Australian Muslims’ orientations to secular society: Empirical exploration of theoretical classifications

Abstract

This article applies Q methodology in order to explore Australian Muslims’ orientations to the secular society in which they live. The analysis is guided by some theoretical claims that are made about the dispositions of Muslims who live in Western societies. While a simple ‘closed’ versus ‘open’ dichotomy has some plausibility, deeper investigation reveals four empirical types: respectively, semi-engaged, coexisting, assertively religious and untroubled participant. These four types vary in the same direction from the theoretical specification that informed the search for the type in question in a way that appears to reduce the tension between the type and the norms of secular society. Generalizations commonly made in both popular and scholarly discourse about the problematic character of Muslim orientations to secular society appear not to apply in Australia.

Keywords

Australian Muslims Islam Q methodology secular societies

The relationship between Islam and Western secular societies has received a great deal of both scholarly and popular attention in recent years. Most Western societies are home to substantial Islamic minorities, whose relationship and commitment to democratic and secular institutions and norms have been questioned by politicians and journalists (for a demonstration of how negative associations dominate media coverage of Muslims in the UK, see Greater London Authority, 2007: 17–30; for Australia, see Saniotis, 2004). This questioning can draw on Huntington’s (1996: 217) claim that because Islam seeks a total order that subsumes politics, Islam and secular society are incompatible, such that their relationship can only become part of the larger drama of the ‘clash of civilizations’. For Huntington, the global clash plays out at ‘fault lines’ where different communities encounter one another. Such lines are especially problematic where Islamic communities are a minority, because in Islam the foremost political membership is in the umma of all believers, not the state. Huntington’s critics could note that contemporary conflicts are the result of symbol manipulation rather than deeply rooted identities (Kaufman, 2001), but such manipulation by both radical Islamists and Western critics of Islam is rife. In Australia no less than other Western countries, fears of radicalism in Muslim communities that would exacerbate division have been raised in the media and occasionally exploited by politicians (Kolig and Kabir, 2008; Rane et al., 2011: 125–6). Some scholars for their part worry about second-generation Muslim immigrants seeking identity in an Islamization of individual and social life that risks isolation and provides fertile ground for radicalization (Leiken, 2005).

Rather than coming to a summary judgment about Islamic populations in their entirety, a more nuanced perspective would point to the variety of orientations toward the secular (and sometimes multicultural) context in Islamic communities in Western liberal democracies (see for example Modood, 2010; Morsi and Little, 2010). Cultures that look monolithic from the outside may prove internally polyvocal (Benhabib, 2002).

Here we examine the orientations and identities found within Muslim communities in Australia, to test claims that have been made about the kinds of attitudes that ought to be found. We deploy a well-developed approach to the systematic study of subjectivity, Q methodology,¹ appropriate because subjective dispositions are the main issue here.² Q methodology is based on a subject’s reaction to a set of statements about a particular domain. Individual profiles (‘Q sorts’) can be compiled and analysed statistically in a search for underlying patterns. There are of course other methods available, ranging from opinion surveys (Pew Global Attitudes Project, 2006) to in-depth interviewing (Poynting, 2009). Q methodology has a particular combination of interpretive depth and systematic statistical purview that makes it especially suited to the exploration of tough questions about the character and variety of dispositions within Muslim communities. We are aware of no comparable Q studies of Muslim minorities in Western societies, though Kanra (2005, 2009) reports on a Q study of secular and Islamist dispositions in Turkey. There are however many Q studies of identities in minority groups in Western societies: involving nationality (e.g. Davis, 1999 on Basques), religion (e.g. Fox, 2003), ethnicity (e.g. John and Montgomery, forthcoming), and sexuality (e.g. Kitzinger and Stainton Rogers, 1985).³

The intent is to search for empirical patterns that correspond to theoretical claims about the variety of orientations that exists. The first such claim is that we can distinguish between ‘open’ and ‘closed’ dispositions in Muslim communities (Celermajer et al., 2007). The second is that the four basic orientations can be categorized as isolationist, Islamist, fundamentalist and participant (drawing on Saeed, 2008; the labels are ours). We do our best to locate the kind of picture that has been conjectured by observers and theorists, but should also be prepared to find something different. Thus our inquiry is in a broad sense hypothesis testing, the hypotheses referring to the existence of particular orientations. These hypotheses are not causal in nature though. Q is in its essence an interpretive methodology (though, unlike most interpretive methods, constrained by quantitative results), and so cannot by itself test causal claims. Our interpretations will be guided by what others have said about the orientations that should exist, but will not stop there. In the empirical tests, there should be no false positives; we can be confident that any perspective we find does have a real existence among Australian Muslims. Any false negatives (i.e. failing to identify a perspective that does exist) would lie in inadequate variety in subject selection, which is why we devote a great deal of effort to maximizing this variety.

Our results show it is possible to distinguish plausibly between ‘open’ and ‘closed’ orientations to secular society within Muslim communities in Australia. However, a more fine-grained result locates four distinct types: respectively, semi-engaged, coexisting, assertively religious and untroubled participant. What is striking about all four types is that they depart in the same direction from the theoretical specification (respectively, isolationist, Islamist, fundamentalist, participant) that informed the search for the type in question. This direction appears to reduce tension between the type and the norms of secular society. We have no directly comparable evidence using the same method from other Western societies. But our results suggest that many of the generalizations noted at the outset, commonly made in both popular and scholarly discourse about the problematic character of Muslim orientations to secular society, do not apply in Australia.

Methodology

Q methodology begins by modelling an individual’s orientation to a particular topic or domain. In the case at hand the individuals are Australian Muslims, the domain is that of the relationship between Islamic communities and Australian secular society. Each individual subject is confronted with (in this study) 34 relevant statements from the domain in question and asked to order them into a set of nine categories from ‘most agree’ to ‘most disagree’ in a quasi-normal distribution (other studies can use different numbers of statements and categories). The ordinal ranking of statements an individual produces is a ‘Q sort’. There are thousands of ways that 34 statements can be sorted into nine categories, so the Q sort models the subject’s own subjective construction, carried out with substantial freedom. The Q sort requires interpretation by the analyst because an individual’s placement of any one statement in the nine categories only takes on meaning in relation to his or her placement of all the other statements. So, for example, an individual’s strong agreement with a statement such as ‘the essence of democracy is majority rule’ means something very particular if a subject’s reaction to other statements indicates that he or she equates ‘majority’ with the majority ethnic group in a society (see the study of discourses of democracy in Serbia reported in Dryzek and Holmes, 2002: 62–73). That is, the positive reaction to the statement about majority rule is not just endorsement of an abstract democratic principle, rather it also indicates a hierarchical relationship between ethnic groups.

For this study, the 34 statements used were selected to pick out key themes from a larger body of statements generated mainly in discussion groups, but supplemented by statements made on television, in newspaper articles and in letters columns. One discussion group was organized at an Islamic Centre and involved only Muslims, another group was made up of non-Muslims, two other groups mixed Muslims and non-Muslims. This mechanism for generating statements means they are all drawn from natural language, not made up by the researcher. The advantage here is that the subjects who subsequently perform the individual Q sorts can give an account of themselves in terms of the language of the society in which they live and talk, which is more meaningful than anything made up by the researcher (Dryzek and Berejikian, 1993; Kitzinger, 1986). The disadvantage of natural language is that some statements may be ambiguous, but ambiguity is resolved by how the subject reacts to one statement in the context of the other statements (Dryzek and Berejikian, 1993: 51), so the analyst can interpret what a particular statement actually means to the subject (as in the ‘majority rule’ example just mentioned). However, some Q methodologists prefer unambiguous statements; and, conversely, there is nothing to stop researchers using other techniques, such as sample surveys, from using natural language.

Patterns across the Q sorts of different individuals can be summarized and compared using factor analysis (Dziopa and Ahern, 2011: 40–1). While familiar ‘R’ factor analysis seeks patterns across variables, Q factor analysis seeks patterns across individuals (the Q/R terminology was established long ago in psychology, the letters do not actually mean anything). The results of the Q factor analysis should give us a picture of commonalities and differences in the orientations of Australian Muslims toward secular society. Each factor represents an ideal type, which can be described in terms of how a hypothetical individual loading 100 percent on that factor would order the 34 statements. In practice, we generally find some subjects have significant loadings on more than one factor (and almost never find an actual individual loading 100 percent on a factor).

Factor analysis (of any kind, Q or R) does not yield a single unique solution. Factors can be rotated using either statistical or judgemental criteria to look at the results from different points of view. This is actually an advantage in testing for the presence or absence of a theoretical point of view hypothesized by other scholars, such as those we will introduce below. We can rotate factors judgementally to try to find the theoretical point of view in question (Brown, 1980: 33). If the theoretical point of view has no significant presence, no amount of rotation will find it. If we look for a theoretical point of view postulated by a scholar but, no matter how hard we try, end up finding something different, that in itself is instructive.

Q is an intensive methodology and so normally works with small numbers of subjects (see any Q textbook, e.g. Brown, 1980: 191–4). If a sample of subjects is well selected with a view to obtaining the full variety of subjective positions in a population, then adding more subjects contributes little information. In its approach to generalization, Q is different from survey research, which can estimate the proportion of the population that subscribes to a particular point of view. However, generalization of a kind is still possible in Q because each factor is itself a generalization:

factor X demonstrates how persons of this type think in general. Q factors denote qualitative differences in perspective that increases in quantity cannot affect, any more than gallons of paint could somehow testify to differences between red and blue that could not be documented as well using a thimbleful of each. (Brown, 2009: 240)

Small numbers do not therefore compromise test-retest reliability and replicability, as empirical studies have shown (Nicholas, 2011).

For this study, 33 subjects were selected, maximizing variety in terms of ethnic background, length of residence in Australia, gender, age, involvement with Islamic organizations (from zero to extensive) and religiosity (from secular to devout). Interviews were carried out in Sydney, Canberra and Melbourne in 2008 and 2009. In Sydney, most of the interviews were done in Auburn, using snowball sampling in its large Lebanese and Turkish communities and smaller African community. In Melbourne, the participants were attendees at an Islamic conference. In Canberra, participants were mostly recruited via the Islamic Centre. Individuals with particular characteristics were sought using our contacts with the organizations in question and with individual members of particular communities (including those with no involvement in Islamic organizations).

The hypotheses

Surveying the articles in a special issue on ‘Australian Muslims and Secularism’ of the Australian Journal of Social Issues, Celermajer et al. (2007: 3–4) conclude that, reacting to current challenges in the media, public policy and public opinion:

For some [Muslims], this has meant vigorously asserting the compatibility of their faith commitments with full and active Australian citizenship. For others … it has meant turning to sources of information and discourses that place Islam at the centre, not at the margins, taking refuge in a more closed and separate sense of identity, and in some cases, imposing these more closed inflexible interpretations of Islamic identity on others, principally women.

What this summary assessment suggests is that our empirical inquiry should look first for two Muslim viewpoints, one open in its orientation to Australian secular society, the other closed. Our methodology does indeed enable us to look for a two-factor solution along these lines, and that is where we start.

Saeed (2008: 207–14) develops a more nuanced fourfold taxonomy of Muslim attitudes to their participation in Western societies (not just Australia). His categories (his title for each is a bit of a mouthful) are:

‘Ethno-national/traditionalist/non-ideological/isolationist’, who ‘have little interest in local issues: politics, economic or legal’ (2008: 209). We will call this type ‘isolationist’ for short, because it is isolation that ought to characterize its relationship to the broader society.

‘Transnational/semi-traditionalist/ideological/semi-isolationist’, who ‘emphasise Islamisation of every aspect of life’ (2008: 210), so let us call this category ‘Islamist’ for short.

‘Transnational/traditionalist/non-ideological/isolationist’, who ‘want an Islam that remains unquestioned by scholarly inquiry or threatened by “unconventional” interpretation’, and has a radical sub-type that is ‘agitating for a struggle against the West’ (2008: 211). Let us call this type ‘Fundamentalist’ for short, because its core motivation is a fixed view of Islam that cannot change in any confrontation with secular society.

‘Indigenous/non-traditionalist/non-ideological/participant’, who ‘are developing strategies to deal with the question of integration and adaptation without assimilation. For these Muslims in the West, Australia, for example, is not a foreign country but their home’ (2008: 212). So we can straightforwardly call this category ‘Participant’ for short.

If we compare these four theoretical types with the two types characterized by Celermajer et al. (2007: 3–4), type 4 is ‘open’, types 1, 2 and 3 are ‘closed’.

Saeed’s classification is based on his own intuitive grasp and personal knowledge of Islamic communities rather than systematic empirical study. We can undertake such a study. To begin, we can try to match his four types with a four-factor solution. At the outset, we allow that it would be astonishing to find perfect correspondence between four factors and Saeed’s four types – but they do provide useful guidance, even though we end up with a somewhat different set of four categories. First, however, let us examine a two-factor solution guided by the ‘open’ and ‘closed’ types specified by Celermajer et al. (2007).

Results: two-factor solution

For the two-factor solution, judgemental rotation of the two factors⁴ was oriented by the Q sorts of two individuals who in individual interviews seemed to exemplify key aspects of closed and open orientations.⁵ The ‘open’ individual was a self-described moderate, the ‘closed’ individual one who emphasized the need for Muslims to establish Sharia law, strongly prioritized religious practices over human rights, and strongly opposed the idea that mosques and schools preaching extremism should be closed down. (Support for Sharia law is not necessarily indicative of a closed position, given the variety of ways it can be implemented, and does not have to follow the hardline version applied in Saudi Arabia or Afghanistan under the Taliban; see Ahmed, 2009.)⁶ We now report on the differences between the two factors, based on how a hypothetical individual loading 100 percent on each factor would sort the 34 statements.⁷ We report only the statements whose score differs by 4 or more on the 9-point scale (which runs from -4, most disagree, to +4, most agree). These idealized Q sorts are obtained from the factor analysis of all 33 subjects, not from the Q sorts of the two individuals whose sorts guided the rotation. They represent how a hypothetical individual (archetype) loading 100 percent on the factor in question would order the statements, not any actual individual.

Table 1 suggests that there is indeed a clear difference between open and closed orientations among Muslim Australians. Factor A seeks integration into secular society, is happy with life in that society, rejects Sharia law, and opposes the idea that there must be conflict between Islam and the West. Factor B, in contrast, does not believe Muslims in Australia have equal rights with others, does not see effective integration as possible, seems not to be proud to be Australian, believes in the duty to pursue Sharia law, and rejects the idea that mosques and schools preaching extremism need to be closed down (probably because it would reject the notion of ‘extremism’ this statement implies).

Table 1.

Two-factor solution – factor scores of statements distinguishing between ‘open’ and ‘closed’ factors.

Statement	Factor AOpen	Factor BClosed
1. Muslims and non-Muslims have the same rights in Australia.	+3	−3
11. Muslims cannot be fully integrated into Australian society because some Islamic principles are against democratic and secular principles of Australian society such as women and gay rights.	−2	+2
12. If Muslims are expected to learn Australian values non-Muslim Australians should also learn Islamic values.	−1	+3
14. Australia’s democratic and secular values are an extension that reinforces the very basic values and concepts of our practice in Islam and I just wish to say I have no problems whatsoever. So living in Australia is an absolutely fantastic experience.	+2	−2
21. I am a Muslim and a proud Australian. I have never imposed my belief on others and have no intention of turning Australia into an Islamic state.	+4	−3
26. The ultimate goal for a Muslim should be to establish Sharia law no matter where they live.	−4	+3
27. Mosques and Islamic schools that are used for preaching hatred and extremism should be closed down.	0	−4
28. The conflict between Western and Islamic values will be there forever. Dialogue cannot fix the problem.	−3	+1

Note: Scale runs from –4 (disagree most strongly) to +4 (agree most strongly).

These results seemingly corroborate strongly the Celermajer et al. (2007) categorization of Australian Muslims into two clear camps. However, if we dig a little deeper, a more complex picture emerges. To begin, we had to try quite hard to isolate factor B. Of our 33 subjects, only four had a significant loading on factor B, without also loading significantly on factor A (at the .01 level of significance).⁸ Given we went out of our way to find variety among our subjects, and extracted factors specifically designed to locate ‘closed’ viewpoints, we still had a hard time finding closed viewpoints.

Further subtlety emerges when we look more closely at the profile for the ‘closed’ factor B. The statements reported in Table 1 are those for which there is a strong contrast between the two factors, thus highlighting the ‘closed’ aspects of factor B. But there are some statements which get a common reaction from the two factors. For example, statement 30, ‘If the media spent more time reporting on good Muslims, rather than radicals, common people would not be suspicious of every Muslim they see on the street’, gets the highest possible +4 ranking in factor B (and +3 in factor A). This indicates in factor B a desire for coexistence with non-Muslims, if not necessarily integration. Similarly, statement 13, ‘Most Australians are fearful of Muslims simply because they know nothing about Islam’, receives a +4 ranking in factor B, indicating perhaps a desire to be understood, if not accepted.

Our provisional conclusion is that the two-factor solution distinguishing between ‘open’ and ‘closed’ perspectives among Australian Muslims tells part of the story – but not the whole story, which we now flesh out by looking at the theoretically guided four-factor solution.

Results: four-factor solution

Judgemental rotation to a four-factor solution was informed by the perspectives of four individuals among the 33 interviewed who seemed best to express the four theoretical types specified by Saeed (2008), which we abbreviated to isolationist, Islamist, fundamentalist and participant. For these four theoretical types, the best fit we could find among our subjects was for a ‘participant’, actually the person who had the highest loading on the ‘open’ factor in our two-factor solution (not the same person we initially identified as ‘open’ to inform the two-factor rotation). The fit between any single subject and the other three theoretical types was less close. The closest we could find to a ‘fundamentalist’ was the same person who informed rotation toward the ‘closed’ factor in the two-factor solution, who stressed the need for Muslims to establish Sharia law and for religion to take precedence over human rights. For the ‘isolationist’ theoretical type, we located a person who was quite sceptical about Muslim engagement with secular society and Australian democracy in particular. For the ‘Islamist’ theoretical type we located an individual who believed strongly in Sharia law and the Islamization of life but did not want to impose this on others, and sought coexistence with secular Australia. The four-factor solution represents a picture of the factor space in its entirety, informed by a theoretical classification. It is superior in that it produces clearer differences between the factors than does the two-factor solution,⁹ but it is better to think of it as enabling more detail to be read, just as a large-scale map is more detailed than a small-scale map of the same territory.

The results for each of the four factors are reported in Tables 2 –5,¹⁰ and the four factors are compared in Table 6 (reporting only the statements whose score differs by 4 or more across any two factors). Again, the factor scores represent how a hypothetical individual loading 100 percent on the factor in question would rank the 34 statements. The four factors are reported in an order that preserves consistency with Saeed’s (2008) ordering of his ideal types.¹¹

Table 2.

Factor A, semi-engaged.

Defining statements (four-factor solution)
29. The media and politicians often create stereotypes about Muslims.	+4
30. If the media spent more time reporting on good Muslims rather than radicals, common people would not be suspicious of every Muslim they see on the street.	+4
21. I am a Muslim and a proud Australian. I have never imposed my beliefs on others and have no intention of turning Australia into a Muslim state.	+3
20. A real dialogue with Muslim people is the key to achieve their integration into Australian society. The effort should come from both sides.	+3
2. As long as Muslims are allowed to practise their religion there is no problem for them living in a secular society like Australia.	+3
3. Muslims never believe that Australia is their home. They only want to benefit from its wealth.	−3
18. Muslims are small in number but want to come here and change the Australian way of life.	−3
27. Mosques and Islamic schools that are used for preaching hatred should be closed down.	−3
26. The ultimate goal for a Muslim should be to establish Sharia law no matter where they live.	−4
5. Muslims should participate in Christmas celebrations to show their willingness to be part of Australian society.	−4

Note: Scale runs from –4 (disagree most strongly) to +4 (agree most strongly).

Table 3.

Factor B, coexisting.

Defining statements (four-factor solution)
2. As long as Muslims are allowed to practise their religion there is no problem for them living in a secular society like Australia.	+4
21. I am a Muslim and a proud Australian. I have never imposed my beliefs on others and have no intention of turning Australia into a Muslim state.	+4
14. Australia’s democratic and secular values are an extension that reinforces the very basic values and concepts of our practice in Islam and I just wish to say I have absolutely no problems whatsoever. So living in Australia is an absolutely fantastic experience.	+3
20. A real dialogue with Muslim people is the key to achieve their integration into Australian society. The effort should come from both sides.	+3
1. Muslims and non-Muslims have the same rights in Australia.	+3
12. If Muslims are expected to learn Australian values non-Muslim Australians should also learn Muslim values.	−3
27. Mosques and Islamic schools that are used for preaching hatred should be closed down.	−3
34. The more Muslims integrate into Australian society the less Muslim they become.	−3
3. Muslims never believe that Australia is their home. They only want to benefit from its wealth.	−4
5. Muslims should participate in Christmas celebrations to show their willingness to be part of Australian society.	−4

Table 4.

Factor C, assertively religious.

Defining statements (four-factor solution)
30. If the media spent more time reporting on good Muslims, rather than radicals, common people would not be suspicious of every Muslim they see on the street.	+4
13. Most Australians are fearful of Muslims simply because they know nothing about Islam.	+4
2. As long as Muslims are allowed to practise their religion there is no problem for them living in a secular society like Australia.	+3
12. If Muslims are expected to learn Australian values non-Muslim Australians should also learn Muslim values.	+3
29. The media and politicians often create stereotypes about Muslims.	+3
18. Muslims are small in number but want to come here and change the Australian way of life.	−3
19. Tolerance has its limits. Religious activities should be banned if they breach fundamental individual rights.	−3
34. The more Muslims integrate into Australian society the less Muslim they become.	−3
3. Muslims never believe that Australia is their home. They only want to benefit from its wealth.	−4
27. Mosques and Islamic schools that are used for preaching hatred should be closed down.	−4

Table 5.

Factor D, untroubled participant.

Defining statements (four-factor solution)
20. A real dialogue with Muslim people is the key to achieve their integration into Australian society. The effort should come from both sides.	+4
21. I am a Muslim and a proud Australian. I have never imposed my beliefs on others and have no intention of turning Australia into a Muslim state.	+4
16. It is only a small minority who creates problems, not the majority of Muslims. Therefore, division between Muslims and non-Muslims are not as deep as it is presented.	+3
29. The media and politicians often create stereotypes about Muslims.	+3
30. If the media spent more time reporting on good Muslims, rather than radicals, common people would not be suspicious of every Muslim they see on the street.	+3
3. Muslims never believe that Australia is their home. They only want to benefit from its wealth.	−3
11. Muslims cannot be fully integrated into Australian society because some Islamic values are against democratic and secular principles of Australian society such as women and gay rights.	−3
28. The conflict between Western and Islamic values will be there forever. Dialogue cannot fix the problem.	−3
15. Racism is alive and simmering in Australia and it is targeting Muslims.	−4
26. The ultimate goal for a Muslim should be to establish Sharia law no matter where they live.	−4

Table 6.

Four-factor solution, factor scores of statements distinguishing between factors.

Statement	Factor Asemi-engaged	Factor B coexisting	Factor C assertively religious	Factor Duntroubled participant
1. Muslims and non-Muslims have the same rights in Australia	+2	+3	−2	+2
2. As long as Muslims are allowed to practise their religion there is no problem for them living in a secular society like Australia.	+3	+4	+3	0
5. Muslims should participate in Christmas celebrations to show their willingness to be part of Australian society.	−4	−4	−2	0
11. Muslims cannot be fully integrated into Australian society because some Islamic values are against democratic and secular principles of Australian society such as women and gay rights.	+1	0	0	−3
12. If Muslims are expected to learn Australian values non-Muslim Australians should also learn Muslim values.	+1	−3	+3	−1
13. Most Australians are fearful of Muslims simply because they know nothing about Islam.	+1	−2	+4	0
14. Australia’s democratic and secular values are an extension that reinforces the very basic values and concepts of our practice in Islam and I just wish to say I have absolutely no problems whatsoever. So living in Australia is an absolutely fantastic experience.	+1	+3	−1	+2
15. Racism is alive and simmering in Australia and it is targeting Muslims.	+1	+1	+2	−4
21. I am a Muslim and a proud Australian. I have never imposed my beliefs on others and have no intention of turning Australia into a Muslim state	+3	+4	0	+4
26. The ultimate goal for a Muslim should be to establish Sharia law no matter where they live.	−4	+2	+1	−4
27. Mosques and Islamic schools that are used for preaching hatred should be closed down.	−3	−3	−4	+2
29. The media and politicians often create stereotypes about Muslims.	+4	0	+3	+3

The four-factor result showed that each of the four individuals targeted to most closely approximate the theoretical ideal type could load significantly at the .01 level on a factor that did not also have a significant loading from one of the other targeted individuals. Thus the fourfold theoretical typology of Muslims does have some plausibility, yielding empirically four points of view that are mutually exclusive in a statistical sense. However, a more nuanced and different story emerges once we examine the content of the four factors. Of the four theoretical types, only ‘participant’ emerges unscathed from the empirical analysis. The other three types need to be modified substantially in their meaning.

The extraction of Factor A was informed by an individual who seemed closest to an isolationist perspective. Table 2 shows factor A has some isolationist tendencies in seeking to keep Muslims distant from the larger society (for example, vehemently rejecting the idea that Muslims should participate in Christian celebrations). But the isolationist aspects in practice are much more strongly associated with Factor A’s clear opposition to any notion of imposing Muslim ideas on the rest of society, or changing anything about that larger society. Factor A does not really reflect the ‘little interest in local issues’ or yearning for Islam as it was ‘back home’ postulated by Saeed (2008: 209). It is not the wish to be isolated that really energizes factor A, but rather dismay about the way Muslims are portrayed by politicians and the media (see Table 2 statements 29 and 30). This corresponds with the survey finding reported by Rane et al. (2011: 133) that the media and the political system are the two Australian institutions that Australian Muslims trust least. A’s isolationism is further attenuated by its agreement (+3) with statement 20 about the need for dialogue. So, rather than being characterized as isolationist, or even semi-isolationist, Factor A is better re-titled semi-engaged.

The extraction of Factor B was informed by a subject with what we call an Islamist point of view. Islamism here follows Saeed’s (2008: 210) description of commitment to ‘Islamisation of every aspect of life’. Among the four factors, Factor B most supports the idea that it is the duty of Muslims to establish Sharia law (Table 6, statement 26) – a position strongly rejected by the semi-engaged Factor A, as well as Factor D. Factor B also strongly rejects the idea that Muslims should participate in Christian celebrations (Table 3, statement 5). But clearly Factor B only wants Sharia law for Muslims, for it strongly asserts there is otherwise no intention to impose Muslim ideas on the Australian state (statement 21). For the most part, Factor B is at peace with secular society and the place Muslims can occupy within it, and is much less anxious on such matters than Factor A. Indeed, semi-Islamist, let alone Islamist, is too strong a description to apply to it. Factor B is confident of its Islamic identity, but is best described as ‘coexisting’ in its orientation to secular society.

Factor C’s extraction was informed by an individual with a fundamentalist point of view, committed to a traditionalist and unquestioning view of religion. More so than the other factors, it does accept that there might be a clash between Muslim principles and secular Australian ones (Table 6, statement 14). But this acceptance falls far short of Saeed’s (2008: 211) postulate of complete rejection of the norms of Western society. Factor C believes that if non-Muslim Australians understood more about Islam they would not fear it (Table 4, statement 13), and that if there is to be an exchange between Muslims and non-Muslims, it should go both ways, such that non-Muslims should be required to learn Islamic values (Table 4, statement 12). Unlike the other factors, it does not believe Muslims currently have equal rights in Australia (Table 6, statement 1), but sees no problem in coexisting with secular society if Muslims can practise their religion. It is lukewarm about establishing Sharia law (Table 6, statement 26). This factor is assertive in that it wants any exchange with secular society to be truly reciprocal, rather than seeing it as a question of fitting in with secular society (which is what factor D seeks). If there is conflict between religion and secular rights, rights should give way (Table 4, statement 19). Overall, though, it is hard to retain the description ‘fundamentalist’ for this factor. Assertively religious is better.

Factor D for its part was informed in its extraction by a subject with a participant point of view. Unlike Factors A, B and C, the empirical result strongly corroborates the theoretical point of view, characterized by Saeed (2008: 214), in terms of its wish to take a full place in secular society, whose laws should be honoured. Factor D strongly rejects any incompatibility between Muslim values and secular democratic principles (Table 5, statement 11), as well as the idea that anti-Muslim racism is a problem in Australia (statement 15). It is strongly opposed to the idea that Muslims should try to introduce Sharia law (statement 26), perhaps on the grounds that it would be bad for Muslims as well as anyone else. Alone among the four factors, it agrees with the idea that mosques and schools preaching extremism should be closed down (Table 6, statement 27). Its main anxieties, shared with factor A, concern the way Muslims are portrayed by the media and politicians. Of the four factors, Factor D is most enthusiastic about integration with Australian society (Table 5, statement 11), and so the title untroubled participant is warranted.

Looking at the results in their entirety, there is a common and striking difference between the four types specified by Saeed (2008), and the four factors that result from the empirical analysis. Table 7 summarizes the differences in labels between those that our empirical analysis suggests is most appropriate, and the corresponding theoretical type that guided the factor rotation. To put it simply (using our terminology, not Saeed’s), in Australia the isolationists are not very isolationist, the Islamists are not very Islamist (to the point where the description ‘Islamist’ is actually not appropriate), and the fundamentalists are not very fundamentalist. However, the participants prove to be very participatory.

Table 7.

Contrast between theoretical types and types emerging from the four-factor solution.

Theoretical type	Type found empirically
Isolationist	Semi-engaged
Islamist	Coexisting
Fundamentalist	Assertively religious
Participant	Untroubled participant

The four types we found empirically can be further illuminated by the four types specified a priori in a study of British Muslims by Modood and Ahmad (2007) (which does not of course share our methodology). There is a good match between two of their types and two of ours. Our ‘assertively religious’ type is similar to their ‘Traditional Islam – reasoning from faith and first principles but doing so in the way of the traditional ulemma or, more likely, in a way not opposed to traditional Islamic learning’ (Modood and Ahmad, 2007: 194). Our ‘untroubled participant’ type resembles their:

Existential Muslim – arguing in a more existential and pragmatic way, for example, linking the communities and institutions that one belongs to.… Or to treat being ‘British Muslim’ as a hyphenated identity in which both parts are to be valued as important to oneself and one’s principles and belief commitments. (2007: 194)

The other two Modood and Ahmad types are:

Modernist Islam – reasoning from faith and first principles, but doing so in ways that draw upon modernist ideas within an Islamic methodology (2007: 194)

and

Philosophical Muslim – reasoning from first principles but without much systematic reference to Islam and drawing more on modern western theory, ethics and principles, including arguments about multiculturalism, equality, and so on. (2007: 194)

It would be neat if we could map these two types onto our semi-engaged and coexisting types, but really none of the Modood and Ahmad categories quite captures our ‘coexisting’ type. And both ‘Modernist Islam’ and ‘Philosophical Muslim’ are somewhat more insistent than our ‘semi-engaged’ type on the need to argue positions in society’s larger communicative processes. This lack of a precise match reflects the fact that the Modood and Ahmad categories are developed in a study of high-profile Muslim intellectuals and activists, whereas ours is a study of ordinary people (only a few of whom have any public profile or activist leanings). Still, the very fact that we can draw some connections between the four Modood and Ahmad categories and our four types is instructive because their study is explicitly confined to ‘moderate Muslims’ (a term they use apologetically [2007: 193–4], but cannot come up with a better one). In comparative light, the universe of discourses among Australian Muslims seems to correspond roughly to what is in the UK the set of moderate discourses. This drives home our main finding about what the four-factor solution reveals: that in Australia, any expectations based on Muslims’ orientations in other Western societies are not confirmed. As we have pointed out, compared to expectations, in Australia the isolationists are not very isolationist, the Islamists are not very Islamist, the fundamentalists are not very fundamentalist – but the participants are very participatory. Quite how matters got to be that way in Australia lies beyond the scope of the analysis presented in this article. Morsi and Little (2009) argue that liberalism in Australia has been successful in affecting or even producing Muslim identities, leading to the depoliticization and privatization of faith. (While Morsi and Little condemn this development because they dislike liberalism, if what they say is correct, it is by no means obvious that it is a bad thing, compared with the more isolating alternatives.) On the other hand, the cause might lie in the diverse sources of Muslim migration to Australia; or in the policies of government; or in some combination of these factors.

Conclusion

At a general level, the applicability of a distinction between ‘open’ and ‘closed’ Muslim perspectives toward Australian society was corroborated by our two-factor solution, though the ‘closed’ factor was actually quite attenuated in any rejection of Australian society and its secular norms. The four-factor solution reveals a more interesting story. While there does exist an untroubled participant orientation strongly related to an ‘open’ perspective, three other orientations can be found. Departing from theoretical specification of what these three might be, we identified them as, respectively, semi-engaged, coexisting, and assertively religious. In all four cases, the factor derived from the Australian results is less problematic in its orientation to secular society than the theoretical expectation. And while we have no directly comparable empirical evidence using the same methodology from other Western societies, we suspect that Australian Muslim communities may be less problematic in this respect than their counterparts in other countries. The Modood and Ahmad British study provides some support for this suspicion. The relationship between Australian Muslims and secular society is certainly less problematic than the theoretical specification of four types would lead us to expect. It would be very instructive to carry out a systematic comparative study, taking the methodology we have developed in this article and applying it to Islamic communities in other Western societies.

This study is not a sample survey, so we cannot say what proportion of Australian Muslims subscribes to each of the four factors we have identified empirically. That is the nature of an interpretive study and a methodology that seeks interpretive depth. But given that, for all four factors, the results point to less problematic relationships between Muslims and secular society than the theoretical specification would lead us to anticipate, we are confident that the results do tell us something important about the relationship between Muslims and secular society in Australia in general. There is considerable variety and nuance in the orientation of Australian Muslims to the larger society in which they live, but the kinds of extremes occasionally postulated in political discourse and the media do not seem to have a substantial presence. In particular, there is no evidence that Huntington’s (1996) ‘clash of civilizations’ has any explanatory power at all when it comes to Australian Muslims. Fears expressed in the media and occasionally in more scholarly works (Leiken, 2005) of a generalized Islamic radicalization do not stand up in light of the kind of polyvocality (not the simple fact of polyvocality) we have found among Australian Muslims.

Footnotes

Acknowledgements

For help and advice, we thank Selen Ayirtman Ercan, Steven Brown, and Simon Niemeyer.

Funding

This research was supported by Australian Research Council Discovery Grant DP0773626.

Notes

Author biographies

John Dryzek is Australian Research Council Federation Fellow and Professor, School of Politics and International Relations, Australian National University. Recent books include Foundations and Frontiers of Deliberative Governance (Oxford University Press, 2010).

Bora Kanra is Visiting Research Fellow, Centre for Deliberative Democracy and Global Governance, Australian National University. He is the author of Islam, Democracy and Dialogue in Turkey: Deliberating in Divided Societies (Ashgate, 2009).

References

Ahmed

(2009) ‘The Taliban’s Perversion of Sharia Law’, The Guardian 3 May.

Atkin

(1974) Mathematical Structure in Human Affairs. London: Heinemann Educational.

Benhabib

(2002) The Claims of Culture: Equality and Diversity in the Global Era. Princeton, NJ: Princeton University Press.

Brown

S.R.

(1980) Political Subjectivity: Applications of Q Methodology in Political Science. New Haven, CT: Yale University Press.

Brown

S.R.

(1984) ‘Q Analysis: Caveat Emptor’, Operant Subjectivity 8(1): 6–17.

Brown

S.R.

(2006) ‘A Match Made in Heaven: A Marginalized Methodology for Studying the Marginalized’, Quality and Quantity 40: 361–82.

Brown

S.R.

(2009) ‘Q Technique, Method and Methodology’, Field Methods 21: 238–41.

Celermajer

Yasmeen

Saeed

(2007) ‘Introduction’, Special Issue: ‘Australian Muslims and Secularism’, Australian Journal of Social Issues 42(1): 3–6.

Davis

T.C.

(1999) ‘Revisiting Group Attachment: Ethnic and National Identity’, Political Psychology 20: 25–47.

10.

Dryzek

J.S.

Berejikian

(1993) ‘Reconstructive Democratic Theory’, American Political Science Review 87(1): 48–60.

11.

Dryzek

J.S.

Holmes

(2002) Post-Communist Democratization: Political Discourses across Thirteen Countries. Cambridge: Cambridge University Press.

12.

Dziopa

Ahern

(2011) ‘A Systematic Literature Review of Q Technique and its Methodology’, Methodology 7(2): 39–55.

13.

Fox

J.C.

(2003) ‘A Typology of LDS Sociopolitical Worldviews’, Journal for the Scientific Study of Religion 42(2): 279–89.

14.

Greater London Authority (2007) The Search for Common Ground: Muslims, Non-Muslims, and the UK Media. London: Greater London Authority.

15.

Huntington

S.P.

(1996) The Clash of Civilizations and the Remaking of World Order. New York: Simon and Schuster.

16.

John

Montgomery

(forthcoming) ‘Socialization Goals of First-Generation Immigrant Indian Parents: A Q-Methodological Study’, Asian-American Journal of Psychology 3, forthcoming.

17.

Kanra

(2005) ‘Democracy, Islam and Dialogue: The Case of Turkey’, Government and Opposition 40(4): 515–40.

18.

Kanra

(2009) Islam, Democracy and Dialogue in Turkey: Deliberating in Divided Societies. Aldershot: Ashgate.

19.

Kaufman

S.J.

(2001) Modern Hatreds: The Symbolic Politics of Ethnic War. Ithaca, NY: Cornell University Press.

20.

Kitzinger

(1986) ‘Introducing and Developing Q as a Feminist Methodology’, pp. 151–72 in Wilkinson

(ed.) Feminist Social Psychology. Milton Keynes: Open University Press.

21.

Kitzinger

Stainton Rogers

(1985) ‘A Q Methodological Study of Lesbian Identities’, European Journal of Social Psychology 15: 167–87.

22.

Kolig

Kabir

(2008) ‘Not Friend, Not Foe: The Rocky Road of Enfranchisement of Muslims into Multicultural Nationhood in Australia and New Zealand’, Immigrants and Minorities 26: 266–300.

23.

Leiken

R.S.

(2005) ‘Europe’s Angry Muslims’, Foreign Affairs 84(4): 120–35.

24.

McKeown

Thomas

(1988) Q Methodology. Newbury Park, CA: Sage.

25.

Modood

(2010) ‘Multicultural Citizenship and Muslim Identity Politics’, Interventions 12: 157–70.

26.

Modood

Ahmad

(2007) ‘British Muslim Perspectives on Multiculturalism’, Theory, Culture & Society 24: 187–213.

27.

Morsi

Little

(2010) ‘“Just Getting on with my Life”: The Politics of Muslim Identity in Australia’, paper presented at the Australasian Political Studies Association Annual Conference, Melbourne, 27–9 September.

28.

Nicholas

(2011) ‘Reliability in Q Methodology: A Case Study’, paper presented at the Eastern Education Research Association Annual Conference, Sarasota, FL, 25 February.

29.

Pew Global Attitudes Project (2006) The Great Divide: How Westerners and Muslims View Each Other. Washington, DC: Pew Research Center.

30.

Poynting

(2009) ‘The “Lost” Girls: Muslim Young Women in Australia’, Journal of Intercultural Studies 30: 373–86.

31.

Rane

Nathie

Isakhan

Abdalla

(2011) ‘Towards Understanding what Australia’s Muslims Really Think’, Journal of Sociology 47(2): 123–43.

32.

Saeed

(2008) ‘Muslims in the West and their Attitudes to Full Participation in Western Societies: Some Reflections’, pp. 200–15 in Levey

G.B.

Modood

(eds) Secularism, Religion and Multicultural Citizenship. Cambridge: Cambridge University Press.

33.

Saniotis

(2004) ‘Embodying Ambivalence: Muslim Australians as “Other”’, Journal of Australian Studies 28 (82): 49–59.