Journal of Cultural Economy

ISSN: 1753-0350 (Print) 1753-0369 (Online) Journal homepage: https://www.tandfonline.com/loi/rjce20

THE ONTOLOGICAL POLITICS OF ‘CLOSING THE

GAPS’

Tim Rowse

To cite this article: Tim Rowse (2009) THE ONTOLOGICAL POLITICS OF ‘CLOSING THE
GAPS’, Journal of Cultural Economy, 2:1-2, 33-48, DOI: 10.1080/17530350903063917

To link to this article: https://doi.org/10.1080/17530350903063917

Published online: 24 Jul 2009.

Submit your article to this journal

Article views: 301

Citing articles: 7 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rjce20
 THE ONTOLOGICAL POLITICS OF
‘CLOSING THE GAPS’

Tim Rowse

In Australia and New Zealand, the realization of the knowledge object ‘national population’
makes it necessary to involve Indigenous Australians and Māori in the Census. Both Indigenous
peoples have engaged in the Census and have made use of the resulting official statistics in their
self-representation as peoples not yet accorded social justice. This paper considers two of the
issues of representing Indigenous peoples as populations: where to draw the distinction that
makes the non-Indigenous/Indigenous population binary; and how to prevent the quantitative
representation (‘population’) from subverting the qualitative representation (’people’). That
‘population’ might trump ‘people’ is arguably an effect of the nation-state being a kind of
‘method assemblage’ in which people are arrayed as social entities that are knowable in certain
terms. Drawing on the terms of recent liberal political theory, the paper poses the question of the
‘civicity’ of Indigenous Australians and Māori, concluding that there are ways that Indigenous
intellectuals might use population data to substantiate their claims to people-hood.

KEYWORDS: social justice; political ontology; liberalism; population; Indigenous people;

civicity

The policy of ‘closing the gaps’
a phrase that Australia has borrowed from New
Zealand
aims to achieve social justice expressed as the measured socio-economic
equality of the Indigenous and the non-Indigenous populations. ‘Closing the gaps’ thus
deploys a population binary: Indigenous and non-Indigenous population data configured
in comparable terms. In Australia, this binary is a relatively recent achievement, first made
public by the Commonwealth government in April 1969, using the 1966 Census.
Aborigines were for the first time compared with the total population in respect of
education status and occupation structure.1 To tell the story of the rise and deployment of
the population binary is to write a narrative of what the British sociologist John Law calls
‘ontological politics’. What is ‘ontological politics’?

‘Ontological Politics’
In his book After Method: Mess in Social Science Research (Law 2004), Law makes
explicit what he sees as the commonly accepted metaphysics of Euro-American social
science:
. That there is a reality out there beyond ourselves;
. And it is usually independent of our actions and our perceptions;
. This reality thus precedes us;

Journal of Cultural Economy, Vol. 2, Nos 1
2, March/July 2009

ISSN 1753-0350 print/1753-0369 online/09/010033-16
– 2009 Taylor & Francis DOI: 10.1080/17530350903063917
34 TIM ROWSE

. And it has a discoverable structure: definite forms and relations;

. That are the same everywhere.

When we enumerate, describe and analyse objects such as populations, our commitment
to this common sense metaphysics allows us to think of the reality that we are working on
as singular, constant, passive and universal (Law 2004, pp. 24
25). We therefore assess
methods of data generation and analysis according to how likely they are to approximate
to what is really there. This assessment makes sense if we suppose the object in question
to have the above five characteristics. Law proposes an alternative way to think about
social science methods in which methodological issues are important not because they
determine how close we get to what is really there but because to choose a method is to
begin to construct a reality.
Working through a series of ethnographies of reality construction
in laboratories,
hospitals and other places
Law does two things. First, he shows how methods of rational
investigation construct or perform realities; second, he points to the entrenched habits of
thought and practice that enable us to forget such processes of construction and
performance, and thus to persist in a way of thinking about method in which the
important question is: how accurately does it reveal reality? Law insists that ‘method is not,
and could never be, innocent or purely technical’ because it ‘unavoidably produces not
only truths and non-truths, realities and non-realities, presences and absences, but also
arrangements with political implications’ (2004, p. 143). He calls the means of producing
these arrangements ‘method assemblages’. Any social science method is embedded in a
‘method assemblage’, but accounts of method usually ignore or take for granted parts of
the method assemblage that are the assumed conditions of existence of the method itself.
Law offers us a new question: what reality does a particular ‘method assemblage’ render?
To try to answer this question is to investigate the ‘ontological politics’ of a particular
process of social inquiry.
John Law’s work belongs to a strand of thinking about social science that explores
the ways in which social science plays ‘a very significant role in making up our world, and
the kinds of persons, phenomena and entities that inhabit it’ (Osborne & Rose 1999, p.
368). One example is the opinion poll. Polls, repeatedly administered and publicized,
‘‘‘make up’’ people; people come to ‘‘fit’’ the demands of the research; they become, so to
speak, persons that are by nature ‘‘researchable’’ from that perspective’ (Osborne & Rose
1999, p. 392). It is difficult to imagine the practised idea of the opinionated person being
subject to a counter practice that cultivated the opposite potential: of humans as not
subjects and/or as subjects without any opinion on anything. In this sense, public opinion
pollsters do their ‘making up’ in a power vacuum: there is no comparable agent working
to make up people as a non-public. The nearest we get to making concrete the non-public
of a poll is when the pollster includes ‘no opinion’ as an optional answer. A pollster who
does this will measure the extent to which the public is not opinionated. If a high
proportion of answers is ‘no opinion’, then the poll is, in a sense, a failure: it has not
produced a ‘public’ for an issue (though to characterize the public as ‘indifferent’ may shed
light on an issue). However, a pollster is under no obligation to include a ‘no opinion’
option, and those who exclude that option will harvest data that give the impression that
the entire public is engaged with the polled issue and has an opinion.
Whereas the form of opinion poll is a relation of pollster to respondent that is always
the same (pollster asks, respondent answers), the content of opinion polls is infinitely
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 35

variable. So, within the technology that presupposes and produces ‘the public’ there are
significant choices about what questions to ask, how to word them and how to set out the
options for respondents. Public opinion polls as a ‘method’ assemblage thus produce the
public at two levels of generality that we may distinguish: there is the public, in general,
presupposed and produced as opinionated by the posing of any question; and there are
the particular publics generated by the pollsters’ selection and framing of issues. To study
the ontological politics of the polls, we could thus pursue two lines of inquiry: has there
been a poll (for there are some matters that are unlikely ever to be conceived as matters of
opinion)? And in what terms did pollsters’ questions elicit ‘opinion’?
Similarly, a census administered to households may ask many different questions,
but before we consider the politics of the questions (what features of people are being
enumerated?) we must notice a politics that is implicit in the very existence of the census.
A census presupposes a territory (usually a nation-state) beyond which the census does
not reach and within which it is universally understood and accepted. It presupposes an
interlocutor (most likely an adult in a household) who is capable of imagining, more or less
sympathetically, ‘the state’ as the entity posing the question. A census presupposes a
shared understanding of the key terms of the instrument
such as ‘household’, ‘income’,
‘employment’. For these presuppositions to be made good
that is, for the deep social
infrastructure of the census to exist
is a political achievement, a kind of basic hegemony,
that it is easy to take for granted. In parts of contemporary Australia it cannot be taken for
granted. According to Frances Morphy and her colleagues, field observations of the
administration of the 2001 and 2006 Censuses showed that, in many respects, this
hegemony has not been fully established among the Yolngu in north-east Arnhem Land
(Martin et al. 2002; Morphy 2007).
Historical studies of the census as a political technology are likely to treat as
unproblematic the deep social infrastructure of the census. One of the best known is
Nicholas Dirks’ account of the Censuses of India conducted by British authorities. Dirks
wishes to historicize ‘caste’. He shows that censuses, among other practices, helped to
make caste seem ‘somehow fundamental to Indian civilization, Indian culture, and Indian
tradition . . . capable of expressing, organizing, and above all ‘‘systematizing’’ Indians’
diverse forms of social identity’ (Dirks 2001, p. 5). Dirks’ chapter on ‘The enumeration of
caste’ focuses on the politics of the questions posed in survey and Census instruments

the formation of a schema of caste classification
and he treats as unproblematic the
census administrator’s confident assumption that the object of their inquiry
a territorially
defined population that included persons (whether officials or householders or individuals)
acculturated to respond to questions
existed. I intend no criticism, for what Dirks tells us
is important: that the practical effectiveness of this ‘caste’-oriented way of seeing the social
relationships within the population of India owed much to its congruence with the
perspective of Brahmanic privilege. In historicizing the order of caste, however, it was not
Dirks’ business to account for the possibility of a state-entity posing questions to a
territorially-defined sector of the human population, and the possibility of a respondent-
entity (local officials, householders, individuals) answering.
If Law would accept the two instances of ‘method assemblage’ that I have
discussed
public opinion polls and censuses
then perhaps he would accept that it
may be useful to tease apart the two levels at which these two method assemblages
work. That is, let us distinguish their deep social infrastructure
relatively invariant, once
established
from their ‘semantics’
the contingent terms of elicitation, the questions,
36 TIM ROWSE

the shared concepts put into play in a sequence of question and answer. The ontological
politics of the binary ‘Indigenous/non-Indigenous’ is more a matter of the semantics of
the practice of official statistics, though it rests on the (not yet completed) deep social
infrastructure of census-taking. Thus, in my historical research on the politics of
Indigenous enumeration, I tease these levels apart by posing a series of questions that
historicize ‘Indigenous population’.2
1. When and why did colonial authorities conceive ‘native’ peoples in quantitative terms, that
is, as a ‘population’?
2. By what administrative machineries have they deployed instruments of measure?
3. What have been the terms of those instruments?
4. What has made it possible for people to cooperate? What uses have Indigenous people
found for official representations of themselves as a ‘population’?
5. At what point in history have governments in settler colonial societies produced the
Indigenous/non-Indigenous binary? With what effects?
The material in this paper consists largely of answers to questions 4 and 5. In addressing
these questions through the perspective of ‘ontological politics’ I aim to show how
different ways of statistically assembling Indigenous populations and cultures are related
to different ways of acting on, and being reacted to by, those populations and cultures.

Indigenous Cooperation with Official Data Collection

Australia’s 1966 Population Census asked
13. Race: State each person’s race. For persons of European race wherever born, write
‘‘European’’. Otherwise state whether Aboriginal, Chinese, Indian, Japanese, etc., as the
case may be. If of more than one race, give particulars, for example, ½ European
½
Aboriginal, 34 Aboriginal
¼ Chinese, ½ European
½ Chinese
In publishing 1966 Census results, the Bureau at first obeyed the Australian Attorney
General’s definition of ‘aboriginal native’ (upper case ‘A’ was not then considered
mandatory) as a person of half or more Aboriginal descent. However, both social scientists
and Aborigines themselves argued that many people of less than fifty per cent Aboriginal
descent lived with, and in the manner of, Aborigines and considered themselves to be
Aborigines. After a brief internal debate about whether it would be socially regressive to
recognize and endorse such identifications, the Australian government included a revised
question on ‘race’ in the 1971 Census that constituted a self-identified ‘Aboriginal
population’.
5. What is this person’s racial origin? (If of mixed origin indicate the one to which he
considers himself to belong) (Tick one box only or give one origin only): 1. European
origin, 2. Aboriginal origin, 3. Torres Strait Islander origin, 4. Other origin (give one
only) . . .
The Australian Government’s abandonment of a descent-based delimitation of the
Aboriginal population and switch to self-identification has been hailed by Mick Dodson
(a prominent Indigenous academic and political activist) as conceding the right to ‘self-
determination and self-identification’ (Dodson 2003, pp. 32, 39
40).
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 37

As in Australia, a long-standing official tradition of defining the Indigenous

population by degree of descent came under challenge in New Zealand in the 1970s.
Up to 1974, legislation defined Māori as those of half or more Māori descent. The Māori
Affairs Amendment Act 1974 introduced a new definition: ‘all those with Māori ancestry, no
matter how remote, were Māori for the purposes of the Act’ (Statistics New Zealand 1999,
p. 10). To be consistent, the Census definition had to change. In 1976, as well as asking
people of Māori descent to categorize themselves in terms of their ‘degree of blood’, the
New Zealand Census asked people simply if they were of Māori descent. According to a
subsequent review by the Department of Statistics, the combination of these two
questions produced such confusion among respondents that the data were ‘highly
unreliable’ (Brown 1983, p. 30). This review publicly reconsidered the ethnicity question,
and found that among users of New Zealand statistics (including Māori organizations)
there was a strong desire to change the Census question from ‘degree of descent’ to
‘cultural affiliation’ (Brown 1983, p. 39). All 15 submissions on the ethnic question
‘objected to the requirement to state fractions of racial origin. This was seen as an
‘‘offensive, distressing and overly complex method of assigning people to an ethnic
group’’’ (Statistics New Zealand 1999, p. 11). By then, many government agencies were
already using ‘cultural affiliation’, without reference to fractions of racial descent, in data
collection. To bring about consistency between agencies, and in a concession to popular
understandings of ethnicity as ‘identity’ rather than as precisely graded descent, the New
Zealand Census from 1986 asked respondents to say whether they affiliated solely to Māori
(‘sole Māori’) or whether they saw themselves as belonging to Māori among other ethnic
groups. Respondents did not have to say whether degree of descent was the basis of their
ethnic identity. A further review, in 1988, recommended that the Census find a way to
distinguish ‘between ancestry and cultural affiliation aspects of ethnicity’ (Statistics New
Zealand 1999, p. 13). In the 1991 Census, there were two questions relevant to measuring
the Māori population. One asked: ‘Which ethnic group do you belong to?’ the other: ‘Have
you any New Zealand Māori ancestry?’
This series of steps away from a state-authorized Māori population boundary based
on degree of descent was a response to pressure from both the popular level (Māori freely
interpreting their degree of Māori descent, when answering the Census) and the
organized level (Māori and other organizations, including many government agencies,
needing consistency among ‘ethnic’ statistics). As in Australia, the change to self-
identification has been welcomed by Māori intellectuals. ‘Self-identification underpins
ethnic classification and is a principle of self-determination’ wrote New Zealand Statistics
officers Bridget Robson and Papaarangi Reid (Robson & Reid 2001, p. 11). Official statistics
about Māori help to enforce the Treaty of Waitangi whose Article III is widely interpreted,
by Māori and others, as binding governments to raise Māori standards of living to those of
New Zealanders as a whole. Defective information on the ethnicity of the New Zealand
population and on the social and economic characteristics of Māori weakens ‘our ability to
show disparities, and hence, Treaty breaches’ (Reid 1996, p. 10).
The population binary that measures Indigenous socio-economic deprivation
relative to non-Indigenous Australians has encouraged Indigenous critiques of formal
liberalism. In Achieving Social Justice (Behrendt 2003) Larissa Behrendt criticized those who,
by insisting on the formal legal equality of Australian citizens, objected to special public
policies directed toward Indigenous Australians. An obsession with formal equality, she
argued, had deprived liberalism of a conceptual basis for public policies (‘special
38 TIM ROWSE

measures’) designed to overcome what the she called ‘substantive inequality’ between
Indigenous and non-Indigenous Australians. She commended what she called ‘outcome
focused liberalism’, explaining
Because substantive equality is concerned with the equality of outcomes, it allows for a
balancing of equal measures with special measures, a balance that allows difference and
equality to co-exist. It means measuring the success of legal reforms, legislation and
policies by their ability to create equality. This result-oriented approach rejects formal
equality unless it can produce equality in its outcomes. (Behrendt 2003, pp. 126
127)
In short, between the 1970s and the 1990s, Indigenous intellectuals in Australia and New
Zealand formed enthusiastic constituencies of users of the population binary Indigenous/
non-Indigenous.
In the rest of this paper I want to draw attention to two issues that the use of this
binary has created: where to draw the line between the two entities in the binary, and how
to imagine the coherence, as a people, of the Indigenous population.

Where to Draw the Line in the Binary

When the Commonwealth Bureau of Census and Statistics published The Aboriginal
Population of Australia: Summary of Characteristics (CBCS 1969) with tables comparing
‘Aborigines’ with all Australians, the definition of ‘Aboriginal’ was ‘those persons who
described themselves in the 1966 Census as being 50 per cent or more Aboriginal or
simply as ‘‘Aboriginal’’’ (CBCS 1969, p. 3). Some social scientists cautioned that with this
narrower (‘darker’, one might say) ‘Aboriginal population’ the ‘degree of Aboriginal
differentiation may be somewhat overemphasized and the sharpness of the discontinuity
between Aborigines and non-Aborigines may be exaggerated’ (Broom 1970, pp. 150
151;
Broom & Jones 1973, p. 24). A complete and unbiased description of the Aboriginal
population and a fair-minded account of their relative deprivation required, in these
researchers’ view, that people of less than half Aboriginal descent should be placed on the
Aboriginal side of the binary. A limited release of the 1966 Census data expanded the
‘Aboriginal population’ to include those who had identified themselves in the Census as of
any degree of Aboriginal descent; this revised ‘Aboriginal population’ was then called the
‘identified Aboriginal population’. The Census question from 1971 to 2006 has invited
people to decide their own criteria of ‘Aboriginal’ and ‘Torres Strait Islander’, dismantling
the barrier against people of less than half Indigenous descent.
Consideration of this episode prompts a question for those considering Law’s
‘ontological politics’: Who is authorized to engage in it? Indigenous Australians were not
party to the discussions that took place among bureaucrats and social scientists between
1968 and 1970 about re-composing the population binary. Indigenous Australians in the
1960s made no secret of their dislike of being classified by degree of descent, and it is
likely that the officials who made the decision to include those of lesser descent, under
academic advice, also took note of that widespread sentiment. Better that the Census
address this pan-Aboriginal ideology empathetically than continue to apprehend it as a
source of defective answers to the ‘race’ question. Indigenous Australians’ engagement
with the Census was at stake. I wonder what they would have said about the claim that
this change in the composition of the binary would show a narrower socio-economic gap,
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 39

that is, a society closer to attaining social justice. In the practice of ‘ontological politics’,
‘Who’s in the loop?’ is an inescapable issue.
Māori have been part of the New Zealand discussion about where to draw the line
that defines the binary Māori/non-Māori. Three Māori populations can be generated from
the New Zealand Census.
. Ancestral Māori (who are entitled to be on the Māori Electoral Roll and to avail
themselves of the Waitangi Tribunal, but who do not necessarily identify as members of
the Māori ethnic group)
. Ethnic Māori (who are generally assumed to be ‘ancestral Māori’ and who also identify
themselves as belonging to Māori and possibly to other ethnic groups)
. Sole Māori (the subset of Ethnic Māori who say they do not belong to any other ethnic
group).

The Department of Social Policy’s Chief Research Analyst, Simon Chapple (Chapple
2000) questioned the utility
for social policy
of defining the Māori population as the
‘ethnic Māori’, that is, all those who said that at least one of their ethnic identities was
Māori: some 524,000 people in the 1996 Census. Chapple urged the government to define
the Māori population, for social policy purposes, as the 274,000 ‘sole Māori’
those who
do not identify with any ethnicity other than Māori. That is, Chapple was proposing that as
many as 48 per cent of ‘ethnic Māori’ could be reclassified as ‘non-Māori’ in the
comparison of Māori and non-Māori socio-economic characteristics.
Chapple made a number of observations about the Ethnic Māori. They were highly
exogamous. They differed enormously in the strength of their attachment to Māori
customs, associations and political parties. A significant minority changed their answers to
the ethnicity question between the 1991 and 1996 Censuses. There was a range and
distribution in the incomes of Ethnic Māori that closely resembled the range and
distribution of non-Māori incomes. He made the same observation about life expectancy
at birth. He thus drew attention to the within-ethnic-group differences that are obscured
when averages or medians of the groups are compared. Chapple was critical of Statistics
New Zealand for assigning all those Māori with multi-ethnicity to a single ethnicity: Māori.
Summing up, he wrote:

Popular rhetoric to the contrary, Māori do not share a common experience of socio-
economic disadvantage. The Māori ethnic group is not a group whose boundaries are
well defined by socio-economic failure. Socio-economic differences amongst Māori as a
group overwhelm socio-economic differences between Māori and other groups. Māori
ethnicity is not socio-economic destiny. Seeing members of the Māori ethnic group as
socio-economic failures defined only by the average Māori group outcome is to
perpetuate a highly inaccurate stereotype. (Chapple 2000, p. 110)

‘It is sole Māori with low literacy, poor education, and living in geographical concentrations
that have socio-economic problems, not the Māori ethnic group as a whole’ (Chapple
2000, p. 115). He expressed the fear that New Zealand’s policies of Indigenous rights were
likely to be of most benefit to already well off Māori.
Responding to these observations, Māori sociologist Tahu Kukutai (Kukutai 2003)
analysed data gathered by the ‘New Zealand Women: Family Education and Employment’
(NZWFEE) survey of 1997. This survey included questions that allowed those Māori who
40 TIM ROWSE

identified with more than one ethnic group to say whether they considered themselves
primarily Māori or primarily non-Māori. Two out of every five of such respondents said that
their primary identity was Māori, and the survey showed that these people were similar to
‘sole Māori’ in their socio-economic characteristics. Kukutai argued that they should be
included in the socially vulnerable ‘Māori’ population, along with the ‘sole Māori’.
Conceding Chapple’s point that the wider Māori Ethnic Group population was hetero-
geneous in its social characteristics (Kukutai 2003, p. 96) and that it is necessary for social
policy to distinguish Māori according to socio-economic deprivation, Kukutai argued that
it would be better to make Māori ‘primary identity’ the criterion of the ‘Māori population’.
Two features of this New Zealand discussion should be noted. First, neither side was
treating the Māori population (and hence the binary of which it is a part) as a given, as if it
were a natural phenomenon; it is agreed that there are multiple ‘Māori populations’ and
more than one Māori/non-Māori binary is possible. Second, whatever their differences,
both Chapple and Kukutai highlighted socio-economic criteria for determining the Māori/
non-Māori boundary. Ethnic identity is thus a necessary but not a sufficient element in the
definition of the Māori population when ‘social justice’ is operationalized as ‘Closing the
Gaps’.

Peoples or Populations? Liberal Political Ontology

This leads to an important question: Are the Indigenous populations produced by
Censuses, official statistics and consequent analytical work Indigenous peoples? As
Indigenous peoples have mobilized politically, they have insisted that they are more
than mere ‘populations’ (the term used in the ILO’s 1957 Convention 107 on the
‘Protection and Integration of Indigenous and Other Tribal and Semi-Tribal Populations’),
they are ‘peoples’ (as in the 1989 ILO Convention 169 ‘Concerning Indigenous and Tribal
Peoples in Independent Countries’). As De Costa points out, between the two Conventions
‘indigenous populations’ fell out of favour in global human rights discussions and
‘Indigenous Peoples’ became standard usage (De Costa 2006, p. 133).
To be a ‘people’ is to imply that the ‘population’ coheres in some way that is more
than the coded equivalence of each instance. It is not just that, on Census night, many
households answer a certain question in such a way as to be classified as ‘Indigenous’

though that is certainly a necessary part of what happens. It is also that those giving that
answer are in some respects a ‘group’ whose members share
at the very least
an
identity term, but possibly much more: religious beliefs, ritual practices, territory, historical
narratives, kinship networks, customary and voluntary associations, governmental institu-
tions. Māori and Aborigines would say that they are more than just a population category
because their past and current relationships with one another and their shared
relationship with those on the other side of the binary bind them into people-hood.
Here it is relevant to notice the problem of ‘political ontology’ in liberal political
theory. There is a neo-liberal or contractualist ontology that insists that a people or a
society is not anything more than an aggregate of individuals who contract with one
another to give effect to their common purposes. The justice norms of society can be
accounted for only by reference to those contracts made by individual agents. Society is
not itself a moral agent; ontologically a society is no more than an ensemble of contracting
individuals. Those dissatisfied with this view are prompted by an intuition of an ontology
of collective being, a sense that ‘society’ is something more than the sum of its contracting
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 41

individual members. But what is it about the ‘togetherness’ of individual agents that
constitutes something real beyond their individual agency? Reading the work of John
Rawls, Phillip Pettit has inferred the bases of a collective quality that Rawls calls ‘civicism’
and whose existence makes a social collective a ‘civicity’. The social dynamic that produces
‘civicity’ is debate. In controversy in a ‘well-ordered’ and democratic society, even when
people cannot come to agreement, there accumulates among the disputants a shared
sense of what is a relevant consideration: ‘the commonly authorized presumptions and
valuations of the civicity’ (Pettit 2005, p. 169). ‘Democratic society is not a group agent of a
solidarist kind. But neither is it a mere aggregate of separate individual agents. It is,
precisely, a civicity’ (Pettit 2005, p. 170).
In his exposition of Rawls’ political ontology, Pettit makes available the terms in
which we might think of Māori and Aborigines as peoples, and not just as aggregates of
households that have ticked the ‘indigenous identifier’ question in a Census or survey.
That is, an Indigenous people generates its distinct ‘civicity’ when people engage with one
another, even disputatiously, and invoke a shared set of considerations relevant to
determining a Māori, Aboriginal or Torres Strait Islander perspective on a matter that
excites their concerns.
However, Pettit undercuts the promise of this possible ontology of Indigenous
people-hood when he turns to the issue of justice between peoples (Pettit 2005, pp. 171

172). In his discussion of Rawls’ skepticism about a cosmopolitan theory of justice, Pettit
does not challenge Rawls’ assumption that how peoples relate to one another is a
question about ‘the international scene’ (Pettit 2005, p. 172). That is, in his discussion of
relationships between peoples Pettit effectively aligns the terms ‘people’, ‘society’ and
‘nation’. In this perspective, it is not possible to pose the question that confronts many
Indigenous people: on what terms are peoples within a nation-state to live together? Pettit
does not ask whether, within a nation-state, there could be an Indigenous ‘civicity’
through whose debates Indigenous people try to come to terms with their position within
the wider, encompassing nation-state (a society with its own ‘civicity’). Pettit notes that
underlying Rawls’ political ontology is the idea that a well-ordered, democratic society ‘will
not be regulated by any single comprehensive doctrine’ (Pettit 2005, p. 170):
contemporary democracies accommodate a plurality of comprehensive doctrines. Further,
following Rawls, Pettit postulates that, in the orderly disputations that arise from such
plurality, each democratic society will generate a single overarching ‘civicity’. The ultimate
frame of any ‘civicity’ is the polity within which disputants co-exist, and Pettit offers no
challenge to the assumption that a polity implies a nation-state. In this way, it is
reasonable for Pettit to assume that each democratic nation-state generates a single
‘civicity’
even if that assumption finesses the issue of whether, within a nation-state,
there may be more than one ‘people’ and more than one ‘civicity’.
Those who propose a political ontology of Indigenous peoples thus find themselves
facing a powerful adversary: a liberalism that at first sight grounds ‘civicity’ in people-
hood, is actually grounding ‘civicity’ in nation-state-hood. For liberals such as Pettit, this is
no problem: he aligns the terms ‘people’ ‘society’ and ‘nation’ in a way that assumes
nation-states to be the basis of a civicity. The point of Indigenous political theory, as I
understand it, is to question that alignment, imagining nation-states to be the political
framework in which the ‘civicity’ of a colonized people engages with the imposed national
‘civicity’. I suggest that Law’s notion of ‘method assemblage’ can help us to understand
the strength of the presence of ‘nation-state’ in such liberal imagining of ‘people-hood’ as
42 TIM ROWSE

we find in Rawls and Pettit. A national Census is a ‘method assemblage’ that enables some
realities and disables others. In what follows I will argue that in official statistical
representations of the Indigenous, it is easier to see the Indigenous as a ‘population’ than
as a ‘people’. As population data acquire prominence in the representations that liberal
settler colonial societies such as Australia and New Zealand afford to Indigenous
minorities, representations of Indigenous ‘people’ are at risk of being undermined by
the currency of ‘population’-based representations of the Indigenous.
It is now relatively easy to be a ‘population’. The Census in each nation is very
undemanding: it simply addresses each individual or household, and it does not require of
each responding unit that it give an account of its relationships with similar responding
units. The coding of the Census creates the aggregation ‘Māori ’ or ‘Aboriginal’; whether
there is more to the aggregation than that depends on political and social activity (for
example, by state agencies, by Indigenous organizations and by families and individuals)
external to the Census-taking itself.
In New Zealand, as Andrew Sharp has pointed out, Māori have presented in more
than one way the ties that bind them as a people.

Māori in traditional kin groups such as whānau, hapū, and iwi (sometimes called families,
sub-tribes, and tribes); Māori in Māori voluntary organizations, such as modern ‘urban
iwi’; and Māori conceived of as members of the ethnic or racial group Māori. Each kind of
grouping has a different ruling justification for existing
roughly, blood kinship, consent,
and the sharing of genetic and cultural material. But in each case the ruling justification is
not the only one. (Sharp 2002, p. 15, emphasis in original)

Sharp argues that, as an ethnic group, largely constituted by the classifications of official
statistics, Māori do not constitute a ‘team of action’. That is, as an ethnic group, Māori can
be acted upon, but they cannot act for themselves. When Māori act for themselves, they
do so in those ‘descent’ and ‘consent’ groupings in which members work hard to
institutionalize cohesion and collective capacity. I suggest that we can accept Sharp’s
three part schema as a general model for the various ways that Indigenous peoples
achieve representation as a collective. In the terms of his schema, we can understand the
project of ‘self-determination’ as an effort by Indigenous people to get others (the state
most of all) to deal with them in terms of their corporate capacities as rights-bearing
descent and consent groupings, albeit with reference to their quantified characteristics as
an ethnic population. Their people-hood lies in the combination of the three modes
distinguished by Sharp, and to sustain this combination requires political effort at many
levels. Political regimes that are hostile to self-determination will attach little or no legal/
political significance to formations built on Indigenous ‘descent’ and ‘consent’ ties (while
allowing such ties as features of individuals’ private existence), favouring instead the
representation of Indigenous people as a mere population category (whose quantified
socio-economic characteristics mark them as needful).
When the Australian government (under John Howard 1996
2007) championed the
‘practical’ and disdained the ‘symbolic’ aspects of ‘reconciliation, it addressed Indigenous
Australians as a category of the population whose socio-economic disadvantage, well
documented in official statistics, demanded remedies. The Howard government discour-
aged and distanced itself from those formations of Indigenous Australians in which they
presented themselves as rights-bearing and politically self-conscious networks of kin and/
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 43

or associations. The formation that mattered more to the Howard government was the
statistically-defined constituency of need, not the politically-affirmed constituency of right.
By appealing to ‘equality’, Howard shared some ground with the Aboriginal and
Torres Strait Islander Social Justice Commissioners. In a series of annual reports about
‘social justice’, beginning in 1993, Commissioner Mick Dodson and his successor Bill Jonas
used official statistics to measure progress from injustice to justice in terms of (for
example) comparative rates of imprisonment and numbers of people per dwelling. In 2000
Jonas quoted the UN’s Committee to Eliminate Racial Discrimination (or CERD) as
acknowledging that:
. Indigenous disadvantage is the result of systemic discrimination;
. the appropriate benchmark by which to measure progress is one of equality between
Indigenous and non-Indigenous Australians;
. the government is obligated to take sufficient steps (or special measures) to achieve such
equality;
. there must be adequate monitoring and evaluation of progress, including measuring
effectiveness through benchmarking and standard setting; and
. real progress requires the effective participation of Indigenous people in decision making
(including through the representative voice of ATSIC). (ATSISJC 2000, pp. 47
48)
The Howard government’s promotion of ‘practical reconciliation’ as equality stimulated
the Aboriginal and Torres Strait Islander Social Justice Commissioners to clarify their
understanding of ‘equality’ as more than simply distributive. They understood the
inequalities addressed by ‘practical reconciliation’ as an effect of historically constituted
structural political inequality between a colonized minority and the colonizing power.
They worked from the conviction that from the inequalities revealed and quantified by the
population binary we can infer a structural political inequality between peoples.
However, this inference is fragile to the extent that statistics are the coin of what is
socially real, for the organs of Aboriginal and Torres Strait Islander people-hood are not
made visible by the Census: they are less real social entities than the aggregated attributes
of individuals and households recorded as Indigenous. To supplement the Census, the
Australian government has conducted the National Aboriginal and Torres Strait Islander
Social Survey (NATSISS) in 1994 and 2002, eliciting data relevant to the issue of Australian
Indigenous cohesion. Less than half (47.5 per cent) of the Indigenous people in non-
remote Australia (where three out of four Indigenous Australians live) see themselves as
belonging to a clan, tribal or language group. In ‘remote’ and ‘very remote’ regions, three
out of four people (76 per cent) identify themselves as part of a clan, tribal or language
group (ABS 2004). So these traditional institutions of social cohesion may not be relevant
to all in the Indigenous ‘population’ and the reach of such institutions seems to be less in
the cities and towns than in the remote regions. Whether this makes it harder to see
Indigenous Australians as a people depends on the weight that one attaches to ‘clan, tribe
or language group’ as forms of collective presence. These are not the only possible bases
of Indigenous Australians’ collective deliberation and action: since the 1970s, the
Commonwealth has encouraged Indigenous Australians to formalize their associations
and to develop municipal bodies out of former missions and settlements; and there are
informal kin-based networks that population data do not capture.
In New Zealand, by contrast, Māori have persuaded the government to inscribe
Māori political entities within the Census itself. Since the 1991 Census, Māori have been
44 TIM ROWSE

asked to what iwi they belong. About one in five ‘ethnic Māori’ do not give an answer. If
such people are not affiliated to an iwi then it brings into question, in one observer’s view,
the significance of iwi in Crown/Māori settlements under the Treaty of Waitangi. For the
Crown to deal with the Māori population through iwi, argues Elizabeth Rata, has resulted
in ‘the exclusion and dispossession of detribalized Māori from ownership of the means of
production’ (Rata 2002, pp. 173
174). She infers that a significant minority of the Māori
population does not share in the resources held by iwi on behalf of the Māori people. We
could say that her ontological politics privileges the population as the grounds from which
to assess the institutions of Māori people-hood.
We can read the work of Mason Durie and his colleagues at Massey University as a
response to such a criticism. Their Māori Specific Outcomes and Indicators (2002) seeks
representation of Māori people-hood in the idiom of Māori population. Their aim is to
improve the capacity of official statistics to measure the progress of the Māori people as a
whole, according to ‘objectives and aspirations, unique to Māori, that [reflect] cultural
rather than generic ideals’ (Durie et al. 2002, p. 33). Not only do Māori want what all New
Zealanders want
‘a good job, access to education, to be healthy, or to own their own
home’ (Durie et al. 2002, p. 33)
they wish also to maintain their distinct heritage: ‘a secure
identity, te reo Māori [language], tikanga Māori [custom, rules], and having access to those
institutions which nurtured the sense of being Māori (whānau, hāpu, and iwi, marae and
hui)’ (Durie et al. 2002, p. 33). Whereas Rata, thinking of all those Māori who are not
connected with an iwi, finds fault with public policy that makes iwi central to Māori well
being, Durie and his co-authors propose to evaluate public policy by asking, among other
questions, how much it encourages each Māori individual and whānau to form or to
strengthen affiliation with iwi and with other social units such as marae and hapu that
Durie and his colleagues see as essential to the well-being of Māori.
Two issues arise in their exposition of Māori people-hood in terms that statistics
could substantiate: the characterization of the unaffiliated individual, and the content of
Māori norms of wellbeing. Durie and his co-authors acknowledge that ‘hapu and iwi
activities do not form part of the experience of all Māori’ (Durie et al. 2002, p. 52) and they
affirm that ‘a positive outcome would necessarily deconstruct access boundaries
to te ao
Māori [the Māori domain] or te ao Pakeha [the Pakeha domain], leaving the individual with
choice and a genuine sense of control’ (Durie et al. 2002, p. 34, emphasis added). These
words evoke the unaffiliated Māori as a sovereign individual, choosing between ethnically
defined worlds for his/her primary identity and way of life. An alternative evocation is to be
found in a later passage in which the Report seeks to evaluate the extent to which public
policy fosters Māori well-being by enabling ‘participation and achievement in Māori
society. Active participation in the Māori world is closely linked to a secure cultural identity’
(Durie et al. 2002, p. 51, emphasis added). These words imply that a person of Māori
descent who does not participate in the Māori world is insecure in identity: his/her non-
participation is not an expression of individual autonomy but a mark of his/her regrettable
deracination. The Report’s notion of Māori well-being seems, for the most part, to be
grounded in this communitarian presumption that a person of Māori ethnic identity is
better off choosing to participate in ‘Māori society’
measured in terms of iwi affiliation.
When Durie and his colleagues interviewed Māori experts about what they thought
were the criteria of Māori well-being, interviewees differed on how to judge whether a
Māori organization was an effective vehicle of Māori autonomy.
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 45

Most participants considered that the test was whether an organization employed Māori
values as well as Māori staff and operated for the benefit of Māori people. However, a few
participants did not accept that the modus operandi was the essential point
rather it
was primarily a question of control. In their opinion, provided the day-to-day decisions
were made by Māori, tino rangatiratanga [the Treaty right of Māori control over Māori
affairs] was being exercised. (Durie et al. 2002, p. 43)
That is, as Māori experts conceive the project of Māori autonomy, ‘the benefit of Māori’ is
not necessarily to be reckoned in terms of specified organizational procedures, employ-
ment patterns, values or outcomes; that benefit is assumed by some simply to follow from
Māori control. This is a normatively ‘thin’ criterion of Māori-ness, and it would hardly satisfy
a critic such as Rata who invokes conventional liberal democratic norms of accountability
and fairness when she assesses New Zealand public policy (including Treaty settlements);
she worries that the Crown effectively empowers Māori elites who are not necessarily
accountable to Māori masses (Rata 2006).
The intention of Durie and his co-authors in their 2002 study was to develop a
statistical framework ‘not to compare Māori with non-Māori but to assess outcomes
against Māori norms’ (Durie et al. 2002, p. 48). But did they state an agreed corpus of
‘Māori norms’? Their presentation of individual autonomy is unresolved about the nature
of attachment to community, and their brief consideration of criteria for ‘community’
strength does not raise the issue of the distribution of power and resources within a
community or among Māori at large. Whereas Rata, as critic, appeals to what she takes to
be universal democratic norms about fairness of distribution, absence of exploitation,
accountability of ‘tribal’ leaders to members (see especially Rata 2003, pp. 55
56), the
report by Durie and his colleagues neither endorses nor criticizes such norms: it is silent on
the norms relevant to judging the processes internal to Māori collectives. To return to
Pettit’s terms, the report does not substantiate Māori people-hood in the terms of a
specific civicity or normative framework.3

Conclusion: The Nation-State as a Method Assemblage

Andrew Sharp’s Justice and Māori (Sharp 1997) pointed out that New Zealand’s
debates about social justice have mingled distributive, reparative and sovereignty issues
(Sharp 1997). It is likely that his observation applies to all liberal settler-colonial societies. In
this paper I have argued that official statistics are significant in this field of discursive
possibilities. Nation-state method assemblages such as standardized processes of data
collection enable a rich quantitative representation of an Indigenous population. When
composed into a population binary, these statistics reveal stark disparities. The question
for Indigenous intellectuals and for social scientists is what conventions of interpretation are
built into and built upon this quantitative archive, and what forms of action are made
possible by it? How do statistics figure in the presentation of social justice as a problem of
distribution among individuals, among households, and between peoples? And if statistics
have difficulty substantiating peoples, do they disable arguments for reparation between
peoples and for re-establishing Indigenous autonomy?
As expounded by Law and Urry (2004), the issue of ontological politics ‘is not simply
how what is out there can be uncovered and brought to light . . . it is also about
what might be made in the relations of investigation, what might be brought into
46 TIM ROWSE

being . . . which realities might we try to enact? . . . what realities do the current methods
of social sciences help to enact or to erode? And what realities might they help to bring
into being or to strengthen?’ (Law & Urry 2004, p. 396). I have argued that the advent of
the Antipodean politics of ‘Closing the Gaps’ reveals the nation-state itself as a method
assemblage in which a certain ontological politics is powerfully institutionalized. I have
drawn attention to two realities that official statistics enable: the Indigenous population
and the population binary of which it has become a part. In the history of the Australian
and New Zealand population binaries, we can see a more or less deliberated discussion of
two issues in the construction of the realities of injustice.
First, the composition of the binary: on which side of the Indigenous/non-
Indigenous distinction should certain ambiguous (because of mixed descent, or because
of multiple ethnic identification) cases be counted?
The second issue is about the degree to which an Indigenous population (an artifact
of the use of some ‘Indigenous identifier’ question) is to be addressed and understood as
an Indigenous people (a complex political artifact, a dense texture of kinship, association
and normative regulation). To understand the Indigenous population primarily in terms of
its aggregate socio-economic features (the mean, average, median) is consistent with
more than one approach to justice. On the one hand, aggregating devices such as means
and averages that measure the socio-economic inequalities between two sub-populations
raise the question of distributive justice between the two halves of the population binary:
why are non-Indigenous (on average) better off than Indigenous? It is possible
but more
contentious
to proceed from there to an evocation of the Indigenous population as a
people who have been unjustly treated: their averaged socio-economic deficits are
explained by their common experience, as a people, of colonization. To the extent that
that account is accepted, other kinds of social justice claims
for reparations, for forms of
autonomy
are enabled. On the other hand, on the basis of official statistics, there is more
than one question of distributive justice to ask. A population can be disaggregated to
households and even to individuals, so that analysis can construct significant distinctions
among the ‘people’. It may highlight socio-economic equivalences across the binary (poor
Indigenous having needs similar to poor non-Indigenous) and it may highlight differences
within each side of the binary (between unaffiliated and affiliated Māori; between male
Aborigines/female Aborigines). Evocations of Indigenous people-hood may deploy
population statistics in such a way as to document the grievances of a people, but
alternative analytics of justice will complicate (and perhaps undermine) the assertion of
people-hood.
Rather than conclude on this cautionary note
emphasizing the risk of official
statistics to Indigenous agendas
I pose this challenge to the discourse of Indigeneity: to
admit the political relevance of the issue of distributive justice (and other kinds of justice)
within the Indigenous people. I see two reasons for doing so:
1. The diagnosis of the bases of Indigenous disadvantage might be illuminated by studying
the history and circumstances of the strata of Indigenous people who are not so
disadvantaged.
2. A people (a ‘civicity’) asserts its existence partly in terms of its norms, so a discussion about
‘Indigenous norms’ is an unavoidable and desirable feature of the discourse of people-
hood. As well as invoking aggregate statistics (such as rates of employment for the entire
Indigenous population) it is necessary to face up to Indigenous socio-economic diversity
 THE ONTOLOGICAL POLITICS OF ‘CLOSING THE GAPS’ 47

and to discuss what differences of Indigenous outcome are consistent with social justice
(because they reflect normative diversity among the Indigenous people) and what
differences of Indigenous outcome are an affront to Indigenous standards of fairness.
By deploying in different ways the statistical archive that we use to represent an
Indigenous population, the Indigenous people can be represented in a more complex way,
allowing other questions of social justice to be raised.

ACKNOWLEDGEMENTS
Research on this paper was funded by the Australian Research Council (DP0665866). I
would like to thank Jon Altman, Emma Kowal, Manuhuia Barcham, two anonymous
referees and the editors for their comments.

NOTES
1. When the Indigenous population is a small proportion, there is little difference between
an Indigenous/non-Indigenous comparison and an Indigenous/total population compar-
ison. The earliest version of the binary in Australia was in the latter terms.
2. I share my ARC grant work with Dr. Len Smith (ANU) and we have employed Dr Tiffany
Shellam (Deakin University).
3. I do not infer that Māori lack a normative culture. My point is that it is evidently difficult
to specify its content as a distinct normative universe within New Zealand society.

REFERENCES
ABORIGINAL AND TORRES STRAIT ISLANDER SOCIAL JUSTICE COMMISSIONER (ATSISJC) (2000) Social Justice
Report 2000 [Online]. Available at: http://www.hreoc.gov.au/social_justice/sj_report/sj
report00/index.html (accessed 20 November 2007).
AUSTRALIAN BUREAU OF STATISTICS (ABS) (2004) National Aboriginal and Torres Strait Islander Social
Survey 2002 (Cat no. 4714.0), ABS, Canberra.
BEHRENDT, L. (2003) Achieving Social Justice: Indigenous Rights and Australia’s Future, The
Federation Press, Annandale.
BROOM, L. (1970) ‘Educational status of Aborigines’, Australian and New Zealand Journal of
Sociology, vol. 6, pp. 150
156.
BROOM, L. & JONES, F. L. (1973) A Blanket a Year, Australian National University Press, Canberra.
BROWN, P. G. (1983) An Investigation of Official Ethnic Statistics, Occasional Paper no. 5 (Cat
no.20.005), Department of Statistics, Wellington.
CHAPPLE, S. (2000) ‘Māori Socio-economic disparity’, Political Science, vol. 52, no. 2, pp. 101
115.
COMMONWEALTH BUREAU OF CENSUS AND STATISTICS (CBCS) (1969) The Aboriginal Population of
Australia: Summary of Characteristics (CBCS Cat reference 2.23).
DE COSTA, R. (2006) A Higher Authority: Indigenous Transnationalism and Australia, University of
New South Wales Press, Kensington, NSW.
DIRKS, N. (2001) Castes of Mind: Colonialism and the Making of Modern India, Princeton University
Press, Princeton & Oxford.
48 TIM ROWSE

DODSON, M. (2003) ‘The end in the beginning: re(de)finding Aboriginality’, in Blacklines:

Contemporary Critical Writing by Indigenous Australians, ed. M. Grossman, Melbourne
University Press, Carlton, Vic, pp. 25
42.
DURIE, M, FITZGERALD, E., KANI KINGI, T., MCKINLEY, S. & STEVENSON, B. (2002) Māori Specific Outcomes
and Indicators, report prepared for Te Puni Kokiri, Ministry for Māori Development, Te
Putaki a Toi School of Māori Studies Massey University Palmerston North.
KUKUTAI, T. (2003) The Dynamics of Ethnicity Reporting: Māori in New Zealand, discussion paper
prepared for Te Puni Kōkiri, Ministry of Māori Development/Te Puni Kōkiri.
LAW, J. (2004) After Method: Mess in Social Science Research, Routledge, London.
LAW, J. & URRY, J. (2004) ‘Enacting the social’, Economy and Society, vol. 33, no. 3, pp. 390
410.
MARTIN, D.F., MORPHY, F., SANDERS, W. G. & TAYLOR, J. (EDS) (2002) Making Sense of the Census:
Observations of the 2001 Enumeration in Remote Australia, Monograph no. 22, CAEPR
Research, Canberra.
MORPHY, F. (ED.) (2007) Agency, Contingency and Census Process, Monograph no. 28, CAEPR
Research, Canberra.
PETTIT, P. (2005) ‘Rawls’s political ontology’, Politics, Philosophy and Economics, vol. 4, no. 2, pp.
157
174.
OSBORNE, T. & ROSE, N. (1999) ‘Do the social sciences create phenomena? The example of public
opinion research’, British Journal of Sociology, vol. 50, no. 3, pp. 367
396.
RATA, E. (2002) ‘The transformation of indigeneity’, Review, XXV, pp. 173
195.
RATA, E. (2003) ‘Late capitalism and ethnic revivalism: a ‘‘New Middle Age’’?’ Anthropological
Theory, vol. 3, no. 1, pp. 43
63.
RATA, E. (2006) ‘The political strategies of ethnic and indigenous elites’, in Public Policy and
Ethnicity: The Politics of Ethnic Boundary Making, eds E. Rata & R. Openshaw, Palgrave,
Basingstoke, pp. 40
53.
REID, P. (1996) ‘The Paua, the mermaid and others’, Proceedings of the Māori Users of Statistics
Conference, 20
21 November 1996, Statistics New Zealand/Te Tari Tatau, Wellington, pp.
7
11
ROBSON, B. & REID, P. (2001) Ethnicity Matters: Māori Perspectives (Review of the measurement of
ethnicity, main paper), Statistics New Zealand/Te Tari Tatau, Wellington.
SHARP, A. (1997) Justice and the Māori, 2nd edn, Oxford University Press, Auckland.
SHARP, A. (2002) ‘Blood, custom, and consent: three kinds of Māori groups and the challenges
they present to governments’, University of Toronto Law Journal, 52, pp. 9
37.
STATISTICS NEW ZEALAND (1999) Measuring Māori Ethnicity in the New Zealand Census, Statistics
New Zealand Background Paper, Social Policy Division, Wellington.

Tim Rowse, Professorial Fellow, Centre for Citizenship and Public Policy, University of
Western Sydney, Bankstown Campus, Locked Bag 1797, Penrith South DC, NSW
1797, Australia. Email: T.Rowse@uws.edu.au