1 Self-Determination as a Political Cleavage: The Norwegian Sámediggi Election of 2009 1 Johannes Bergh Institute for Social Research, Oslo, Norway [email protected]Jo Saglie Institute for Social Research, Oslo, Norway & Sámi University College, Guovdageaidnu, Norway [email protected]Paper presented to workshop 25, ‘Indigenous Politics: Mobilization, Representation, Internationalization’, ECPR Joint Sessions of Workshops, St. Gallen, 12–17 April 2011 Introduction The 2009 elections marked the 20 th anniversary of the democratically elected institution designed to represent the Norwegian Sámi people: the Norwegian Sámediggi (the Sámi Parliament). Observers of Norwegian Sámediggi election campaigns are often struck by the degree to which there is agreement on most political issues. Apparently, the ideological differences that are on display in national Norwegian elections are a less prominent feature of Sámediggi elections. What, then, explains voting in these elections? Are there ‘unspoken’ political cleavages that materialize at the ballot box, without being an explicit part of election campaigns? Are Sámediggi elections a reflection of the national Norwegian political landscape? Or, are Sámediggi elections fundamentally apolitical, in which case voting is determined by habit, attachment to individual candidates or perhaps some minor non- ideological issues? If there are political cleavages at work in Sámediggi elections that are not immediately apparent for observers of the campaigns, a natural place to start looking for them would be in the nature of the Sámediggi itself. The Sámi parliament is an attempt to establish an amount 1 We would like to thank the Sámediggi for granting us permission to draw a sample from the Sámediggi electoral register for the purpose of conducting an election survey. We also thank the Research Council of Norway for funding of the Norwegian Sámi Election Study.
22
Embed
Nationhood as a Political Cleavage: The Norwegian Sami ...€¦ · Sámi People’s Party – Sámeálbmot bellodat 3.9 2.3 1 - Local Sámi lists Reindeer Herders’ List – Johttisápmelaččaid
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Non-Reindeer Herders’ List – Dáloniid listu 1.8 1.3 1 -
Deanu Joint List – Deanu oktasašlistu - 1.2 - -
Ofelaš - .8 - -
Sjaddo - .6 - -
Finnmark List – Finnmárkkulistu 1.8 - 1 -
Southern List – Åarjel læstoe 1.6 - 1 -
Kárášjoga Badjeolbmuid listu 1.2 - - -
Mid-Nordland List – Midtre Nordland tverrpolitiske liste .4 - - -
Turnout 72.6 69.3
Total number of seats 43 39
Source: Statistics Norway (2005, 2009)
The NSR was also an established organization before the Sámediggi was founded. It was –
and still is – a cultural organization, as well as a political movement. The NSR worked for
Sámi rights and for the revitalizing of a Sámi identity during the 1970s and 1980s – often with
6
the Labour Party as its main opponent, for instance during the controversial damming of the
Alta river.3 The NSR stood as an all-embracing Sámi movement struggling against
Norwegianization, bringing together Sámi activists with otherwise diverging political views.
In Gaski’s (2008: 11) words, the NSR started out as a ‘catch-all party’.
According to Gaski (2008), NSR’s and Labour’s 2005 election campaigns lacked
explicit ideological positions: ‘throughout the campaign it was difficult to distinguish between
the two parties’ agendas’ (Gaski 2008:2). However, her discourse analysis also calls attention
to the role of national identity in the two main opponents’ self-presentation. Being an
independent Sámi organization, the NSR can easily depict Labour’s Sámi politicians as
subservient to the Norwegian Labour Party headquarters in Oslo. As a division of a
Norwegian party, Sámi Labour cannot be a true spokesman for Sámi interests – according to
the NSR.
Nevertheless, Labour’s Sámediggi politicians seem to speak more strongly for Sámi
self-determination than before. The actual policy differences between Sámi Labour and the
NSR appear to have become less prominent, also regarding Sámi self-determination. This
gives, however, room for conflict between Sámi Labour and the main Labour Party
organization – especially in Northern Norway, where scepticism against Sámi rights and self-
determination is widespread among non-Sámi (Norwegian and Kven4) voters. Since elections
for the Norwegian Parliament and the Sámediggi are held on the same day, this creates some
strategic dilemmas for the Labour Party. Prior to the 2009 election campaign, for example, the
leader of Labour’s Finnmark County branch said: ‘I was appalled when I saw [Labour’s]
Sámediggi election manifesto’ (NRK Sápmi 2009). When Labour candidates for the
Sámediggi speak for more Sámi self-determination while the same party’s candidates for the
general election speak against, the NSR can score some easy points. A rhetorical solution to
Labour’s dilemma has been to emphasize the mutual interests of the people in the North.
According to Gaski (2008: 14–15), the Sámi Labour Party prefers to address ‘ordinary
people’ rather than ‘Sámi people’.
Both the main opponents lost ground in the 2009 Sámediggi elections, although they
remained the largest parties by far (see Table 1). A newly formed Sámi party, Árja (the
Northern Sámi word for ‘commitment’) broke through as the third largest party, with ten per
3 The building of the Alta river dam and hydroelectric plant around 1980 was opposed by Sámi political activists
and others. The dam was built in a traditionally Sámi area, and flooded parts of what was seen as Sámi-owned
land. The protests in connection with the Alta river dam project led to an awakening of Sámi political
consciousness, as well as increased attention to Sámi interests in national Norwegian politics. 4 Descendants of Finnish immigrants who settled in Northern Norway during the 18
th and 19
th century.
7
cent of the votes and three seats. In its campaign, the party spoke for the rural and coastal
Sámi population outside of reindeer herding, and emphasized the role of the elderly and the
importance of traditional knowledge. After the election, Árja entered a coalition with Labour
and some minor lists, and Árja’s leader was elected vice president of the Sámediggi.
A Norwegian political party – the right-wing populist Progress Party – also had its
breakthrough in Sámi politics in 2009. The party had fielded candidates before, but no
representatives had been elected. In 2009, however, the party emerged as the fourth largest in
terms of votes, winning three seats. The Progress Party’s Sámi policy differs radically from
all its competitors: its position is that the Sámediggi should be abolished. According to the
Progress Party, the Sámediggi is a case of ethnic discrimination. The party’s argument is that
the Sámi and other nationalities have lived side by side in the North for centuries, and that
none of these groups should be given special rights. Unlike the case of the Labour party,
campaigning for two simultaneous elections do not seem to create any difficulties for the
Progress party. The party’s criticism of so-called Sámi privileges are met with response from
many non-Sámi voters in the traditional Sámi settlement area, but it is noteworthy that the
Progress Party also finds support within parts of the Sámi community.
So far, we have looked into the policies and arguments of the political actors. Gaski
concluded her discourse analysis of the 2005 election campaign that although the political
actors argued on the basis of self-determination,
we still do not know anything about the attitudinal bases for cleavage formation in the
Sami society; if the framing of the opposition lines which are initiated by the political
actors reflects attitudes in the electorate or has any resonance with the voters. But the
results indicate that the nature of social and ideological divisions which structure the
Sami party competition do not necessarily follow the patterns for cleavages in the old
democracies in the West. Rather one can expect cleavage formation along other lines,
reflecting the historical legacy of the Sami society (Gaski 2008: 17–18).
In the following analyses, we will look into the question raised by Gaski: What can survey
data from the Sámi Election Study tell us about political cleavages in Norwegian Sámi
politics?
8
Data: the 2009 Sámediggi Election Survey
The 2009 Norwegian Sámi Election Study was carried out in September–October 2009,
starting immediately after the 2009 Sámediggi election. The survey sample was drawn from
the Sámi electoral register. The data was collected by means of telephone interviews, where
the respondents could choose to be interviewed in either the Norwegian or Northern Sámi
languages. The fieldwork was carried out by TNS Gallup. 973 respondents participated in the
survey. 731 of these said that they had voted in the election, and provided an answer to the
question of who they had voted for. This gives us a sufficient number of respondents to
analyse voting for each of the four largest parties – Labour, NSR, Árja and the Progress Party
– separately.5
Attitudes and Political Differences among Sámediggi Voters
The Sámediggi election survey contains a limited number of attitudinal questions. We
therefore have to bear in mind the possibility that significant attitudinal or ideological
differences are present in Sámi society, without being covered by this survey. There are,
however, a number of policy questions in the survey pertaining to issues that have been
debated in the Sámediggi or in public debates in Sámi society at large. One example of that is
the question of ownership of natural resources in the traditionally Sámi areas of Northern
Norway. There is a growing mining industry in the area, that some feel should financially
compensate Sámi society for the use of natural resources. Another issue covered in the survey
deals with the education of Sámi children. Schools that have Sámi pupils, and that teach the
Sámi language, are also part of the Norwegian education system. There may be some
disagreement about the extent to which the Sámediggi should have the power to control the
curriculum and mode of education at these schools. There are also some questions about trust
in and satisfaction with the Sámediggi. The Progress Party policy of abolishing the Sámediggi
is covered in a survey question, and there is one question about the power of the Sámediggi
vis-à-vis the national Norwegian parliament.
5 The interviewers attempted to phone 6487 individuals. However, this apparently large non-response includes
both those who declined to be interviewed and those the interviewers could not reach (no answer, only
answering machines etc.). The percentages of men and women are approximately the same in the sample and the
electoral register, but our sample is geographically skewed: the constituency of Ávjovári is underrepresented
(15.4 % of the sample against 25.6 % of the electoral register), while the constituencies of Southern Norway and
Gáisi are somewhat overrepresented. Labour and NSR voters are overrepresented in the sample, whereas Árja
and the Progress Party are somewhat underrepresented.
9
The Sámediggi election survey furthermore includes a left–right self-placement index
on a scale from 0 to 10. As a proxy for political preferences in national Norwegian politics,
we have created a dummy of voting in the national parliamentary election.6 Left-of-centre
voting7 is coded as 1, all others are 0. Table 2 displays a rotated factor analysis of these
attitudinal variables in the Sámediggi election survey.
While the survey questions seem to cover a range of political issues, table 2 reveals
only two attitude dimensions in this set of questions.8 Factor 2 reveals a left–right ideological
divide that is central to Norwegian politics, which also appears to be present among Sámi
voters. We therefore expect left–right ideological placement to have an effect on voting in
Sámediggi elections.
Table 2 Factor analysis of attitudinal data in the Sámi election study of
2009. Varimax rotated factors
Factor 1 Factor 2
The Sámediggi should have substantial influence
over the education of Sámi children* -.59 -.12
The Sámediggi should have increased influence
over fisheries in Sápmi* -.80 -.08
The Norwegian parliament should have the power
to overturn decisions made by the Sámediggi* .69 .04
The mining industry in Sápmi should financially
compensate Sámi society* -.69 -.03
The Sámediggi should be closed down* .77 .23
The Sámediggi should have less influence over
natural resources in Sápmi* .73 .08
Political trust towards the Sámediggi
(0=no trust; 10=very high trust). .79 .10
How satisfied are you with the job the Sámediggi has done
since its formation? (1= very satisfied; 4=not satisfied) -.74 -.15
Left–right self-placement scale
(0=far left; 10=far right) -.16 -.85
Voted for a left-of-centre party in the Norwegian
parliamentary election (1=yes; 0=no) .07 .88
* 1=strongly agree; 5=strongly disagree
6 Voting occurred on the same day as the Sámediggi election.
7 This category includes three political parties: the Labour Party, the Socialist Left Party and the Red Party.
8 The survey included four other questions on political attitudes, concerning a) gender equality; b) the use of
cross-country motor vehicles; c) the political influence of inner Finnmark (the core Sámi settlement area) in the
Sámediggi, and d) whether Norwegian parties should participate in Sámediggi elections. Preliminary analyses
showed that these questions did not load clearly on any substantial factor, and they were not strongly linked to
party choice. These four questions are therefore not included in the analyses.
10
Factor 1 appears to be more complex, since it covers several policy issues. These
issues are analytically distinct from left–right ideology. The factor analysis indicates that they
are derived from some other underlying ideological or value dimension. What, then, do these
questions have in common, that would induce voters to answer them in a congruous manner?
While they deal with policy issues that are not directly linked, they could also be interpreted
as dealing with the role of the Sámediggi and of Sámi society more generally. On the one
hand, there may be a group of people who favour a strengthening of the Sámediggi, of other
Sámi institutions, and of Sámi society in general. While this group probably does not favour
the establishment of a Sámi nation, they do favour institutional and other changes that move
in the direction of nation-building. Such a group of voters would be inclined to say that the
Sámediggi should have increased power over any policy-area, to believe that the mining
industry should compensate Sámi society, and to express trust and satisfaction with the
Sámediggi.
One the other hand, there may be a group of voters who are sceptical of the Sámediggi
and believe that Sámi society has already gone too far in the direction of nation-building.
Such a group would express low levels of trust, and would not favour increased powers to the
Sámediggi. They may also be sceptical of Sámi claims of ownership of natural resources.
We create an index of the eight questions that load strongly on factor 1. We believe
this index taps into peoples’ beliefs about Sámi self-determination, so we dub it the ‘Sámi
self-determination index’. If there is a political cleavage within Sámi society related to the
issue of nation-building, we expect this index to have an effect on voting in the 2009
Sámediggi election.
The other major cleavage that we hypothesise has an effect on voting is left–right
political ideology. We will use the left–right self-placement scale to measure those political
differences. We will not include a measure of voting in the national parliamentary elections as
an independent variable when explaining voting in the Sámediggi election, as the two
variables are strongly linked.
Explaining Voting in the 2009 Sámediggi Election
Since the Sámediggi election survey of 2009 is the first of its kind, our analyses of voting in
that election will have to start from scratch when it comes to identifying relevant independent
variables. Our main goal is to test the effect of the two ideological dimensions: Sámi self-
11
determination and left–right ideology. However, we also look for other relevant explanatory
variables.
There is one feature of Sámediggi elections that especially make them distinct from
most other elections: the ethnically based electoral register. The number of people enrolled in
the register has grown steadily since its formation in 1989. While those that enrolled at the
time of the formation of the Sámediggi tended to reside in the core Sámi areas, the growth
that has occurred since comes mostly from other areas of Norway (Pettersen 2010). It seems
reasonable to expect that those who registered in 1989 have a stronger identity as Sámi
political activists than those who enrolled more recently. Generally speaking, the time of
registration in the electoral register is a politically relevant variable that could influence
voting. As a point of departure in our exploration of voting in the 2009 Sámediggi election,
we look at differences in voting by time of enrolment.
Ap
NSR
Arja
FrP
Other Sámi lists
Other
Norwegian
parties
0
10
20
30
40
1989 1990 - 1999 2000 - 2009
Figure 1 Voting in the Sámediggi election of 2009, by the
time of enrolment in the electoral register. Percent
12
Figure 1 reveals a number of notable differences in voting between those who registered as a
voter in 1989 compared to those who signed up at a later point. These differences reflect
changes in the fortunes of the parties through the history of the Sámediggi. The NSR was the
dominant party at the time of the formation of the Norwegian Sámi Parliament. While the
Labour Party started out as a smaller party, it has now outgrown the NSR in terms of vote-
share and representation in the Sámediggi.
The NSR clearly has its strongest supporters among those who registered in 1989,
while those who enrolled at a later stage tend to favour the Labour Party, and also to some
extent other national Norwegian political parties.9 Unless the trend that is evident for the NSR
is broken, this does not bode well for the fortunes of that party in future elections.
The overall trend in figure 1 is strong support for national Norwegian political parties
among newly enrolled voters, while Sámi lists, including the NSR, are favoured by those who
enrolled in 1989. This indicates that time of enrolment in the electoral register is a useful
independent variable when studying voting. Still, the differences in figure 1 could be
explained by a number of factors that are not just about the time of enrolment.
First, there are a number of social background variables that are related to time of
enrolment that could also be related to voting. Age is an obvious example. You cannot sign in
to the electoral register until the age of 18, so recently enrolled voters are on average quite a
bit younger than those who registered in 1989. We will also control for gender, since there
have been some historic variations in the gender-composition of the Sámi electorate.
Finally, we aim to control for some kind of identifier of Sámi identity. Multiple ethnic
identities are common in Norwegian Sámi society; especially so in the northernmost part of
the country. Those who enrolled in the electoral register at the time of the formation of the
Sámediggi may have had an especially strong and unequivocal sense of Sámi identity. First
language is a possible identifier of such an identity. People whose first language is Sámi are
likely to have a strong sense of Sámi identity. It should be noted, though, that the previous
Norwegian policy of suppression of the Sámi language has led a number of people and
families to abandon the Sámi language. By identifying people whose first language is Sámi as
having a strong Sámi identity, one has to keep in mind that such an identity is not the
exclusive prerogative of those who speak the language. With that caveat, we use first
language as an independent variable in our analyses.
9 Due to a small N we had to merge all these parties into a single category.
13
Given the differences in voting by time of enrolment, as well as the success of two
new political parties in 2009, we believe there may be a lack of satisfaction with ‘business as
usual’ in the Sámediggi among some groups of voters. The Sámi self-determination index
includes measures of political trust and satisfaction with the Sámediggi. What is not covered
in that index is satisfaction with the two main political parties and the job that they have done
in the last 20 years. We include two independent variables in the analyses that measure the
voters’ rating of the NSR and of the Labour Party on a 0 to 10 scale.
Furthermore, we make use of a question that asks the voters to pick one factor, of the
following three, that especially influenced their choice of party in the election: 1) political
issues, 2) individual candidates on the list, and 3) general trust in the party/list. We expect
Árja voters to pick alternative 1, ‘political issues’, to a larger extent than the other voters. As a
newly formed party, Árja could not rely on established party loyalties or trust in experienced
politicians (although some Árja candidates had experience from other parties). We use a
dummy-variable in the analysis to identify those who chose ‘political issues’ as a reason for
their vote choice.
Another possible reason for discontent with the work of the Sámediggi and of the two
main political parties could be the issues that have been prioritized over the years. A group of
survey questions asks about the Sámediggi’s priorities of various issues in previous years. The
respondents are asked if the Sámediggi have focused ‘too little’, ‘too much’ or ‘just right’ on
each specific issue. Among those issues are four that have been central to the work of the
Sámediggi since its founding: education, language, culture, and reindeer-herding. Responses
to the first three (education, language and culture) are correlated10
and load on a single factor
in a factor analysis, so we have merged those into a single index. We include the final issue-
rating with respect to reindeer-herding as a separate independent variable in our analyses.
These variables are indicative of peoples’ satisfactions with the issue priorities of the
Sámediggi. We expect supporters of the two new political parties to be sceptical of previous
issue priorities.
Multivariate analyses of voting in the 2009 Sámediggi election
An analysis of voting in the Sámediggi election survey is limited by the number of
respondents in the survey. 731 respondents answered the question about voting in the election,
10
The correlations (Pearson’s r) are: (education and language) .28, (education and culture) .21, and (language
and culture) .33.
14
of which a little over 200 voted for Labour, and a similar number for NSR. The third and
fourth largest parties Árja and the Progress Party received 62 and 48 votes respectively from
these respondents.
We analyse support for each of the four largest parties in the election by use of logistic
regression analysis. Logistic regression is preferable to Ordinary Least Squares (OLS)
regression in this case, given that we use dichotomous, and in some cases highly skewed,
dependent variables.
The first model in each regression contains four social background variables: gender,
age, time of enrolment in the electoral register, and first language. The second model
introduces the variables related to issues and issue priorities. Third, we include our two
measures of political cleavages: the Sámi self-determination index and the left–right self-
placement index. Finally, we control for rating of the NSR and the Labour Party. When
analysing voting for the Labour Party or the NSR, we exclude the rating of the same party as
an independent variable; it is too highly correlated with voting and therefore obscures our
other findings.
Table 3 displays the analysis of voting for the Labour Party in the 2009 Sámediggi
election. Of the social background variables, first language is the only significant predictor of
Labour Party support. People whose first language is Sámi tend not to vote for the Labour
Party. If first language is seen as an indicator of Sámi identity, this seems to characterize other
voters more than Labour Party voters.
Our second model indicates, in accordance with our expectations, that Labour Party
voters tend not to state that political issues are the reason for their vote choice. Rather, these
voters are motivated by trust in the party and individual candidates on the list. Model 3 shows,
unsurprisingly, that the left–right self-placement index has a significant effect on voting for
the Labour Party. The Labour Party is a left-of-centre party, which is also reflected in the
ideological composition of its voters. The Sámi self-determination index has a weak negative
effect on support for the Labour Party. These voters probably favour the status quo; they are
not looking for changes in the direction of Sámi nation-building.
Finally, rating of the NSR has a significant negative effect on Labour Party support.
This indicates that the two parties are seen as opposites. Those who favour the Labour Party
tend to be critical of the NSR. We expect to find a similar effect vis-à-vis the Labour Party
among NSR voters.
15
Table 3 Modelled support for the Labour Party in the Sámediggi election of 2009.
Logistic regression coefficients
Modell 1 Modell 2 Modell 3 Modell 4
Gender (male=1) -.09 -.08 -.01 -.05
Age (years) .01 .01 .01 .01
Enrolled in the electoral register in 1989? (1=yes;
0=later) -.18 -.22 -.17 -.12
First language (Sámi=1; Other=0) -.88 ** -.85 ** -.72 ** -.86 **
Reason for vote-choice: political issues (=1) -.37 * -.34 -.46 *
The Sámediggi has emphasized the issues of education
language and culture too much (=-1); just right (0); too
little (=1).
.30 .30 .33
The Sámediggi has emphasized the issue of reindeer
herding too much (=-1); just right (0); too little (=1). -.10 -.05 .01
Sámi self-determination index (1 to 10) -.10 * .07
Left (0) to right (10) self-placement index -.17 ** -.19 **