFL.POSS = reflexive
possession, SAP = speech act participants, ST = part of stem.
obligatorily (see §3.2 below). The possessum on the other hand cannot be a pronoun in
Hoca\k, there are no expressions like French le mien, la mienne «mine». The following series
of examples show that neither the empathy hierarchy √ except with regard the pronoun/noun
distinction √ nor the distinction between relational vs. non-relational nouns have any effect on
the expression of attributive possession in Hoca\k. E 4 is an attributive possessive relation with
a proper name as possessor and a kinship term as possessum. The relation is inherent and
inalienable. The definite article is required.
Petergá hi»a\crá
/Peter-gá hi»a\c- rá/
P.- PN father-DEF
«Peter»s father»
The possessive relation in E 5a is a part-whole relationship with a human possessor
and a body part term as possessum. The possessive relation is inherent and inalienable. The
same holds for the examples in E 5b-c. The whole possessive noun phrase needs to be
specified by a determiner, i.e. the definite article, a demonstrative pronoun, or the indefinite
article. The determiner in this position controls the reference of the whole expression. If there
is a definite article following the possessor (cf. E 5b), then it is the possession of a specific
definite possessor. If the indefinite article follows the possessor (E 5c), it is the possession of
an indefinite or unspecific possessor.
hinú\k his`ja-rá
woman face- DEF
«The woman»s face»
woman-DEF face- DEF
«The face of the (specific) woman»
woman-INDEF face- DEF
«The face of a woman»
The possessive relations in E 6 and
E 7 are alienable. Both contain non-relational nouns as possessa. The possessor in E 6 is a
human being (proper name), the possessum is a (domestic) animal. In
E 7, the possessum is an inanimate object.
Petergá s`u\u\krá
/Peter-ga s`u\u\k-ra/
Peter- PN dog- DEF
«Peter»s dog»
John-ga hiratí-ra
J. -PN car- DEF
«John»s car»
The possessive relation in E 8 includes a body part term as possessum (inseparable,
inalienable) with a non-human possessor. The example in
E 9 represents a part-whole relation with an inanimate object as possessor and an inanimate
object as possessum (separable, alienable). Both possessors in E 8 and
E 9 can be interpreted either as specific or as generic.
wijúk huu-rá
cat leg-DEF
«The leg(s) of the/a cat»
waz`a\tíre hogis-rá
«The wheel(s) of a/the car»
The type of construction employed for the expression of possession in the preceding
examples is also used for the expression of spatial relations. There are numerous local nouns
such as coowé «front part», na\a\ké «back part» rook «inside», hihák «top, surface», and so on
which are used to express the specific local relation of an object vis-à-vis the local region of
another object. The local nouns are the possessum in these constructions. They designate the
local region of the possessor. The possessor functions as the reference point of the
localization, it represents the object with regard to which another one is localized, cf. the
examples in
E 10a-c. The clitic =eja «there» is a local adverb frequently used in these constructions, others
are also possible.
E 10
s`u\u\krá hirarúti coowéja «aks`a\na\
/s`u\u\k-rá hirarúti coowé=eja «aks`a\na\/
dog- DEF car
front.of-there be.lying-DECL
The dog is (in a lying position) in front of the car.
s`u\u\krá hirarúti hihákeja jeena\
/s`u\u\k-rá hirarúti hihák= eja jeena\/
dog- DEF car be.standing-DECL
The dog is (in a standing position) on the top of the car.
s`u\u\krá hirarúti rookéja na\ks`a\na\
/s`u\u\k-rá hirarúti rook= éja na\ks`a\na\/
dog-DEF car
inside.of=there be.sitting-DECL
The dog is (in a sitting position) inside of the car.
The constructions in E 4 to
E 10 are structurally equivalent. The possessor precedes the possessum independent of
the semantic class of the possessor or the relationality of the possessum. From the point of
view of markedness theory, all these constructions are equipollent, i.e. equally marked. The
predictions of the prototype approach are not reflected in these data. Relational nouns (kinship
nouns, body part nouns, local nouns) with an inherent argument slot for the possessor are
possessed structurally in the same way as non-relational nouns. The same type of
juxtaposition can also be used on the level of word formation in order to create new nominal
concepts. The preceding modifying noun specifies the following noun in order to create a
more specific nominal concept. It is a classifying technique; cf. the examples in E 11a-c.
E 11
«tree-leaves/ leaf(s) of a tree»
waminá\k huu
«chair-legs/ legs of a chair»
3. Nominalized possessive clauses
With a lexical human possessor
The juxtaposition of two nouns is a general construction type to express possession and other
binary relations. There are, however, alternative possessive constructions especially for
possessive relations with kinship terms, domestic (pet) animals, and other animate and
inanimate nouns as possessa and lexical human possessors. The alternative constructions are
nominalized variants of possessive predications employing different possessive verbs for
different types of possessa. The nominalized possessive clauses appear in the same syntactic
position as juxtaposed nouns, i.e. in a noun phrase position. This is the reason why they are
not subsumed under attributive possession in §2. These nominalized possessive clauses rather
represent a transition from attributive to predicative possession. The general structure of the
alternative possessive constructions is given in Figure 5.
Figure 5: Structure of nominalized possessive clauses
PROj-PROi-Verb of possession-DET]
If the possessor is a lexical human noun, this construction type competes with the
juxtaposition dealt with in the preceding section. If the possessor is a speech act participant or
a third person, the construction in Figure 5 is the only possible construction on the level of the
noun phrase. In this case, the verb of possession is pronominally inflected for the person
category of the possessor. If the possessum is a multitude of entities, the verb of possession
exhibits another marker, a third plural undergoer marker, to indicate the plurality of the
possessum. This means, person and number of the possessor and the possessum are crossreferenced in the verb of possession utilizing the two different series of pronominal prefixes,
the actor series for the possessor and the undergoer series for the possessum (see below §3.2).
The alternative construction of
E 7 (see above) is given in E 12. The verb haní\ «to own» is a regular (lexical) transitive
verb employed for the possession of alienable entities such as inanimate objects, artifacts,
animals, and so on. It is restricted to human possessors. Part-whole relations with inanimate
possessors are never expressed with this construction. In order to compare the structural
properties of both constructions, see the clauses in E 13a-b.
E 12
John-gá hiratí hani\- rá
J.- PN car own-DEF
«John»s car«
E 13
John-gá hiratí-ra
hacáa- na\
J.- PN car- DEF 1SG.see-DECL
«I see John»s car.»
John-gá hiratí hani\- rá hacáa- na\
J.- PN car own-DEF 1SG.see-DECL
«I see John»s car.»
The example clause in E 13a contains a juxtaposition of two nominals, the example clause in
E 13b contains the nominalized clause with haní\ «to own». Both clauses have the same
translation, but speakers indicate that they prefer the nominalized variant over the juxtaposed
variant. The same constructional pairs exist for possessive constructions with kinship terms
and pet animals (domestic animals), cf. the examples in E 14a-b and
E 15a-b. The a) clauses represent the juxtaposition construction (they are repeated here for
convenience), the b) clauses represent the nominalized alternative construction. In all cases,
the nominalized constructions are preferred by speakers.
E 14
E 15
Petergá s`u\u\krá
/Peter-ga s`u\u\k-ra/
Peter- PN dog- DEF
«Peter»s dog»
Peterga s`ú\u\k ni\i\híra
/Peter-ga s`ú\u\k ni\i\hí- ra/
Peter-PN dog
«Peter»s dog»
Petergá hi»a\crá
/Peter-gá hi»a\c- rá/
P.- PN father-DEF
«Peter»s father»
Peterga hi»á\c hiirá
/Peter-ga hiȇ\c hiiP.- PN father make.kin -DEF
«Peter»s father»
The verbs of possession that are used in the b) clauses are restricted in their usage. The verb
=hii «X makes Y a kin» can only be used with kinship terms or with terms designating close
friends. This verb is homophonous with the causative auxiliary =hii «to cause». There are
reasons to believe that both verbs are historically cognate, and that they should be considered
as different usages of one verb rather than homonyms. The main reason for this analysis is
that the causative verb =hii has an irregular personal inflection, and the possessive verb =hii
shows exactly the same irregularities. A literal translation of
E 15b is then «the one Peter makes (his) father». The three possessive verbs used in
nominalized possessive clauses are summarized in Figure 6.
Figure 6: Verbs of possession in Hoca\k
Verbs of
Lexical human
Kinship (including close social relations
such as friendship)
Pet animals (usually domestic animals
such as dog, cat, horse, etc.)
Inanimate objects (including body parts)
The possessive verb ni\i\hí is used only with pet animals. Usually, pet animals are
domesticated animals such as cats and dogs etc. The boundaries of this class are not clear cut.
Historically, ni\i\hí is presumably a combination of *ni\i\ «to live, living thing» which does not
occur independently in Hoca\k and the causative auxiliary =hii. This verb shows the same
inflectional irregularities as the causative verb =hii.
All three verbs form the same type of possessive construction with lexical human
possessors (and pronominal possessors, see §3.2 below). There is no difference between them
with regard to structural markedness. But they are definitely more marked than the
corresponding juxtaposition constructions. That the nominalized constructions are preferred
over the juxtaposed constructions makes the situation even worse for the prototype approach,
because it is the relational nouns as possessa that have a preferred alternative construction that
belongs to the type of a more established possessive construction. The fact that there are
alternative possessive constructions for kinship terms (inalienable possession) confirms the
particularity of this class of nouns. On the other hand, there are alternative constructions for
pet (domestic) animals and inanimate object including body parts too, and they are
structurally not more marked than the kinship possessions.
With a pronominal possessor
As was indicated in the preceding section, verbs of possession have to be used in nominalized
constructions in case that the possessor is a speech act participant or third person, i.e.
expressed by a pronominal affix. No alternative constructions do exist for this case. The
paradigm of possessive forms with a kinship term as possessum is given in Figure 7. The
paradigms for the possession of pet (domestic) animals and inanimate object are with a
pronominal possessor are given in Figure 8 and Figure 9.
Figure 7: Possessive paradigm for kinship terms
Person category
Rel. Noun : hicu\wí\
hicu\wí\ haará
hicu\wí\ raagá
hicu\wí\ hiirá
hicu\wí\ hi\hirá
hicu\wí\ hi\hiwíra
hicu\wí\ haawíra
hicu\wí\ raawíga
hicu\wí\ hiírera
«aunt (father»s sister)»
«my aunt»
«your aunt»
«his aunt»
«my and your aunt»
«our aunt»
«our aunt»
«your aunt»
«their aunt»
Figure 8: Possessive paradigm for pet (domestic) animals.
Person category
Noun: s`u\u\k
s`u\u\k ni\i\háara
s`u\u\k ni\i\nára /-ga
s`u\u\k ni\i\híra
s`u\u\k ni\í\hira
s`u\u\k ni\i\háwira
s`u\u\k ni\í\hiwira
s`u\u\k ni\i\náwira/-ga
s`u\u\k ni\i\hírera
«my dog»
«your dog»
«his dog»
«our dog»
«our dog»
«our dog»
«your dog»
«their dog»
Figure 9: Possessive paradigm for inanimate object.
Person category
Non-Rel. Noun: waz`a\tíre
waz`a\tíre haaní\na\
waz`a\tíre has`i\ní\na\
waz`a\tíre hanína\\
waz`a\tíre hi\i\nína\\
waz`a\tíre hi\i\ní\wi\na\\
waz`a\tíre haaní\wi\na\
waz`a\tíre has`i\ní\wi\na\
waz`a\tíre haní\i\nera
«my car»
«your car»
«his car»
«our car»
«our car»
«our car»
«your car»
«their car»
If the possessum is a plurality of individuals, the verb of possession is always marked for that,
cf. the example in E 16 for the possession of a kinsman. The same holds for all three verbs of
E 16
hicu\wí\ wahaará
/hicu\wí\ wahaará/
aunt 3PL.U-1SG.A.make.kin-DEF
«my aunts»
Note that the personal inflection of the verb of kinship possession =hii is identical in
form with the causative verb =hii. The kinship term hicu\wí\ «aunt (father»s sister)» has a
variant form that is used for address purposes, cu\wí\ «(my) aunt !». These address forms of
kinship terms √ often simply lacking the initial syllable hi- √ cannot occur in a possessive
construction as exemplified in Figure 7. This seems to be a general rule. Further examples are
given in E 17a-c. The reason for this restriction seems to be that the usage of address terms
usually implies the kinship relation designated by the kinship address term and EGO, the
speaker. Since the possessor can easily be inferred in such a communicative situation, there is
no need to specify it.
E 17
*nu\nú\ haara
* ku\nú\ haara
*heena\ haara
«older sister»
«older sister (address term)»
«my older sister»
«first son»
«first son (address term»
—my first son»
—second son»
«second son (address term)»
—my second son»
There is another kind of variation in the paradigm of kinship possession that may be
rooted in the mutual knowledge of the interlocutors. The common determiner in possessive
constructions with a kinship term is the definite article √ra. However, in the second person
singular and plural the determiner is √ga, a deictic element also used for the indication of
proper names. Lipkind claims that √ga has to be used exclusively in these instances (cf.
Lipkind 1945:31), but Hoca\k speakers gave me forms that show that there is actually a choice
between -ra and -ga in the second person and in the first person inclusive dual form3; -ga is
ungrammatical in all other person categories. Phil Mike indicated to me that this choice has to
do with the mutual knowledge of the kinsman by both interlocutors. The definite article is
used, if the speaker does not know the kinsman (assuming that the hearer knows his or her
kinsmen), but √ga is used when both interlocutors know the person talked about (which is
I am particularly grateful to Henning Garvin helping me to collect the relevant forms here.
more naturally the case if the speaker talks about the kinsman of the hearer). This could also
explain why √ga is not allowed, if the possessum is plural.
The demonstrative suffix √ga is also used with the address forms of kinship terms
indicating the first person as possessor. Lipkind (1945:31) says that all kin terms with initial
hi- take haará «my» in the first person, the few ones without it take solely √ga instead; the
reason is that the shorter forms are terms of address while the hi- forms are terms for
reference. E.g. the form cu\wí\ is the address term corresponding to hicu\wí\ «aunt», hence the
1SG possessive form is cu\wi\-gá which translates literally «that aunt» implying that everybody
knows that she is the aunt of the speaker (EGO), it is a kind of reduced form of speaking, the
address term implies that the person so addressed has the kin relation designated by the term
toward the speaker. It is an effect of the Empathy hierarchy. Shared background knowledge of
the possessor plays an important role here (cf. also Heine 1997:26f). This can also be
interpreted as an instance where the inherent relationality of kin terms leads to a structural
reduction of the expression of possession confirming the prediction of the prototype approach.
From the data presented in this section it is evident that the empathy hierarchy plays a
significant role in the expression of possession. If the possessor is high on this hierarchy i.e. a
SAP or third person, no matter whether the possessum is relational or not, the nominalized
construction has to be used instead of the juxtaposition. The predictions of the prototype
approach would be in this case that the constructions with a possessor high on the empathy
hierarchy, i.e. constructions that are closer to the prototype notion of possession, should be
structurally less marked than the construction with a possessor low on the hierarchy. Exactly
the reverse is the case. The possessive constructions presented in this section are all
structurally more marked than the juxtaposition constructions. They are constructions that
belong rather to the established types of possessive construction than to inherent ones. This
holds in particular for possessa that are kinship nouns, pet (domestic) animal nouns, and body
parts. It holds also for inanimate possessa, but they have never been hypothesized to be
relational nouns. The data presented so far indicate that <the relationality of noun does not
play a role in the expression of possession in Hoca\k.
There is an important constructional similarity between kinship possession and
localization. The causative auxiliary =hii that is used for possession of kinship is also used for
the localization of an object with regard to another object employing local nouns for this
purpose. The local constructions with =hii «make» are illustrated in E 18 and E 19. The
structure is as follows: =hii is the verb of a subordinated clause together with a local noun
figuring as undergoer and possessum, respectively, and the actor as possessor. This
subordinated clause predicates the local position of the argument in the matrix clause, cf.
Figure 10. The examples in E 18 and E 19 illustrate this construction. The difference between
them is that the former has a lexical nominal as actor/ possessor of =hii while the latter has a
pronominal actor/ possessor.
Figure 10: Structure of localization by means of =hii.
Object to be Possessor
Reference point for
Local relational noun
localization (always
(local region of
of Being
E 18
s`u\u\krá Billgá na\a\ké hiieja jeena\
/s`u\u\k-rá Bill-gá na\a\ké
hii= eja jeena\/
dog-DEF Bill-PN back.part make=there be.standing-DECL
«The dog is (standing position) behind Bill»
(lit. «The dog is (standing) where Bill makes the back»)
E 19
s`u\u\krá na\a\ké haaeja jeena\
eja jeena\/
/s`u\u\k-rá na\a\ké
dog-DEF back.part 1SG.A.make=there be.standing-DECL
«The dog is (standing position) behind me»
(lit. «The dog is (standing) where I make the back»).
The parallel between kinship possession and possession of local nouns, i.e.
localization is another piece of evidence that the predictions of the Seilerian prototype
approach are not reflected in the Hoca\k data. According to Seiler, local nouns like kinship
nouns are relational nouns with a inherent semantic argument slot for a possessor. It is these
nouns that do not need relation establishing grammatical means. However, in Hoca\k it seems
to be the other way round. It is precisely this relational classes of nouns that need
constructions for established possession in Hoca\k.
4. Predicative Possession in Hoca\
The verbs of possession listed in Figure 6 above and described in the preceding sections are
the principal means to predicate possession in Hoca\k. Their meaning/ function is exactly the
same, whether they are used as predicates in possessive clauses or in nominalized clauses.
The structural difference is that there is no nominalizing determiner such as a definite article,
or a demonstrative pronoun at the end of the construction. Instead, a declarative suffix or
another sentence final suffix is required in this position. The whole clause is not subordinated
to another clause. Cf. the general pattern in Figure 11. The BELONG TO-type of possessive
predications (cf. Figure 3 in §1.3 above) does not exist in Hoca\k.
Figure 11: Structure of possessive clauses
Some examples will illustrate the structure of a predication of possession employing the three
verbs of possession. In
E 20a-b, possession is expressed with the verb haní\ «to own» for the possession of inanimate
(alienable) objects, and non-pet (non-domestic) animals. E 21a-b contains possessive
predications with the verb for pet (domestic) animals, and E 22a-b shows predicative
possession with kinship terms. All a) clauses have lexical possessors; all b) clauses have
pronominal possessors. It is obvious that they are structurally exactly parallel. They differ
from the nominalized possessive clauses in chapter §3 in several respects. First, the verbs
have a declarative suffix or some other sentence-final suffix instead of the nominalizing
determiner; secondly they can be modified more freely with regard to tense, mode and aspect,
while this is not possible with the nominalized counterparts. And thirdly, the possessum in the
predications is much more flexible with regard to determination and modification than the
possessum in the nominalized possessive clause, e.g. the possessum in the predication can
have the indefinite article, the one in the nominalized possessive clause not.
E 20
E 21
E 22
Billgá waz`a\tíreizà\ hani\í\na\
/Bill-gá waz`a\tíre-izà\
hani\- na\/
Bill- PN carINDEF own-DECL
«Bill has/ owns a car.»
waz`a\tíreizà\ haaní\i\na\
ha- haní\- na\/
«I have/ own a car»
Billgá wijukíz`a\ ni\i\híina\
/Bill-gá wijuk-íz`a\
Bill-PN cat- INDEF
«Bill has a cat.»
wijukíz`a\ ni\i\háana\
cat- INDEF
«I have/ own a cat.»
Billgá hisu\kíz`a\
Bill-PN make.kin-DECL
«Bill has a younger brother.»
haana\ 1SG.A.make.kin-DECL
«I have a younger brother.»
Other types of possessive predication such as «X is my Kin» or «This is my Kin, Pet,
etc.» are formed in Hoca\k with very similar constructions; cf. the examples in E 23, E 24, and
E 25.
E 23
Billgá hi»á\c haána\
/Bill-gá hi»á\c haána\/
Bill-PN father 1SG.make.kin-DECL
«Bill is my father» (lit. «I make Bill (my) father»)
E 24
hi»á\c haána\
DEM.PROX father 1SG.A.have.kin-DECL
«This is my father.»
E 25
hicakóro haána\
DEM.PROX friend 1SG.A.have.kin-DECL
«This is my friend»
Finally, there is at least one kinship verb4 in Hoca\k that predicates the type of kinship
relation instead of predicating possession as such with certain selectional restrictions. This
type of possessive construction is very different from the ones dealt with up to this point. The
kinship verb is ki\i\nú\p «be sibling (including cousins)» and kiikí\nu\p «be siblings to each other»,
compare the examples in E 26a-b. No verb of possession is used in these cases.
E 26
heena\ga kunu\gá «ee ki\i\nú\ps`a\na\
ki\i\nú\p- s`a\na\/
second.son-PN first.son-PN be.sibling-DECL
«Heenaga is a brother to Kunu\ga.» (White Eagle 1988:62)
«We are cousins.» (White Eagle 1988:61)
The Hoca\k data illustrating the predication of possession do not reflect the predicted
asymmetries with regard to the hypothesized prototype of possession. The usage of the three
verbs of possession is mutual exclusive which leads to a classification of possible possessa:
kinship nouns, pet (domestic) animals, and the rest. The rest includes body parts, inanimate
objects, artifacts, but not local nouns. Part-whole relations cannot be predicated by =hii, haní\,
and ni\i\hí too, because they do not allow non-human possessors. The constructions with these
three verbs of possession are strictly parallel; there is no markedness relation observable
among them that are iconically in line with the closeness or deviation from the prototype.
5. Indirect participation
There are two different types of secondary possession marking, i.e. external possession, in
Hoca\k. One is based on a systematic polysemy of a benefactive application (cf. §5.1 and 5.2
below), the other one may be termed reflexive possession (cf. §5.3). Both types of external
possession marking have in common that the possessive relations between two participants in
the clause are in a way secondary, mostly because one of the participants is only an indirect
participant (cf. Lehmann et al. 2000). The possessive relations are not predicated by means of
verbs of possession. The participants of the relevant clauses are primarily participants of the
Kinship terms are verbs in some Australian and North American Indian languages (cf. Evans 2000)
action predicated by the main verb. It is hypothesized that external possession is restricted to
relational nouns such as kinship terms and body part nouns, cf. Seiler (1988:91) and Lehmann
(1998:10). The prototype approach would predict for these construction types that a)
possessive relations close to the prototype are more easily expressed by external possession,
and that b) possessive relations close to the prototype are less marked in external possession
constructions than possessive relations deviant from the prototype. Both types of external
possession in Hoca\k will be introduced and discussed in the subsequent section in particular
with regard to these predictions.
Benefactive applicative
Hoca\k has a very productive benefactive applicative gi- that increases the argument frame of
the verb by one. Semantically, the new argument is sometimes a sympatheticus (cf. Lehmann
et al. 2000:12ff), i.e. a participant which is indirectly affected by the event designated by the
verb. Such an indirect participant is often expressed by what has been termed traditionally
dativus ethicus. Since there is no case marking in Hoca\k, this term would be inappropriate.
Examples of a sympatheticus participant introduced by the benefactive application are given
in E 27 and E 28. Most often, the newly introduced participant has a benefactive relation to
the state of affairs designated by the verb; cf. the examples in E 29 and E 30.
E 27
Sympatheticus interpretation of the gi- argument
hicawí\-gi- t'é
'widower, (lit. his wife died to him and he was affected by that)'
wife- BEN-die
E 28
Sympatheticus interpretation of the gi- argument
'to laugh, to smile'
'to laugh at so.'
'I laughed at him'
/hagi- híks`a- na\/
1SG.A-BEN-laugh- DECL
There are two meaning nuances of the benefactive role in Hoca\k. Dependent on the semantics
of the verb and the discourse context, the action designated by the verb can be interpreted as
being instigated in favor of (or 'in behalf of') the benefactive participant or as being instigated
instead of the benefactive participant5. Accordingly, the derived form in E 29a could also be
glossed 'to cry instead of someone' which is pragmatically unlikely in this case. There are,
however, verbs such as gima\ní\ 'to walk for someone' which are more likely to be interpreted
as 'to walk instead of someone'. Pronominal marking of the arguments is √ with the exception
of the 3pl √ire √ always prefixal. The benefactive argument is cross-referenced by a
pronominal prefix of the undergoer series morphologically preceding the gi- formative, cf. E
E 29
gi- with intransitive active verb
'to cry for someone'
'they cried for me'
/hi\gi- g`ák- irena\/
The gi-derivation is possible with intransitive active (cf. E 29) and with transitive
verbs (cf. E 30). Transitive verbs with gi- become three place verbs with the actor (actor
series of pronominal affixes) and the benefactive (undergoer series of pronominal affixes)
usually marked pronominally. The patient is third person singular in the overwhelming
number of cases and hence remains zero. However, it is in principle possible to elicit forms
with three pronouns, if the patient is a speech act participant showing that a preference for
either primary object marking or secondary object marking (cf. Dryer 1986) is not
grammaticalized in Hoca\k.
E 30
gi- with transitive verbs
'to hold sth.'
'to hold sth. for someone'
ma\a\shíroji\na\ hani\gítukoss`a\na\
ha- ni\gí- tukoss`a\na\/
hammer- DEF ST-1->2- BEN-1SG.A.hold-DECL
'I hold the hammer for you.'
The argument structure of the benefactive derivations with intransitive active and transitive
verbs is schematically summarized in Figure 12.
Both meaning nuances are formally distinguished in Lakhóta. There are two different so-called dative series of
pronouns, the first one meaning 'on behalf of' the second one meaning 'instead of', cf. Boas & Deloria 1941:87;
Helmbrecht 1998:169-174.
Figure 12: The basic argument structure of benefactive derivations I.
However, the derivation with gi- is systematically ambiguous in Hoca\k in that the benefactive
participant can be interpreted in most cases as the possessor of the patient. This type of
external possessor marking is most obvious with intransitive inactive verbs which will
therefore be treated below, but is present in other verb classes too.
A nonnon-actor is possessor of undergoer
The type of external possession marking to be described here is mostly given preference in
discourse over possessive constructions of the nominalized possessive clause type (cf. §3).
The benefactive argument introduced in the argument frame of the verb can be systematically
interpreted as the possessor of the undergoer, i.e. either possessor of the undergoer of the
intransitive (inactive) verb or the possessor of the undergoer of the transitive verb. The first
possibility is illustrated in E 31a-c, where the intransitive inactive verb s`is`ré is derived by the
benefactive applicative gi-. The resulting verb can have two interpretations, in the first one
'sth. is broken for him' the added argument is interpreted as being affected by the breaking, in
the second one, the benefactive argument is interpreted as the possessor of the undergoer 'his
is broken'.
E 31
gi-derivation with intransitive inactive verbs
'to break, to be broken'
'sth. is broken for him'/ 'his is broken'
waz`a\tírera gi-s`is`réena\
«His car is broken»
gi- s`is`ré- na\/
carDEF BEN-broken-DECL
The benefactive applicative derives verbs that have two undergoer arguments. The first and
primary undergoer is the «car» in E 31c. It is not cross-referenced on the verb, because it is a
Information of the morphological structure of the Hoca\k verb, in particular with regard to the position of the
pronominal affixes in relation to the other affixes can be found in Susman (1943:104-110), and Helmbrecht (in
3SG noun phrase. If it were plural, it would get a marker on the verb. The second undergoer is
the benefactive argument. This undergoer is only indirectly affected by the event designated
by the verb. It is usually interpreted as the possessor of the primary undergoer. The same
constructions are possible also with body parts as possessum. Kinship terms are not possible
in this example for semantic/ pragmatic reasons, but in other verbs they are.
With transitive verbs, two interpretations are systematically available. The added
argument is either the benefactive of the action, or the possessor of the patient of the action.
This is illustrated in E 32a-c. The action is conducted in favor of the benefactive who in turn
may be taken as the possessor of the patient.
E 32
gi-derivation with transitive verbs
'to find sth.'
'to find sth. for someone'
waz`a\tírera hi\i\gí'eena\
'he found the car for me'/ 'he found my car'
/waz`a\tíre-ra hi- hi\gí- 'e- na\/
The systematic interpretation of the benefactive as possessor is, however, not possible
with gi-derivations of intransitive active verbs. Here, only the benefactive meaning is
available, cf. the example in E 33a-c. If the possessor of the 'younger brother' is someone else
haara 'my kin' has to be inflected for another person category. If haara 'my kin' is dropped in E
33c the translation would be 'the younger brother cried for me'. The argument introduced by
gi- cannot be interpreted as possessor.
E 33
gi-derivation with intransitive active verbs
'to cry'
'to cry for someone, to yell for someone'
hiisú\k haará hi\gig`áks`a\na\
'my younger brother cried for me'
haará hi\gi- g`ák- s`a\na\/ my 1sg.U-Ben-cry- Decl
The possibilities of external possessor marking by means of the benefactive applicative gi- are
summarized in Figure 13.
Figure 13: The argument structure of benefactive derivations II.
Verb(intransitive inactive)
Verb(intransitive active)
The possessor interpretation of the benefactive in the first and the third line of Figure 13 can
be blocked by an explicit possessor marking of the undergoer noun phrase, i.e. by means of a
nominalized possessor construction (cf. section §3). An example is given in E 34. The 2SG
possessor of 'the car' contrasts with the 1SG benefactive in the verb hence the 1SG
benefactive can no longer be interpreted as the possessor of the patient noun phrase.
E 34
waz`a\tírera has`i\ní\na\ hi\i\gí'eena\
gí- 'e- na\/
/waz`a\tíre-ra has`i\ní\na\ hi- hi\car- DEF your
'he found your car for me/ *he found my car'
The possessors in such a secondary possessive relation (external possessive construction) are
usually high in the empathy hierarchy. They are human beings (all person categories) referred
to by means of pronominal affixes, or lexical noun phrases. Inanimate possessors
(benefactives) are not possible. The possessum, on the other hand, can be almost everything,
from kinship nouns, body parts, human beings, to inanimate objects. Part-whole relations and
local relation cannot be expressed with this type of construction. The data presented show that
the relationality of nouns does not play a significant role with regard to markedness of these
possessive constructions.
The possessive constructions by means of nominalized possessive clauses (see chapter
§3) and the type of external possessive marking by means of the benefactive application are to
some extent in competition with each other. This can be demonstrated with the clauses in E 35
and E 36. The former one uses the benefactive applicative strategy, the latter the nominalized
possessive clause strategy. If speakers have a choice they always prefer the benefactive (giderivation) strategy.
E 35
waz`a\tírera ni\i\gítuz`aana\
ni\i-\ gí- tuz`aa- na\/
carDEF 1->2-BEN- DECL
‘I wash your car’ (or ‘I wash the car for you’)
E 36
waz`a\tírera has`i\ní\na\ tuz`áana\
/waz`a\tíre-ra has`i\ní\-ra
tuz`áa- na\/
‘I wash your car’
Reflexive possession
The second type of external possession in Hoca\k is marked by the prefixes kara-/ k- 'one's
own'. The choice between both forms that are functionally equivalent is morphologically
determined by the conjugation type of the verb. Verbs that require the second conjugation
take k-, verbs that require the first conjugation take kara-. In both cases the result is a verb of
the first conjugation.
The prefix kara-/ k- can be used only with transitive verbs. In general, it designates
that the transitive actor possesses the transitive undergoer. Since kara-/ k- are possible only
with transitive verbs they represent a highly reliable test for transitivity (syntactic valence) of
verbs in the Hoca\k lexicon. There are, however, also cases where kara-/ k- seems to have
itself a transitivizing function, cf. E 37a-b.
E 37
Transitivizing function of kara-/ k(a)
na\a\zí` \
'to stand, to stand up'
na\a\-kára-z`i\ 'to stand up for his own'
An example of the typical function of kara-/ k- is given in E 38a-c. The transitive verb gigú\s
'to teach someone' changes its meaning to 'to teach one's own' with kara-/ k-, cf. E 38b. It is
indicated that the undergoer of the transitive verb belongs to the actor. There is almost no
restriction with regard to the semantic class of the possessum. Animate and inanimate object
can be possessed, body parts, kinship relations and close social relations such as friendship.
What has be observed several times in other domains of possessive marking holds also for
reflexive possession, the relationality of the possessum does not play a role.
E 38
A possesses U in transitive verbs: kara-/ k(a)
'to teach someone, to council someone'
'to teach one's own'
ni\i\kjá\k waakáragigu\ss`a\na\
'I taught my children'
/ni\i\kjá\k wa- hakáragigu\s- s`a\na\/
The reflexive possession marker kara-/ k- can be used with plain non-derived transitive verbs,
but also with verbs that have received their second argument by way of some derivational
processes. If verbs are transitivized by means of one of the instrumental prefixes kara-/ kindicates that the patient is possessed by the actor, cf. E 39a-b.
E 39
'to break sth. by striking'
boo-kára-xux 'to break his own by striking'
If the verb has received a new (second) argument by means of one of the locative applicatives,
kara-/ k- refers to this newly introduced argument, i.e. it indicates that the actor owns the
place with regard to which the action is instigated. An example with the verb hat'á\p 'to jump
on sth.' is given in E 40a-c. The clause in E 40c illustrates that the place (locative) of the
action which is the new argument is the possessum of the second person actor.
E 40
'to jump on sth.'
'to jump on one's own'
waarúcra harakárat'a\ps`a\na\
harakárat'a\p- s`a\na\/
You jumped on your table.
Ambiguities and uncertainties about the interpretation of kara-/ k- arise with
ditransitve verbs and verbs that have received a third argument by means of some derivation.
Both undergoer arguments are marked on the verb (if they are not 3sg) by pronominal
prefixes of the undergoer series. Since there are two pronominal affixes of the same undergoer
series, some uncertainty about the correct assignment of the semantic roles patient and
recipient to the two undergoer pronouns arise. I assume that in these admittedly extremely
rare cases in actual discourse animacy plays the decisive role, i.e. the human participant is
assigned the recipient role, since this role is prototypically human, and consequently, the
inanimate participant is assigned the patient role, since transitive patients are prototypically
inanimate. The same uncertainty extends now to the usage of the reflexive possessor marking
with kara-/ k-. The transitive patient may be interpreted as being possessed by the actor as
well as the recipient, cf. the examples in E 41a-c. The verb hok'ú\ 'to give sth. to someone' has
three arguments, the actor, patient and recipient, the latter two are expressed by pronouns of
the undergoer series. As was already said it may be uncertain which one of the two undergoer
participants is patient and recipient respectively, but this is not a problem here. The semantics
of the two undergoer arguments makes it clear that the "candies" are the patient (the object of
giving) and the "children" are the recipient. A similar uncertainty holds with regard to kara-/
k-, either the patient or the recipient may be interpreted as possessed by the actor.
E 41
kara-/ k- in ditransitve verbs
'to give sth. to someone'
ho-kara-k'ú\ 'to give sth. (his own) to someone (his own)'
ni\i\kjá\gra taní\zu` kirikíris` wookárak'u\u\na\
/ni\i\kjá\k-ra tani\zú` kirikíris`
wa- ho- kárak'u\- na\/
child-DEF sugar striped (=candy) 3PL.U-ST-REFL.POSS-give-DECL
«He gave the candy to his children.»/
«He gave his candies to the children.»
The same ambiguity exists also for the verb hoz`ú 'to put sth. in sth.' which has also three
arguments. The expressions in E 42b and E 42c are from different Hoca\k speakers, the former
preferred the goal argument (i.e. the container) to be interpreted as possessed, the latter
preferred the transitive patient (i.e. the papers) to be interpreted as possessed. A schematic
summary of reflexive possessive marking in Hoca\k is given then in Figure 14.
E 42
'to put sth. in sth.'
'I put them in my (container)'
karaz`u- na\/
/wa- ho- há
waagáxra waxupá\na\ waákaraz`uuna\
waxupá\- ra wa- ho- hákaraz`u- na\/
paper- DEF suitcase-DEF
«I put my papers in the suitcase.»
Figure 14: Reflexive possession.
The same uncertainties arise in verbs which have an additional benefactive argument, here the
transitive patient as well as the benefactive may be interpreted as possessed. All these cases
are, however, rare and are rather problematic instances of elicitation than real problems of
grammatical structure.
6. Summary and conclusions
The various possessive constructions in Hoca\k can be divided into two different classes, the
ones that require contextual inference in order to be interpreted as indicating a possessive
relation, and the ones that explicitly mark possession between two nominals. The
juxtaposition and the benefactive applicative belong to the former class of possessive
constructions, the verbs of possession (nominalized possessive clauses and predication of
possession) and the reflexive possession belong to the latter class, cf. the summary in Figure
Figure 15: Relevance of the prototypical features of possession for possessive constructions
Mode of
of possessive
Not Relevant
of Possessum
position on
the Empathy
- only lexical
(no SAP and
third person
- human and
Predication of
Benefactive Reflexive
applicative possession
Not Relevant
depend on
restrictions of
the verbs of
- lexical (always
on verb) and
- only human
Not Relevant
depend on
restrictions of
the verbs of
- lexical (always
on verb) and
- only human
- only
- only
Seiler»s central hypothesis with regard to possession is that there are certain semantic
classes of nouns that are relational in nature. They have an inherent semantic argument slot
for a possessor and that»s the reason why they are able to establish a possessive relation to
another noun without requiring relation establishing grammatical material (cf. Seiler
1983:12ff, 45ff). Since juxtaposition (see also Seiler 1983:14f) and the benefactive
applicative (see particularly the chapter on ∆possessor promotion∆ in Seiler 1983:45f) does
not mark possession explicitly, i.e. presuppose inference of the possessive relation, they are to
be expected to show some restrictions or other effects with regard to the semantic type of the
possessum such as kin terms, and body part terms. This is not the case in Hoca\k. Relational as
well as non-relational noun are allowed as possessa. Juxtaposition in Hoca\k resembles much
more what Seiler said about genitive constructions in Indo-European languages (cf. Seiler
1983:39ff). It is a general construction for many different functions. Possession is only one
among these. Another one is the spatial localization of objects. There are no restrictions with
regard to inherent relational and non-relational nouns in juxtapositions expressing possessive
relations. In addition, there is no such restrictions observable with respect to the benefactive
The other three techniques to express possession have in common that they mark this
relation explicitly, either by means of the selectional restrictions of the verbs of possession, or
by means of a special possessive marker. Seiler»s theory would predict that these
constructions are in particular appropriate to express alienable possession, since they are not
built upon the inherent relationality of inalienable nouns. This, however, is not the case. The
three verbs of possession classify nouns according to kinship nouns, domestic animals nouns,
and rest, but this has nothing to do with the relationality of these nouns. Relevant is here only
the selectional restriction of the verb of possession. Similarly, the reflexive possessive marker
is neutral with respect to the semantic types of the possessum. It can be concluded here too
that the distinction between inherently relational vs. non-relational nouns is irrelevant in these
constructions too. These observations lead to the conclusion that one part of the prototype
notion of possession with regard to the possessum is largely irrelevant in Hoca\k and that there
is no such thing as grammaticalized inherent relational nouns in Hoca\k. This part of the
prototype notion has to be given up for the Hoca\k data.
The other part of the prototype notion of possession does play a role in possessive
constructions in Hoca\k. The possessor has to be a human being, i.e. a third person pronoun or
speech act participant, except for the juxtaposition. Here, lexical possessors of all semantic
classes are allowed leading to constructions that no longer express possessive relations. The
higher the possessor is on the empathy hierarchy the more control he has over the possessum.
The reverse force, the reduction of control of a possessor high on the empathy hierarchy by
means of inherent relational possessa such as body parts, and kinship relations, cannot be
observed in Hoca\k. Possession seems to be solely construed along the control parameter in
Hoca\k. Evidence for this hypothesis is provided by the fact that the etymological source for
two of the verbs of possession =hii «to.make.kin» and ni\i\hí «» is the
causative verb =hii «X causes Y», a verb that predicated maximal control of one participant
over the other. The causative schema =hii «X causes Y» serves as the conceptual source as
well as the source construction for the grammaticalization of kinship possession and domestic
(pet) animal possession in Hoca\k. It should be added to Heine»s list of possible event schemas
that serve as source for the grammaticalization of the predication of possession in the
languages of the world (cf. Heine 1997:47). Further evidence is that the same causative verb
is also used for the expression of local relations in case that the reference point of the
localized object is a human being or speech act participant that has control over the spatial
reference point e.g. because the local region is a part of his or her body.
7. References
Baron, I. et al. (eds.) 2001: Dimensions of possession. Amsterdam & Philadelphia: J.
Boas, Franz & Ella Deloria. 1941. Dakota Grammar. Washington, D.C.: Government Printing
Chappell, Hilary & William McGregor (eds.) 1996. The grammar of inalienability: a
typological perspective on body part terms and the part-whole relation. Berlin:
Mouton de Gruyter.
Dryer, Matthew S. 1986: „Primary objects, secondary objects, and antidative.“ Language
Journal of the Linguistic Society of America 62: 808-845.
Evans, Nicholas 2000: „Kinship verbs.“ In: Vogel, Petra M. & Comrie, Bernard (eds.),
Approaches to the typology of word classes. Berlin & New York: Mouton de Gruyter
(EALT, 23); 103-172.
Hagège, Claude 1993. The language builder. An essay on the human signature in linguistic
morphpogenesis. Amsterdam: Benjamins.
Heine, Bernd 1997: Possession. Cognitive sources, forces and grammaticalization.
Cambridge: Cambridge University Press.
Helmbrecht, Johannes 1998: Universalität und Vagheit semantischer Funktionen.
Untersuchungen zum funktionalen Zusammenhang morphosyntaktischer und
kognitiver Kategorien der Sprache. Unterschleissheim: LINCOM Europa (Edition
Linguistik, 26).
Helmbrecht, Johannes 2002: „Nouns and verbs in Hoca\k“. In: International Journal of
American Linguistics IJAL 68,1:1-27.
Helmbrecht, Johannes (in prep.): A grammar of Hocank
Langacker, Ronald W. 1993: „Reference-point constructions.“ Cognitive Linguistics 4: 1-38.
Lehmann, Christian & Yong-Min Shin & Elisabeth Verhoeven 2000: Direkte und indirekte
Partizipation. Zur Typologie der sprachlichen Repraesentation konzeptueller
Relationen. Muenchen: LINCOM
Lehmann, Christian 1998: Possession in Yucatec Maya. Structures - functions - typology.
Unterschleissheim: LINCOM Europa (LINCOM Studies in Native American
Linguistics, 4).
Lévy-Bruhl, L. 1914: „L’expression de la possession dans les langues mélanésiennes.“
Mémoires de la Société de Linguistique 19(2): 96-104.
Lipkind, William 1945: Winnebago grammar. New York; King’s Crown Press.
Payne, Doris L. & Barshi, Immanuel (eds.) 1999: External possession. Amsterdam &
Philadelphia: J. Benjamins (Typological Studies in Language, 39).
Seiler, Hansjakob 1983: POSSESSION as an operational dimension of language. Tübingen:
G. Narr (LUS, 2).
Seiler, Hansjakob 2001: „The operational basis of possession.“ Baron, I. et al. (eds.),
Dimensions of possession. Amsterdam & Philadelphia: J. Benjamins; 27-40.
Susman, Amelia 1943: The accentual system of Winnebago. New York: Columbia University
[Ph.D. dissertation].
White Eagle, Josephine P. 1988. A lexical study of Winnebago. Lexicon Project Working
Papers 26. Cambridge, MASS.: Centrer for Cognitve Science, MIT.
Zeps, Valdis J. 1994. Dictionary of Winnebago. Manuscript.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF