Navigation – Plan du site

AccueilLes actes4Wednesday, January 23rd 2013‎Sustainable environments and new ...The language of ‘rights’ in the a...

Wednesday, January 23rd 2013‎
Sustainable environments and new economies in Aboriginal Australia

The language of ‘rights’ in the analysis of Aboriginal property relations

Ian Keen

Texte intégral

The analysis of Aboriginal “property”

1In an earlier article (Keen 2010) I looked at the ways in which Aboriginal “property” was (mis)understood by English speaking observers from quite early in the colonial period through to the present. A concomitant of the process of the colonisation of Australia and appropriation of Aboriginal land has been the misrepresentation of Aboriginal concepts and processes of relations to country. Observers interpreted property, especially in relation to land and waters, as if it were akin to English land tenure. I concluded as follows:

…[C]ommentators interpreted Aboriginal possessions not only in terms of the all-encompassing concept of ‘property’, including personal possessions, land, wives (in one view) and ‘incorporeal’ things, but also projecting English social structure onto Aboriginal social relations. The tribe is the equivalent of the nation, with its common language, territory (which the tribe defends) and body of custom. It is divided into family groups, each with an ‘eState’. In several accounts, individuals inherit property in land from the father post mortem and pass it down to sons or to other close relatives in the absence of sons. The ‘family’ is thus constituted as a succession of individual landholders—reminiscent of aristocratic families holding land in entail. Radcliffe-Brown’s ‘horde’ with its property in land is in effect an expanded family. Certain class elements are read into Aboriginal social relations in interpretations of prominent individuals as ‘chiefs’ (for example, Dawson 1881). (The later ‘clan’ model introduces an element with a Scottish but not an English equivalent.)

2While both personal and real property were said to be held exclusively—especially land at the level of tribe—they are not described in terms of commodities. The language of commodities creeps in, however, when it comes to the exchange of moveable items, described as ‘trade’ or ‘barter’. After all, ‘trade’ was a synonym for ‘commerce’ in eighteenth-century England and contrasted with the landed interest.

3[Thus] Aboriginal society—at least in its dimension of ‘property’—is depicted through the nineteenth and early twentieth centuries as a primitive form of English society. Terms such as ‘tribe’ (from Latin tribus) and (later) ‘clan’ (adopted into Gaelic from Latin planta) give the structure its exotic, primitive character—‘tribe’ used initially to label elements of Hebrew society and clan to denote Scottish kin groups. It is only when we come to Warner[‘s ethnography] (1937) that ‘property’ begins to be qualified and the possessions of ‘clans’ are given a special status, more closely reflecting the kinds of distinctions made by Aboriginal people themselves.

4The language of ‘rights’ has dominated anthropological discussions of property, including relations to land tenure (Keen 2011), and discussions of Aboriginal relations to country in the land rights era following the enactment of the Aboriginal Land Rights (Northern Territory) Act (1976), and again following the Mabo case and the subsequent Native Title legislation (Bartlett 1993).

5The language of rights developed during the evolution of the market economy through the Early Modern and industrial eras in Britain and other European countries, and is specific to a particular social formation, albeit widely exported through colonial expansion. Concepts of ‘rights’ and ‘property’ are contested in legal studies and anthropology, however. It is rather extraordinary, then, that concepts whose meanings have been taken to be so problematic are used by anthropologists as if they were transparent instruments for translating concepts in other cultures. It is also strange that they appear to have worked so well, although their success may be illusory in the sense that their use has given rise to distortions and misunderstandings. (Keen 2011)

6We may go further to argue that the very language of “rights” may be misleading when used to translate concepts of possession or property cross-culturally, and that it may bring about a distortion of the ways in which people use language to constitute relations of possession (Keen in press).

“Rights” as an anthropological meta-language

7For the most part anthropologists have construed the concept of “property” in terms of rights, obligations and interests. Indeed, the idea of a “bundle of rights” has been proposed as the foundation for a general approach to comparison (von Benda-Beckmann, von Benda-Beckmann, and Wiber 2006; Keen in press).

Von Benda-Beckmann et al. (2006) propose that the ‘bundle of rights’ concept can be used to express the totality of property rights and obligations, or in relation to a ‘master category bundle’ such as private ownership, or particular property objects such as land, or in relation to valued resources held by a particular person or social unit. Hann (2007: 308) comments that the ‘rights and obligations associated with land, the key factor of production, and with concepts of ownership, both collective and private, can be unpacked with the help of the “bundle” metaphor’ (p.308).

8By the early twentieth century property was no longer regarded in legal theory as ‘absolute dominion’, as in Blackstone’s ideal, or as sui generis, but as a disaggregated ‘bundle of rights’. The modern understanding of property as ‘disaggregated’ is traced to the writings of Hohfeld in the early twentieth century (Hohfeld 1913, 1917; see also Gordon 1995:96), although scholars attribute the expression ‘bundle of rights’ to Maine (Hann 1998).

9Nancy Williams study of Yolngu relations to country is a good example of the “rights” approach to Aboriginal relations to country; she writes of “tenure”, “interests” and “rights” held by members of “corporate” groups and their relatives among Yolngu people.

10The use of the terminology of “rights” as a universal metalanguage is questionable, however, for it is very much culturally specific. The concept of “rights” interposes a claimed or attributed capacity or power as an imaginary object between a possessor and a possession. A right to or in something has the appearance of a kind of possession in itself – something a person “has” (Alchian and Demsetz 1973). This mediating concept enables the things that possessors can do in relation to a possession to be differentiated and divided. Where “rights” and related concepts are used as part of a metalanguage these implications of “rights” may be quite different from, and distort, the indigenous constructions that are constitutive of possession relations (Keen in press).

11Aboriginal languages seem not to have terms translateable as “rights” – how then do people talk about what anthropologists construe as rights? Few ethnographies give examples of Indigenous discourse about possession, but rather translate into anthropologese. A rich source of Indigenous discourse on possession of country is Nicholas Evans’ report written for a Kaiadilt land claim under the Northern Territory Land Rights Act (Evans 1998; Keen in press). Drawing on Evans’ work I look first at Kayardild concepts of possession. (The forthcoming article compares Kaiadilt discourse of possession with Navaho and Southern Song China.)

Kayardild concepts of possession

12Kayardild, a Tangkic language spoken until recently by Kaiadilt people of Bentinck Island in the Gulf of Carpentaria in northern Australia (Evans 1995, 1998), has few verbs of possession and no general verb of possession akin to “have”. The verb karrngija, “keep, keep hold of”, frequently used in discussion of country, can mean “look after”, “guard”, and “be responsible for” something in one’s possession (Evans 1995:55). Yolngu use the word dja:ga (“look after”) in a similar way, as well as ngayathama, “hold”.

13Possession in Kayardild is indicated by possessive pronouns and grammatical cases, which denote a variety of possession relations from inalienable to alienable possession. The simplest construction is apposition – what Evans calls the NOM-NOM frame (Evans 1995, 1998).

14The relation of the person to parts of the body requires apposition or the NOM-NOM frame, and having something on one’s person uses the associative and perhaps proprietive cases. As for potentially alienable possession of objects (“ownership”), the genitive and ablative cases serve for attributive expressions such as “x’s country”, and the utilative and proprietive cases are used for predicative expressions as in “country-having person” (dulkuru dangkaa, countryPROP manNOM). In another mode of indicating possession of country, individuals take the name of a focal site (Evans 1998:96).

Kayardild possessors

15Tables 2 and 3 provide examples of possessive pronouns in Kayardild: “my country”, “his country”, “my sea”, “my spear”; and of collective possession as in “the country of my mother’s mothers”, “other people’s food”, “country-owning group”.

16There are no Kaiadilt land-owning groups akin to “clans”, for owners of the same “country” (dulk) may be connected to it in a variety of ways including spirit conception, birth, and cognatic descent (Evans 1995). Nevertheless, owners of the same country are thought of as a single “country-owning group” (dulkuru jardi) rather than a mere collection of individuals (Evans 1998:53)

Kayardild possessions

17The Kayardild language has no general concept of “property” (Nicholas Evans pers. com.). The same may be said for Yolngu dialects. Kayardild possessions in the examples are identified by nominals in NPs that are the direct objects of verbs such as “trespass” and “steal”. The focus here is on “countries” (dulk) – discrete tracts of land and waters, including streams, sea, islands, and fish traps (Evans 1998).

18Each dulk consists of an area around a cluster of named sites, some regarded as mystically dangerous. Like other Aboriginal groups, Kaiadilt people believe that ancestral beings created the land as well as animal and bird life, and humans and their customs, before being transformed into features of the landscape or “story places” that “bear witness” to these original world-creating exploits (Evans 1995:21). Thus ancestral creation underwrites the inalienable possession of dulk. Many dulk are named after single focal sites, and defined by boundaries or edges where one country “cuts off”, giving way to another, and by lists of places along the coast, paths or creeks, given in explanations for visitors or neophytes (Evans 1998:63-5, 72). In this way the category dulk encompasses topographical discourse, the identity of places, and totemic narratives, under a concept of a unitary object possessed by a group or an individual.

19The enactment of possession relations includes both the use of possessions and transactions characteristic of particular possession regimes. Those to do with Kaiadilt country include succession or inheritance of land and waters, negotiations over the use of other groups’ countries, and action taken in the event of breaches of norms (Evans 1998).

Deontological entailments of possession relations

20Possession relations are characterised by deontic norms governing possession, and give it its character.

21Deontic norms are rules or principles that create obligations and grant permissions and powers (Keen in press). Some are established by statutes and regulations, while others are informal and customary. Scholars often associate norms with modal operators such as “ought” and “must”, but they can be expressed in a variety of ways (Finnemore and Sikking 1998; Hechter and Opp 2001). Deontic norms (Beller 2008) govern possession relations and give them their content; these are the deontological entailments of the connection between the possessor and the thing possessed. The bases of possession are recognized, and deontic norms governing possession are current and enforced, within intersecting and overlapping jural communities of varying scope and power, from the family to the State (Goodenough 1965; Finnemore and Sikking 1998; Hechter, Opp and Wippler 2001; Sripada and Stich 2006). Means are required within jural communities to change norms (Ensminger and Knight 1997), to adjudicate disputes, and to enforce norms governing possession (e.g. Fehr 2004).

22Deontic norms are current and enforced within jural communities of various kinds. Jural communities in Australian Aboriginal societies are typically kin-based networks within which people recruited support to protect their interests and to take redressive action against perceived wrongdoers, now subject of course to structures associated with the nation-State. This structure of “self-help” (Berndt 1965) lay within an overarching regional consensus as to the nature of ancestral law, whose foundation is attributed to creator ancestors (Keen 2004).

23Norms governing possession of country in Kayardild have to do with access to and inheritance of land, waters and their resources (Evans 1998).How do people express norms about possession and possessions in Kayadild?

The expression of norms in Kayardild

24Verbal and nominal case are the main means for expressing deontic modality in Kayardild, especially the potential, hortative, desiderative and oblique cases, together with the counterfactual particle maraka (Evans 1995:264, 428). The examples, which are mainly about land and waters, include:

  • Strong prohibition, with the potential inflection in the negative;

  • Weak prohibition, with the hortative inflection;

  • Expressions of obligation, with the counterfactual particle maraka;

  • Conditional constructions, with a deontic modal (in the imperative mood);

  • Normative Statements linking wrong action to ancestral law in declarative sentences and quoted questions;

  • Lexicalisation of wrong action, such as the verb “steal” (Evans 1998).

25Notice the absence of any terms translated as “rights”.

A brief comparison

26It may be of interest briefly to compare Kayardild constructs of possession with the other cases in the longer paper.

27Kayardild lacks a general verb of possession and relies on wide variety of grammatical cases to express subtle distinctions in varieties of possession relations, as well as possessive pronouns and apposition.

28Navajo also lacks a general possessive verb akin to “own” and “have”, but employs possessive pronominal affixes, what is call a “genitive/BE construction” translateable as “my x exists” (I have a car = my car exists), and certain idioms such as “I pack a gun” in which verbs are specific to object type.

29The expression of possession in literary Chinese of the Southern Song includes a “have” verb (you) and possessive pronouns and particles for kin relations, personal attributes and possessions. It might be tempting to relate this to the Song market economy; “have” verbs, however, exist in a wide variety of languages, including some Australian Aboriginal ones (McGregor 2001), which were not associated with market economies before the British colonisation of Australia.

30The three languages use possessive articles, pronouns and pronominal affixes to identify possessors, and these extend to kinship relations, parts of the body, and personal attributes.

31Kayardild and Navajo have no categories of possessions that can be translated as “property” in general, although Navaho has terms translateable as “junk” or “stuff’ and “that which I carry about”. Literary Chinese has a rich variety of property terms, concomitant with the role of the market in Southern Song society.

32The linguistic means for expressing norms governing possessions vary between the three case studies. Kayardild expresses deontic modality primarily by means of grammatical case as we have seen. Like Kayardild, Navajo lacks modal auxiliary verbs (like “should” or “must”), and uses tense and various verbal idioms to express obligation, permission and prohibition (such as “there is space for” indicating permission). Literary Chinese expresses deontic modality by means of verbal auxiliaries and a particle. There is, however, a striking similarity in the range of expression of norms about possession in Kayardild and literary Chinese, for example in the use of expressions such as “according to law” and “that’s the law”. (I could find no examples of Navaho expression of norms.)

33The ontological correlates of possession also vary. The “consubstantial” relations between people and country typical of Aboriginal cosmologies (Bagshaw 1998; Magowan 2001) contrast with the commodification of land in the Southern Song. The former render land and waters inalienable, as the relation between Pr and Pm becomes akin to the relation of a person to a part of the body and/or to a kin relationship. This kind of ontology also pertained in Navajo culture in relation at least to Diné traditional territory as a whole, reflected in the use-ownership of Navajo land. Land in Southern Song China was measured, bought and sold, as well as inherited. (In the longer paper I briefly related these ontologies to Strathern’s critique of the use of the term “property” for the PNG Highlands.)


34To return to the issue of rights, although the language of rights purports to be able to accommodate variation in “property” relations cross-culturally, expressed as diverse “bundles of right” (von Benda-Beckman, von Benda-Beckman and Wiber 2006), an implication of this talk (and see Keen in press) is that the approach misses the diversity of conceptualisations and expressions of possession relations. The language of rights and obligations tries to capture what people can, cannot, must and must not do in relation to things possessed, according to shared norms and laws. No word or expression that can be translated as a “right” in the legal sense has emerged in Kayardild, however (or the other languages of the comparative study). To express deontic norms governing possession in terms of rights and obligations would require processes of inference and construction. Used as a metalanguage, the language of rights obscures much of the language and culture of possession relations, and by imposing the alien concept of abstract rights as mediating possessions, it may well distort what it describes. Here I have tried to show how what become construed as rights and obligations in anthropological and legal discourse are expressed in Kayardild.

35There is an aspect of Aboriginal relations to country that adumbrates the “bundle of rights” idea, however, and that is the distribution of relations to country among groups and relatives of groups (including especially uterine descendants). The bundle of rights notion distorts these relations, however, through over-specification and rigidity.

36The issue is of more than anthropological importance, for the process of recognising Aboriginal connections to country following Mabo II and native title legislation hinges on the analysis of connection in terms of “rights”. This is inevitable I suppose, since the exercise entails the translation of Aboriginal concepts to those of the Common Law tradition. But in recognising traditional connections, the process inevitably transforms them. Furthermore, the translation of Aboriginal connection into the language of rights facilitates their whittling away to accommodate non-Aboriginal interests.

Haut de page

Pour citer cet article

Référence électronique

Ian Keen, « The language of ‘rights’ in the analysis of Aboriginal property relations »Les actes de colloques du musée du quai Branly Jacques Chirac [En ligne], 4 | 2014, mis en ligne le 13 juin 2014, consulté le 25 juin 2024. URL : ; DOI :

Haut de page


Ian Keen

Australian National University

Haut de page

Droits d’auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés), sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search