Content uploaded by Luke Mansillo
Author content
All content in this area was uploaded by Luke Mansillo on Dec 22, 2017
Content may be subject to copyright.
Loyal to the Crown: shifting public opinion towards the
monarchy in Australia
Luke Mansillo
Department of Government and International Relations, University of Sydney, Sydney, Australia
Over the past half century, the Australian public has remained
divided on the issue of whether Australia should retain the
monarchy or become a republic. Clive Bean found that there had
been remarkable stability on the issue and evidence of a long-
term trend away from support for the monarchy with a sudden
decline in 1992. This article adopts Bean’s longitudinal cross-
sectional methods to examine the social and political basis of
public attitudes. This article analyses the Australian Election Study
(1993–2013) to compare Bean’s results and re-analyse earlier data
from the National Social Science Surveys and Australian National
Political Attitudes surveys (1967–90). Public opinion has been fluid
and is now at a crossroads between the 1980s high and the 1990s
lows. Cohort analysis suggests socialisation impacts long-term
opinions. Gender and ethnic nationalism also influences opinion.
ARTICLE HISTORY
Accepted 19 November 2015
KEYWORDS
Australia; banal nationalism;
monarchy; public opinion;
republic
Successive surveys in Australia have shown increasing levels of support for the monarchy
after a long period where support for the same was widely seen to be in an inexorable
decline. This perception remains in spite of evidence to contrary, notably the failure of
the 1999 referendum to replace the Queen with a Republican government.
This article seeks to explain this increased support for the monarchy utilising 46 years of
academic public opinion data. It compares Bean’s(1993) historical approach to trends in
support for the monarchy with more recent data. The article posits the social forces of pol-
itical socialisation and banal nationalism shape support for the monarchy, and suggests royal
misbehaviour that problematised monarchy as an issue explains short-term opinion change.
An increase in support for the monarchy in the 2000s after the lows of the 1990s should
not surprise even the casual observer of Australian public opinion. Compared with earlier
decades, the 1990s featured significant royal scandals. These include the Queen’s self-pro-
claimed 1992 annus horribilis, Charles and Diana’s 1996 divorce, and Diana’s death in
1997. In parallel with these events, support for the monarchy in Australia precipitously
declined from 58 per cent in 1990 to 40 per cent in 1993, and then declined to 34 per
cent in 1998. Since then, support has risen. By 2010, 42 per cent of the population surveyed
expressed support for the monarchy. From 2010 to 2013, there was another increase in
© 2016 Australian Political Studies Association
CONTACT Luke Mansillo luke.mansillo@Sydney.edu.au
Current address: Department of Government and International Relations, University of Sydney, Mereweather Building, City
Road, Darlington 2006, Sydney, Australia.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE, 2016
http://dx.doi.org/10.1080/10361146.2015.1123674
support –from 42 to 47 per cent. This was lower than the high point of 60 per cent in
1987–88, but significantly higher than the level of support in 1998.
More recent years saw considerable improvements in the public profile of the monarchy,
culminating in Prince William’s wedding to Catherine Middleton in 2011, and the births of
Prince George and Princess Charlotte in 2013 and 2015. This article argues that this
improved profile and the declining number and frequency of royal scandals improved
the Australian public’s image of the monarchy. This enabled younger cohorts of Australians
to develop more positive attitudes towards the monarchy and exposure to everyday rep-
resentations of monarchy in the state’s symbolism –banal nationalism –to repair older
cohorts’attitudes.
This article tracks the social bases of attitudes towards the monarchy over the last half
century. It compares political socialisation cohort effects over time. It tests the role of
civic and ethnic nationalisms in determining attitudes. It disaggregates opinions by
gender to account for differences. Finally, it discusses the possible explanations for
overall opinion movements including royal misbehaviour and government actions.
Trends in public opinion
Fourteen academic surveys, namely the Australian National Political Attitudes study
(ANPA) (1967–79), the National Social Science Surveys (NSSS) (1984–90) and the Austra-
lian Election Study (AES) (1993–2013), were used in this article with two questions on the
monarchy.
1
The first asks respondents ‘how important do [they] feel the Queen and the
Royal Family are to Australia’and the second asks if ‘[they] think that Australia should
become a republic with an Australian head of state, or should the Queen be retained as
head of state’.
2
Overall, since the referendum, monarchism has improved in both absolute
numbers and in its strength. There was stable support for the monarchy for both measures
up to 1990 (see Table 1 and Figure 1). This is reinforced by commercial polling (Figures 2
and 3).
3
Commercial polling shows a leap from 36 per cent republic approval in 1991 to 57
per cent in 1992 (Bean 1993: 196). Goot (1994: 65) also nominates 1992 as when ‘a high level
of support for the republic’emerged. Surprisingly, royal misbehaviour has been a neglected
explanation.
4
The strength of support for the republic has declined. In 1998, 34 per cent believed that
Australia should ‘definitely become a republic’; by 2013 this was 26 per cent. Since 1990,
the year 2013 is the high water mark of both weak and strong monarchist support. There is
a similar level of strong republicans in 2013 to 1993. Weak republicans have fallen away
declining from 33 to 27 per cent. Those who were on the fence but not really did not hold
an opinion -only offering one because they were asked -appear to have shifted, not those
with firm views. The annus horribilis’damage may have been recovered. Wellings (2003:
47–48) observed ‘the failure of the republican campaign cannot be seen as a victory for
monarchism or Australian attachment to Britain’. This recovery, in both total support
and strength, must be seen as a victory for the Palace.
Methodology
This article has four analyses. First, I repeat Bean’s analysis with AES data from 1993 to
2013. Second, I replicate and extend Bean’s cohort analysis with 20-year cohorts and
2L. MANSILLO
Table 1. Attitudes towards the monarchy–republic issue, 1967–2013 (%).
a
1967 1979 1984–85 1986–87 1987–88 1990 1993 1996 1998 2001 2004 2007 2010 2013
Importance of the Queen and Royal Family
Very important 28 25 16 20 18 17 13 12 10 10 10 11 13 16
Fairly important 26 29 29 29 25 28 22 27 21 21 23 25 25 28
Not very important 47 46 31 32 32 31 65 61 70 69 68 64 63 56
Not important at all –– 24 21 25 24 ––––––––
(n) (2007) (1996) (2979) (1517) (1646) (2488) (2335) (1797) (1856) (1980) (1731) (1850) (2055) (3865)
Australia become a Republic
Definitely retain the Queen –– 35 32 33 31 14 12 9 11 11 10 12 15
Probably retain the Queen –– 24 27 26 27 26 29 25 25 27 30 30 33
(total retain Queen) –(64) (59) (59) (60) (58) (40) (41) (34) (36) (38) (40) (42) (47)
Probably become a Republic –– 20 20 22 22 33 30 32 26 29 29 31 27
Definitely become a Republic –– 21 20 19 19 27 29 34 38 33 31 27 26
(total become a Republic) –(36) (41) (41) (40) (42) (60) (59) (66) (64) (62) (60) (58) (53)
(n)–(1864) (2907) (1512) (1640) (2481) (2334) (1730) (1833) (1960) (1722) (1830) (2047) (3841)
a
Percentages do not always sum exactly to subtotals because of rounding.
Sources: Bean (1993); Australian Election Study, 1993–2013.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 3
Figure 1. Public opinion on the monarchy. A. Retain the monarchy. B. Importance of the monarchy.
4L. MANSILLO
Figure 2. Support for the Monarchy, 1969–2014, Newspoll & Roy Morgan with a locally weighted
regression curve trend; n=57.
Figure 3. Support for the Republic, 1969–2014, Newspoll & Roy Morgan with a locally weighted
regression curve trend; n = 57.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 5
compare that analysis to a modified analysis with cohorts structured around eight-year
periods to capture adolescent political socialisation effects. Third, I estimate the effect
of nationalism on attitudes with 1996 and 2001 AES data. Fourth, I disaggregate the
model by gender at high- and low-saliency periods.
The data used are from the ANPA (1967–79), NSSS (1984–90) and AES (1993–2013)
academic public opinion surveys (Aitkin 2004; Aitkin, Kahan and Stokes 2011; Bean, Gow
and McAllister 1998,2001; Bean et al. 2004,2007; Jones, McAllister and Gow 1996; Jones
et al. 1993; Kelley, Bean and Evans 2004a,2004b; McAllister and Mughan 1987; McAllister
et al. 1990, 2010,2013). The Bean (1993) results are reproduced to show the complete time
period. The four analyses use the same variables with a few exceptions. Factor and
reliability analyses indicate that both questions, on the importance of –and whether to
retain the monarchy, are closely related measures of the same underlying concept. The
measures are combined into a simple additive scale that is coded 0 (most pro-republican)
to 1 (most pro-monarchy).
5
All independent variables are dummy coded with the excep-
tions of respondent age, years of education and church attendance as a measure of religi-
osity.
6
The second analysis investigates cohort effects and differs from the first analysis
with years of age replaced with a series of birth-year cohorts coded as dummy variables.
The third analysis investigates the effects of ethnic and civic nationalisms on attitudes. The
independent variables are identical to the first analysis with the addition of civic and ethnic
nationalism scales. The fourth analysis investigates gender through disaggregation at
periods of high and low issue salience.
The four analyses use a wide range of social background factors which influence Aus-
tralian political attitudes and behaviour (McAllister 1992;2011). Each employs an ordin-
ary least squares regression to estimate the effect of each independent variable on attitudes.
The unstandardised regression coefficient increases are interpretable as proportional
increases in pro-monarchy (and anti-republican) sentiments for each unit of change in
the dependent variable.
Public opinion is predicted with the following social background attributes: parental
party preference, gender, age, occupation, subjective social class, religion, church attend-
ance, urban–rural residence, state, birthplace and political party identity. The first analysis
identifies the long-term relationship between each observed social compositional factor
and attitudes. The second analysis on cohorts is in two parts. First, it extends Bean’s
(1993: 202) cohort analysis and second constructs a version to test political socialisation’s
effect. I repeat the 20-year cohorts Bean employed and compare his model to a modified
set of eight-year cohorts to test political socialisation effects between the ages of 12 and 18
(Torney-Purta 2005).
7
The third analysis determines if there is a relationship between
monarchism and two nationalism components: ethnic and civic nationalism. These
measures were produced using principal component analysis from a seven-question
battery included in the 1996 and 2001 AES (Appendix 1). The fourth analysis disaggre-
gates the first model to compare gender effects by analysing men and women separately.
Social and political underpinnings of monarchy–republic attitudes
The first analysis in Table 2 reveals that from 1996 to 2010 parental socialisation did not
have a significant relationship to attitudes on the monarchy. In 2013, parental socialisation
re-emerged as a factor of support for the monarchy after the royal marriage and Prince
6L. MANSILLO
Table 2. Social background and attitudes towards the monarchy–republic issue
a
by year, 1967–2013: unstandardised regression coefficients.
b
1967 1979 1984–85 1986–87 1987–88 1990 1993 1996 1998 2001 2004 2007 2010 2013
Parents’party preference (conservative) .06* .02 .04* .07* .04+ .03+ .05** .02 .05* −.03 .01 .03+ .00 .03*
Gender (male) −.10* −.03+ −.07* −.07* −.04* −.06* −.03* −.01 −.04* .02 −.03+ −.03* −.06*** −.05***
Age (years) .004* .003* .004* .004* .003* .004* .003*** .001** .001** .001** .002*** .000 .000 .001*
Education (years) −.009* −.006* −.018* −.011* −.006+ −.005* −.004 −.009* −.005 −.013*** −.013*** −.014*** −.008* −.007**
Occupation (non-manual) −.03 .01 .01 .01 .01 −.03 −.02 −.04* −.03+ −.04** −.05*** −.02 −.04** −.05***
Subjective social class (middle) −.04* .03+ .02 −.05* .01 −.00 −.06*** −.06** .05** −.06+ .00 .03* −.03* −.02+
Trade union membership .02 −.01 −.01 −.02 −.02 −.03+ −.04** −.01 −.02 −.03 −.03 −.02 −.01 −.01
Religion (Ref: Catholic and other) –– – – – –– – – – – – – –
Protestant .10* .10* .10* .09* .13* .10* .08*** .06** .09*** .08*** .07*** .09*** .11*** .12***
No religion −.11* −.00 −.02 −.08* −.07* −.08* .00 .00 −.00 −.02 −.04+ −.03 −.04* −.03*
Church attendance .10* .08* .15* .17* .15* .10* .03*** .02** .02** .03*** .01 .02** .02* .01*
Urban–rural residence (urban) −.03+ −.02 −.03* −.03 −.03 .00 .01 −.04* −.05** −.06*** −.01 −.03* −.04** −.04**
State (Ref: NSW) –– – – – –– – – – – – – –
Victoria .02 .01 −.01 .05* −.02 .03+ .02 −.03 .02 −.01 −.03 −.01 −.02 .02
Queensland .00 .02 .07* .05+ .02 .05* .04+ .00 .05* −.01 .02 .00 −.02 .03*
South Australia −.04 .01 .04+ .07* −.01 .01 .03 −.01 .05+ −.05 −.03 −.01 −.00 .05***
Western Australia −.02 .04+ .01 .02 −.00 .02 .02 −.02 .03 −.02 −.02 −.02 .00 .01
Tasmania −.03 .08* .08* .05 .01 .05+ −.01 .04 .00 .04 −.02 −.01 .10** −.02
Birthplace (Ref: Australia and other) –– – – – –– – – – – – – –
British Isles .04 .09* .13* .16* .09* .10* .10*** .10*** .08*** .10*** .05+ .07** .12*** .07***
Continental Europe −.03 −.13* −.18* −.17* −.16* −.14* −.08** −.10** −.05 −.12** .07 −.07+ −.06+ −.03
Asia –−.04 −.01 −.12+ .02 −.09+ −.03 −.10* −.01 −.06* −.09+ −.03 −.03 −.02
Political identity (conservative) .03 .16* .18* .21* .16* .15* .22*** .19*** .09*** .12*** .14*** .14*** .15*** .13***
Constant .36 .31 .35 .28 .28 .29 .08*** .24*** .19*** .21*** .21*** .26*** .37*** .34***
R
2
.12 .25 .32 .32 .26 .24 .28 .20 .14 .17 .17 .18 .18 .16
(n) (2054) (2016) (3012) (1528) (1663) (2504) (1737) (1364) (1424) (1532) (1311) (1446) (1747) (3225)
Mean on monarchy–republic scale .53
c
.56 .51 .53 .52 .51 .33 .33 .28 .28 .30 .31 .34 .38
(standard deviation) (.36) (.30) (.34) (.34) (.34) (.34) (.32) (.31) (.29) (.30) (.30) (.30) (.31) (.32)
a
Dependent variable scaled from 0 (most pro-republic) to 1 (most pro-monarchy).
b
+ significant at p= .10; *p= .05; ** p= .01; ***p= .001.
c
All figures have been rounded to two decimal places, with the exception of age and years of education which are to three decimal places.
d
The means and standard deviations of the dependent variables at each time point show consistency. There are minor measurement differences between the ANPA, NSSS and AES,
meaning that the means are not a good guide of the level of public support for the monarchy.
Sources: Bean (1993); Australian Election Study, 1993–2013.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 7
George’s birth. Conservative partisanship has been a stable predictor for support for the
monarchy.
8
Bean (1993: 205) concluded that the partisan divide would be ‘biggest hurdle’
to overcome. The conservative partisanship hurdle republican activists failed to leap.
Parental socialisation since 2010 has regained influence on public opinion. Generally,
older Australians are more monarchist; age had a larger effect from 1967 to 1993 but
since 1996, the effect has significantly lessened. In 2007 and 2010, there was no significant
relationship for age and a small effect in 2013. Urban dwellers remain more republican as
are the better educated. There are more male republicans; the resilient gender divide is
bemusing with women’s education levels reaching parity with men (Evans 1980) along
with their political attitudes (McAllister 2011: 112–20).
Class continues to make a stable impression on attitudes. Those who identified with the
middle-class or non-manual workers were relatively antithetical towards the monarchy
and stable over time. Trade union membership is a poor indicator of support; 1993 was
an exception. Protestant identity remains a stable and strong predictor of support.
Those who claim no religion tend to support republicanism. Religiosity, measured as
church attendance, has become a weaker indicator of support. This I interpret as a
long-term secularisation on the issue.
A person’s birthplace has an enduring effect on opinion. Those born in Asia have lower
levels of attachment to the monarchy; this has weakened in recent years. Those born in
continental Europe were more republican but this trend declined from the 1980s to
2001 and is no longer significant. Those born in the United Kingdom or Ireland retain
a large degree of support compared to those Australian born. Levels of support from
migrants from the British Isles remain stable unlike those from Asia or continental
Europe. This could be an effect on migrants associated with John Howard’s manipulation
of Australian identity to reflect the values of the Anglo–Celtic majority (Johnson 2007);
something this article expands upon.
Some states retain greater residual support for the monarchy. Queensland and South
Australia have had higher support. In 2010, there were high levels of support in Tasmania
when the pregnant Danish Crown Princess Mary, a Tasmanian by birth, visited Hobart the
week prior to the election. Royals apart from the Windsors impact attitudes towards the
Crown.
Cohort analysis: generational political socialisation and attitudes
Table 3 shows in the most impressionable years –being from 12 to 18 –significant
national events have a disproportionate and enduring impact on public opinion (Jennings
and Zhang 2005; Schuman, Akiyama and Knäuper 1998; Schuman and Rogers 2004;
Torney-Purta 2005). Critical events leave their mark on children and adolescence that
quickly change political views. Dennis and Webster (1975) found the opinions towards
the American president of those as young as seven years old shifted after the Watergate
scandal. Sears and Valentino (1997) hold that pre-adult socialisation produces stable pre-
dispositions that are catalysed by exogenous political events. Events socialise attitudes only
when they are salient. This allows longstanding predispositions to be socialised episodi-
cally rather than incrementally. The 1975 Whitlam dismissal, the 1992 annus horribilis
and lead-up to the 1999 referendum provide three critical episodes; first misuse of royal
prerogative, second royal misbehaviour and third campaigning in the lead-up to the
8L. MANSILLO
Table 3. Cohort analyses.
1967 1979 1984–85 1986–87 1987–88 1990
a
1993 1996 1998 2001 2004 2007 2010 2013
Bean’s cohort analysis
Born before 1920 .13*** .10*** .08*** .06* .10*** .02 .05+ .02 .11* .16** −.02 .06 −.04 −.06
(proportion in cohort) (.41) (.21) (.14) (.09) (.09) (.09) (.06) (.04) (.03) (.01) (.01) (.01) (.00) (.00)
Born 1920–39 (Ref.) –– – – – –––––––––
(proportion in cohort) (.44) (.31) (.26) (.31) (.27) (.30) (.32) (.25) (.26) (.25) (.20) (.18) (.10) (.05)
Born 1940–59 −.03 −.08*** −12*** −.11*** −.09*** −.11*** −.09*** −.05* −.03 −.04+ −.05* −.07*** −.12*** −.05*
(proportion in cohort) (.15) (.42) (.46) (.43) (.46) (.40) (.40) (.39) (.39) (.36) (.38) (.40) (.33) (.30)
Born 1960–79 –−.05 −.14*** −.15*** −.10*** −.13*** −.09*** −.04+ −.04+ −.04+ −.07*** −.06** −.12** −.07*
(proportion in cohort) –(.04) (.14) (.14) (.18) (.21) (.22) (.32) (.32) (.33) (.29) (.28) (.36) (.39)
Born 1980 onwards –– – – – ––––−.02 −.03 .01 −.05+ −.05*
(proportion in cohort) –– – – – ––––(.05) (.08) (.10) (.21) (.26)
Socialisation cohort analysis
Born before 1892 .09 .22+
(proportion in cohort) (.01) (.00)
Born 1892–99 .14** .13+ .03 .03 .03
(proportion in cohort) (.04) (.01) (.01) (.00) (.01)
Born 1900–07 .09** .11** .15*** .04 .15* .11 .04 .08 .01 .02
(proportion in cohort) (.10) (.05) (.03) (.01) (.01) (.01) (.00) (.01) (.00) (.00)
Born 1908–15 .06+ .09** .12*** .12** .08+ .08+ .08* −.11 .11 .28 .09 .05 .05 .02
(proportion in cohort) (.13) (.09) (.06) (.04) (.04) (.03) (.04) (.01) (.01) (.01) (.00) (.00) (.00) (.00)
Born 1916–23 .04 .08* .14*** .11*** .07* .06* .07* .04 .08+ .04 .18*** .10+ −.08 .14+
(proportion in cohort) (.16) (.11) (.10) (.08) (.08) (.09) (.07) (.06) (.03) (.03) (.02) (.02) (.01) (.00)
Born 1924–31 −.05+ .00 .06** .03 −.02 .02 .00 .06 .01 .07* .15*** .08* .03 .12**
(proportion in cohort) (.19) (.15) (.10) (.12) (.11) (.12) (.12) (.07) (.07) (.07) (.06) (.05) (.03) (.02)
Born 1932–39 (Ref.) –– – – – –––––––––
(Continued)
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 9
Table 3. Continued.
1967 1979 1984–85 1986–87 1987–88 1990
a
1993 1996 1998 2001 2004 2007 2010 2013
(proportion in cohort) (.21) (.12) (.10) (.17) (.12) (.14) (.14) (.14) (.17) (.15) (.16) (.15) (.07) (.05)
Born 1940–47 −.07* −.05 −.04+ −.05* −.06* −.06* −.06* −.01 −.02 −.00 −.02 −.02 −.10** .01
(proportion in cohort) (.15) (.14) (.15) (.17) (.17) (.13) (.15) (.13) (.13) (.13) (.12) (.14) (.11) (.10)
Born 1948–54 −.36 −.08** −.07*** −.07** −.10*** −.12*** −.08*** −.03 −.04 −.01 .02 −.04+ −.12*** −.05+
(proportion in cohort) (.00) (.16) (.18) (.15) (.17) (.15) (.15) (.15) (.14) (.13) (.15) (.16) (.11) (.12)
Born 1955–63 −.06* −.09*** −.13*** −.12*** −.09*** −.10*** −.03 −.02 −.05* −.01 −.07** −.15*** −.08**
(proportion in cohort) (.17) (.22) (.20) (.21) (.19) (.17) (.20) (.20) (.19) (.18) (.18) (.18) (.16)
Born 1964–70 −.12*** −.17*** −.09** −.10*** −.10*** −.05 −.04 −.01 .01 −.06* −.10** −.03
(proportion in cohort) (.05) (.05) (.09) (.11) (.09) (.14) (.12) (.13) (.10) (.10) (.15) (.13)
Born 1971–78 −.13** −.02 −.03 −.02 −.01 −.02 −.01 −.10** −.02
(proportion in cohort) (.03) (.05) (.10) (.10) (.10) (.10) (.09) (.11) (.13)
Born 1979–85 −.01 .01 .04 .05 −.04 −.03
(proportion in cohort) (.03) (.06) (.08) (.07) (.12) (.12)
Born 1986–95 −.07 .01 −.04 −.03
(proportion in cohort) (.01) (.04) (.11) (.15)
(n) (2054) (2016) (2898) (1511) (1638) (2371) (1771) (1433) (1530) (1633) (1417) (1554) (1747) (3225)
a
The 1990 Election panel file held by the Australian Data Archive no longer has all independent variables; the country of birth is missing from this analysis unlike Bean (1993).
+ significant at p= .10; *p= .05; **p= .01; ***p= .001. All figures have been rounded to two decimal places, with the exception of age and years of education which are to three decimal places.
Sources: Australian National Political Attitudes survey, 1967–79; National Social Sciences Survey, Integrated 1984–88 file (1993 release) & Election Panel, 1990; Australian Election Study, 1993–2013.
10 L. MANSILLO
1999 referendum. This analysis sets out to empirically test the generational difference of
public opinion based on their political socialisation. Generational responses to the
annus horribilis, referendum, the Duke and Duchess of Cambridge’s marriage and
Prince George’s birth are covered. Here I discuss some notable cohorts.
Bean’s model finds little divergence between the 1940–59 and 1960–79 cohorts. From
2001 to 2010, the generation born 1980 onwards is indistinguishable from the cohort born
1920–39. The generation born before 1920 held the most monarchist views. Furthermore,
as each cohort in Bean’s model grows older the general trend is for increasing support for
the monarchy. For example, the 1960–79 cohort has unstandardised coefficients range
from −0.15 to −0.10 in the 1980s reducing to −0.12 to −0.04 in the 2000s.
The modified cohort analysis finds childhood and adolescent political socialisation
influences opinion on the monarchy. Those born from 1916 to 1923 experienced the
year of three kings; those born 1932–39 experienced the 1953 Coronation of Queen Eliza-
beth II; those born 1955–63 were socialised when Whitlam was ousted in 1975; those born
1964–70 experienced Prince Charles’s 1981 marriage; those born 1979–85 endured royal
misbehaviour, Diana’s death and the referendum; and those born after 1986 experienced
the new royal generation, namely princes William and Harry, come of age.
The cohort born 1916–23 exhibits significantly higher support than the 1924–31
cohort. This group was of prime socialisation age when the 1936 year of three kings
rocked the Court of St James. From 1979 to 1993, there is evidence to suggest that the
cohort’s higher level of support compared to the 1924–31 cohort could be attributed to
that political experience. The 1932–39 cohort was impressionable when Queen Elizabeth
II was coronated. Compared to the generation born 1940–47 they have higher levels of
support for the monarchy. The different levels of support disappeared in 1996 reappearing
in 2010. The 1955–63 cohort has the most resilient passion against the monarchy. They
have maintained the most longstanding republican sentiments throughout their lifetime.
They were the only cohort to strengthen their anti-monarchist resolve in 2001 following
the referendum; from 2007 to 2013 the cohort exhibits the highest level of republicanism.
This is congruent with research on early experience effects on childhood and adult politi-
cal development (Sears and Brown 2013:59–60). This disposition has lasted through the
cohort’s entire lifespan. Similar to the 1955–63 cohort, those born from 1964 to 1970 were
particularly republican from 1984 to 1993 and again in 2007 and 2010. From 1984 to 1986,
this cohort exhibited stronger anti-monarchical feelings than the 1955–63 cohort. At the
time of the Whitlam dismissal, many were aged between 5 and 11. The cohort responded
in a similar way to the seven- and eight-year-old children did to Watergate (Dennis and
Webster 1975). In 2013, following the royal wedding, there was no significant difference
between the 1932–39 and 1964–70 cohorts. This cohort has not remained set by their early
childhood experiences; they have developed their views during their life course (Sapiro
1994: 204). In 2007, the 1955–63 and 1964–70 cohorts held similar levels of republicanism.
By 2010, a gap between the 1955–63 and 1964–70 cohorts had emerged, and by 2013, the
1964–70 cohort had no significant difference compared to the 1932–39 cohort –unlike the
1955–63 cohort. One cohort warmed to monarchy, while another did not. On the issue,
political socialisation after the age of 12 has more long-lasting impressions than before
the age of 12. The 1979–85 and 1986–95 cohorts exhibit no significant difference from
the 1932 to the 1939 cohort. The first was socialised with royal misbehaviour characteris-
ing the period and Diana’s death and the second the new royals coming of age.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 11
Public opinion is structured around age cohorts created through socialisation around
political events. Since 2007, the generation politically socialised during the Whitlam dis-
missal is the only generation which shows overt negative attitudes towards the monarchy.
This confirms the argument of Sears and Valentino (1997): that exogenous shocks to the
political system crystallise thought. There is evidence to support the claim, the Whitlam
dismissal and not the annus horribilis, crystallised negative thought on the monarchy.
Some cohort opinions evolved more than others during adulthood, as expected (Sapiro
1994), within the period of royal renewal. Royal misbehaviour appears not to have affected
those born after 1979 but temporarily damaged opinion (Figure 2). Those who had devel-
oped the cognitive capacity for abstract thought when Whitlam was dismissed (Sears and
Brown 2013: 60) had greater resilience to changing their opinions later in life as they were
able to recognise the monarchy ‘problem’which their enduring opinion suggests (Billig
2002:7–12).
Monarchy and nation
In nationalism studies, the dominant conceptualisation of nationalism is between civic
nationalism and ethnic nationalism (Özkirimli 2005: 22). Each is tested using 1996 and
2001 AES data. Billig (2002: 34) asserts that, in Britain, ‘the equation of monarchy with
the nation implies an attack upon monarchy is an attack upon the fundamental unique-
ness of the nation’. Some (McGregor 2006: 503) argue that Australia has a ‘composite
nationalism’fusing Britishness into an Australian identity. This is highly contested (War-
hurst 1993).
Both 1996 and 2001 data indicate a clear relationship between ethnic nationalism and
monarchism (Table 4). The relationship to civic nationalism is less clear. In 1996, before
the 1999 referendum campaign, there was a reasonably strong relationship between civic
nationalism and monarchism. By 2001, there was no significant relationship. The partial
coefficient flipped and the p-value is .069. This is approaching statistical significance with a
reduced sample, suggesting that the null hypothesis finding should not be accepted
entirely. This suggests that after the referendum, civic nationalism discontinued its posi-
tive relationship to monarchy but also may have become opposed.
Ethnic nationalism has a strong and stable relationship to monarchism and in part
explains the institution’s longevity. The level of ethnic nationalism increased in Australia
from 1996 to 2001 (Table 5), with the mean position the scale increasing over time. This
fits well with analyses of Howard’s manipulations of values and identity (Johnson 2007).
9
The 2013 AES did not have questions on nationalism. However, the 2013 Australian
Survey of Social Attitudes (AuSSA) did, and was substituted for the AES for this year
(Blunsdon 2015). On identical scales, the mean has become more civic nationalist and
less ethnic nationalist since 2001 (Appendix 1; Table 5). Since 2001, attitudes towards
the monarchy have improved as ethnic nationalism levels have declined. A triangulation
of these data suggests that the link between civic nationalism and the monarchy has prob-
ably returned. This is indicated by the stable link between ethnic nationalism and opinion,
the 11 per cent improvement in monarchy’s support from 2001 to 2013 and declining
ethnic nationalism. Civic nationalism in 1996 predicted support for the monarchy
when overall opinions towards monarchy was 6 per cent lower and overall ethnic nation-
alism levels were higher.
12 L. MANSILLO
A woman’s realm
The previous results show a continual gender divide on opinion. Billig (2002: 173) noticed
‘the differences between males and females, found on the question about “interest”, vir-
tually disappeared on [ …the] abolition of the monarchy.’Unlike Britain, in Australia,
there is a divide on abolition (Table 2). Opinion is disaggregated by gender at periods
of high and low issue saliency: before the 1999 referendum using 1998 data, and following
Prince George’s birth using 2013 data.
When disaggregated, male migrants from the British Isles exhibit greater support.
Female British migrants are not significantly different from women born in Australia
Table 5. Nationalism scale means, 1996, 2001 and 2013.
Mean (μ) 1996 2001 2013 Change (1996–2001) Change (2001–2013)
Ethnic nationalism scale .556 .588 .473 +.032 −.115
Civic nationalism scale .813 .769 .804 −.044 +.035
Difference +.263 +.181 +.331 −.082 +.150
(n) (1693) (1896) (1465) ––
Source: Australian Election Study, 1996 and 2001; Australia Survey of Social Attitudes, 2013.
Table 4. Monarchism and nationalism.
1996 2001
Model 1 Model 2 Model 1 Model 2
bβbβbβbβ
Parents’party preference (conservative) .02 .03 .01 .02 −.03 −.04 −.02 −.03
Gender (male) −.01 −.01 −.01 −.01 .02 .03 .01 .02
Age (years) .001 .07** .001 .06* .001 .07** .001 .05*
Education (years) −.009 −.07* −.007 −.05+ −.013 −.10*** −.010 −.08**
Occupation (non-manual) −.04 −.07* −.04 −.06* −.04 −.07** −.04 −.06*
Subjective social class (middle) −.06 −.09*** −.05 −.08** .06 .04+ .06 .05*
Trade union membership −.01 −.01 −.00 −.00 −.03 −.04 −.02 −.03
Religion (Ref: Catholic and other) ––––––––
Protestant .06 .09** .06 .09** .08 .13*** .08 .12***
No religion .00 .00 .01 .02 −.02 .03 −.01 −.01
Church attendance .02 .08** .02 .08** .03 .10*** .03 .09***
Urban–rural residence (urban) −.04 −.06* −.02 −.02 −.06 −.10*** −.04 −.07**
State (Ref: NSW) –– – –––––
Victoria −.03 −.04 −.02 −.03 −.01 −.01 −.01 −.02
Queensland .00 .00 .01 .01 −.01 −.01 −.02 −.01
South Australia −.01 −.01 −.02 −.02 −.05 −.04 −.04 −.03
Western Australia −.02 −.01 −.01 −.01 −.02 −.02 −.02 −.02
Tasmania .04 .02 .03 .01 .04 .02 .03 .02
Birthplace (Ref: Australia and other) –– – –– –––
British Isles .10 .10*** .15 .14*** −.10 .09*** .13 .12***
Continental Europe −.10 −.07** −.07 −.05+ −.12 −.08** −.10 −.06*
Asia −.10 −.06* −.08+ −.04+ −.06 −.06* −.05 −.05*
Political identity (conservative) .19 .30*** .19 .30*** .12 .19*** .12 .19***
Ethnic nationalism scale ––.26 .15*** ––.20 .13***
Civic nationalism scale ––.12 .05* ––−.09 −.05+
Constant .24 –−.01 –.21 –.16 –
R
2
.20 .23 .17 .18
(n) (1364) (1297) (1532) (1466)
Mean on monarchy–republic scale .33 .33 .28 .28
(standard deviation) (.31) (.31) (.30) (.30)
+ significant at p= .10; *p= .05; **p= .01; ***p= .001. All figures have been rounded to two decimal places, with the
exception of age and years of education which are to three decimal places.
Source: Australian Election Study, 1996 and 2001.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 13
(Table 6). This is true regardless of the monarchy’s issue salience. The class dynamics
differ by gender; women load on class identity while men load on occupation. This is a
product presumably of disrupted labour force status through pregnancy and child-
rearing expectation social norms.
When comparing high- to low-saliency periods, education, age and Asian migrants have
unique features. Education predicts lower levels of support for the monarchy only for
women in periods of low saliency. Age only affects higher levels of support for men at
periods of low saliency. These differences are most probably a function of different interest
levels (Billig 2002: 172–201). Only female migrants from Asia exhibit negative opinions
towards monarchy in periods of low saliency.
Discussion
The observed data indicate opinion instability from 1990 to 2013. First support fell prior to
the 1999 referendum to replace the Queen with a president elected by the parliament.
10
However, following the referendum this pattern has reversed. By 2013, support had increased
to its highest point since 1990. This recovery is not modest but a significant departure from
the late 1990s lows to a more resilient level closer to its historical trend between 1967 and
1990. This represents a crossroads of support for the Australian monarchy. Inferring causa-
tion is difficult with cross-sectional data therefore I provide context and potential sources of
opinion change. Evidence for political socialisation patterns appears to explain the bulk of
opinion change, but it is worth discussing monarchy’s relationship to other social psycho-
logical phenomena.
Both political and royal actors hold some responsibility for public opinion shifts. To
borrow Bagehot’s([1867]2009) terminology, first there is an efficient thesis and a dignified
thesis. These two narratives intertwine yet are inseparable. Focus could be given to either
as both probably helped to shift opinion. The efficient thesis includes monarchy’s dom-
estic politicisation, prolonged debate leading up to the referendum, and Howard’s
manipulation of values and identity (Gulmanelli 2014; Johnson 2007). The dignified
thesis emphasises 1990s royal misbehaviour followed by princes William and Harry
coming of age. Billig’s(1995) banal nationalism is discussed as a social psychological
explanation for support –banal royalty –and Australian nationalism and its relationship
to monarchy.
Efficient
In the 1990s, the new Labor Prime Minister, Paul Keating, put monarchy on the political
agenda. Keating had particular preoccupation with this issue and his efforts to politicise it
saw him speak on the monarchy on 17 occasions between 1992 and 1995 in the House of
Representatives. This topic has not been discussed in the chamber since. A cluster of par-
liamentary debates occurred shortly after the British tabloids featured Sarah Ferguson in a
lewd act and her public separation with the Duke of York. Keating sensationalised mon-
archy as a crosscutting issue and split voters aligned to both major parties (McAllister
2002).
11
Compared to New Zealand prolonged debate may have contributed to greater
Australian republican sentiments (Kullmann 2008).
14 L. MANSILLO
Voters preferred a republic on the condition citizens were to elect the president
(McAllister 2002). Voters sought an elected president to have their own sovereignty
to act in conflict with parliament (Wellings 2003). The added combative element
desired by voters requires the imposition of a complex system of ‘check and balances’
through a popularly elected president holding parliament to account. Such a consti-
tutional change would present major legal and electoral challenges; the latter due to
the low success rate of Australian referenda. Voters demanded more than a nip-tuck
to the monarchy but instead reform of how Australia governs itself. The referendum
lost as voters took a loss-minimisation strategy since ‘voters could not diversify, or
hedge their political risks, they are more likely to vote “No”in the face of uncertainty’
(Davidson, Fry, and Jarvis 2006: 863). Keating may have successfully contributed to
making monarchy unpopular but voters decided that Australian Westminster democ-
racy was better with a Queen than without one.
Less overt than Keating’s attempts to influence opinion, John Howard subtly encour-
aged cultural change (Gulmanelli 2014; Johnson 2007) with the use of populism (Snow
and Moffitt 2012; Wear 2008). The Howard government placed greater emphasis on
Anglo–Celtic values as part of its national narrative (Johnson 2007). Howard held
that Australia did not need to choose between history and geography. This according
to Johnson, Ahluwalia and McCarthy (2010:61–62) allows Australia to be a society
that ‘privileged both Anglo–Celtic identity and Anglo–Celtic values’while trading and
maintaining warm relations with Asia. The monarchy could have benefited from
Howard’s ambivalence to embrace Asia and the more central place for ‘traditional
values’in the state’s construction of citizen identity. Moreover, the empirical evidence
on migrant opinion change from 2001 onwards (Table 2)reflects a move away from
an explicitly multicultural nation to an assimilationist direction after the events of Sep-
tember 2001 (Tate 2009: 117).
The major shifts in majority public opinion occur after the Howard government
(Figure 2). Opinion changes appear to be exogenous to the government’s megalothymic
othering (Huynh 2009b:81–89). For example, dog-whistling on refugees has propagated
the polity since 2001 long before improvements occurred. Othering in popular culture is
widespread and facilitated by the state. An instance being the television programme
Border Force which, since 2004, has ‘alarmed and comforted’its ‘Bogan’viewers who
consume the common plot featuring ‘a suspiciously spluttering narrow-eyed foreigner
who the hunch-smart customs officers identify as having something to hide’(Huynh
2009a: 129–30). The popular consumption of an ethnocentric narrative ‘to its Bogan
audience is clear –there is a nasty world out there, but sit back and relax because gov-
ernment is in control’–all as customs officer uniforms banally display the monarchy’s
symbols (i.e., the Crown). In addition, there are increased visits by Royal Family
members at the invitation of various governments. The efficient part of state had
clear involvements in influencing of public opinion towards the monarchy.
Dignified
The dignified thesis for the change in support centres on the Australian public’s
response to the 1990s royal scandals and new Royals coming of age. Scandals include
the Queen’sannus horribilis, Prince Charles’s separation from Diana, the Princess
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 15
Royal’s divorce and the Duke of York’s estranged wife exposed on the tabloids topless
‘having her toes sucked by a “financial adviser”’ to name a few (Davies 2012). At this
point, Australia’s high support for the monarchy ebbed (Figure 2). Commercial polling
on republican support spiked from 36 per cent in 1991 to 57 per cent in 1992 (Bean
1993: 196). In Britain, Billig (2002: x) challenges whether this had any impact and
asserts that ‘it would be wrong [ …] to chart the decline in monarchy’s fortunes
purely in terms of the behaviour, or misbehaviour, of the Windsors.’Royal fortunes
fade at the moment the dignified part of state appears undignified. Republic proponents
took advantage of these events.
Research into Diana’s death around national mourning in Britain and
Australia (Abrams 1997)isreflected in this article’sfindings. It has been argued Australian
media treated Diana in death more as a media icon instead of an icon of royalty (Duruz
and Johnson 1999). A study of Australian Anglo–Celtic women found participants
had difficulty identifying with Diana in some respects such as the wealth difference
(Black and Smith 1999). Among Australian Anglo–Celtic women their ‘identification
included her physical and character attributes, the mothering role and the universal
tragedy of death.’Women related to Diana as a mother within the Royal Family. The
Durz and Johnson (1999) media ethnography suggests the media represented Diana as
a media icon, but the media reading was as royalty (Black and Smith 1999: 271–72).
Black and Smith’sfinding that age was not a barrier to relating to Diana is mirrored in
the results (Table 6). Australians were critical of the Royal Family’s treatment of Diana
and juxtaposed Diana’s‘human, emotional and loving’mothering to an ‘impassive, cold
and unemotional’Royal Family (Black and Smith 1999: 271). The focus shifted from
Prince Charles once Prince William grew up. As a father, William publically shows affec-
tion as his mother did; unlike his grandmother that Anglo–Celtic women once observed.
With Prince William there is no public point of comparison between cold and warm royal
characters.
Princes William and Harry coming of age have public attention on royalty. The princes
ventured into Iraqi and Afghan theatres and pictures appeared on Australian television
screens –along with William courting, marital pomp and the hype surrounding the
births of Prince George and Princess Charlotte. Improvement to public opinion occurred
as each generation was shown the splendour of the dignified part of state in 2011. This
undoubtedly had an effect with the highest support for the monarchy in polling data
since 1988 being registered six weeks after Prince William’s marriage (Figures 2 and 3).
There were 58 per cent who supported the monarchy while only 35 per cent supported
a republic. This is a similar level to when Prince Charles wedded in 1981. Catherine Mid-
dleton was confirmed to be dating Prince William in 2006, since then there has been
improving support for the monarchy. The spectacle of new royals appears to have main-
tained high support.
12
There are different opinions of women by education levels but not for men when the
monarchy has low saliency as an issue. Billig (2002: 177) finds a household gender division
on interest towards monarchy. It is probable that women with higher levels of education at
a low level of saliency (Table 6) actively considering the issue more.
16 L. MANSILLO
Banal royals
New royals coming of age saw a return of good fortune for the monarchy. Those who
embraced or re-embraced the monarchy probably did as a form of banal nationalism.
Billig’s(1995) account of nationalism as a social psychological phenomenon is dependent
upon symbolism in the nation-state (Day and Thompson 2004: 98). Symbols such as flags
and uniforms propagate the polity.
Monarchy has a subtle presence in Australia but there remain symbols and cultural prac-
tices legitimising the monarchy. A week does not go by without the media mentioning the
Royal Family and symbolism in the form of currency carry Queen Elizabeth II and uniforms
display St Edward’s Crown. This includes popular culture displays such as custom officers’
uniforms on Border Force that constructs a foreign other with the monarchy’s symbols
weekly for two million viewers. Symbols provide a banal reminder to the monarchy’s legiti-
macy: Banal Royalty. These symbols’ubiquity form an ‘everyday a constitution of their
awareness’(Day and Thompson 2004: 99). As the ‘flag repeatedly reminds people of
where they are, and to what they belong’currency has the same effect on the nation’s col-
lective imagining.
13
Continual reminders of monarchy’s presence self-legitimise the Crown. The legitimisa-
tion of this social order results because ‘agents apply to the objective structures of the social
Table 6. Gender and monarchism.
1998 2013
Male Female Male Female
bβbβbβbβ
Parents’party preference (conservative) .05 .07+ .05 .07+ .03 .04 .03 .04
Age (years) .001 .07+ .001 .07+ .002 .08*** .000 −.01
Education (years) −.003 −.03 −.005 −.03 −.004 .03 −.013 −.09***
Occupation (non-manual) −.05 −.09* −.01 −.02 −.06 −.09*** −.04 −.06*
Subjective social class (middle) −.02 −.04 −.07 −.11** −.01 −.02 −.03 −.05*
Trade union membership .01 .01 −.06 −.08* −.03 −.04+ .01 .01
Religion (Ref: Catholic and other) ––––––––
Protestant .08 .14** .10 .17*** .13 .18*** .12 .17***
No religion .02 .03 −.02 −.02 −.04 −.06+ −.03 −.04
Church attendance .02 .08* .02 .08* .01 .04 .01 .04
Urban–rural residence (urban) −.03 −.06 .06 −.10** −.04 .06* −.04 −.06*
State (Ref: NSW) ––––––––
Victoria −.02 −.03 .06 .08* .01 .02 .03 .04
Queensland .05 .07+ .05 .06 .02 .02 .05 .06*
South Australia .05 .05 .06 .05 .10 .09*** .04 .03
Western Australia .00 .00 .06 .05 .03 .03 −.01 −.01
Tasmania −.01 −.01 .01 .01 .00 .00 −.04 −.02
Birthplace (Ref: Australia and other) ––––––––
British Isles .10 .10** .05 .05 .08 .07** .05 .03
Continental Europe −.01 −.00 −.11 −.06 −.07 −.04+ .02 .01
Asia −.01 −.01 −.01 −.01 .04 .03 −.08 −.06*
Political identity (conservative) .11 .19*** .08 .12** .12 .19*** .13 .19***
Constant .13 −.19 −.25 −.39 −
R
2
.14 .15 .17 .15
(n) (717) (707) (1600) (1624)
Mean on monarchy–republic scale .26 .31 .36 .40
(standard deviation) (.29) (.30) (.32) (.32)
+ significant at p= .10; *p= .05; **p= .01; ***p= .001. All figures have been rounded to two decimal places, with the
exception of age and years of education which are to three decimal places.
Source: Australian Election Study, 1998 and 2013.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 17
world structures of perception and appreciation that have emerged from these objective
structures and tend therefore to see the world as self-evident’(Bourdieu 1990: 135).
One generation’s socially constructed relationship to monarchy becomes socially inherited
and the given for successive generations. Australians continue cultural practices, such as
observing the public spectacles of weddings in Westminster Cathedral, a social practice
that sets the bonds between each generation to monarchy as suggested in Table 3.
The gradual ‘modest’improvement that McAllister (2011: 29) observed after the 1999
referendum perhaps was the result of the monarchy’s symbols repairing the Crown’s legiti-
macy through the processes of banal nationalism as the republic’s salience declined. From
1998 to 2007, the AES records an improvement of 6 per cent in monarchy’s support. Then,
a 7 per cent spike from 2007 to 47 per cent in 2013 as the Australian public witnessed the
once in a generation courting of an heir and the pomp of monarchy –a sort of stimulant
‘brilliant to the eye; that which is seen vividly for a moment’(Bagehot 2009: 9). The rapid
improvement in support observed is potentially a period effect. For Billig (1995: 69)
national identity is ‘more than an inner psychological state or an individual self-definition:
it is a form of life that is daily lived’.
The use of national symbols and Royal Family media provides the basis for legitimising the
ancient institution. This is the probable source of the recent gradual improvement in public
sentiments. Monarchy is a deep-seated part of Australian national identity. In national cul-
tural events, such as royal marriages in Britain, Australians wish to keep their monarchy
more. For many Australians, monarchism is banally adopted through exposure to royal
symbols that is unconsciously associated to the nation-state.
Ethnic nationalism
Australian nationalism and the monarchy has undergone little scholarship into how
royalty features in Australian identity. This article demonstrates the relationship
between ethnic nationalism and opinion on the monarchy to be stable but conversely
civic nationalism and opinion to be unstable (Tables 4 and 5). There is significant scholar-
ship ‘[a]gainst persistent trends in Australian historiography, argu[ing] that Britishness
was neither inimical nor incidental to Australian nationalism’(McGregor 2006: 494)
and that Australian nationalism is ‘distinctively Australian while simultaneously and fer-
vently British: a composite nationalism’. McGregor (2006: 503) expands:
The Australian head of state was the British monarch, and many of the symbols of civic
nationhood –anthem, flag, honours, titles –remained wholly or substantially British long
after federation. Australians of the day were not only aware of the Britishness of their
civic institutions, they rejoiced in the fact and regarded their freedom and democracy as
part of their ethnic inheritance as Britons.
Britain and Australia no longer share an anthem, but there is an ethnic and civic
nationalism in Australia that has ‘combined in mutually supportive ways’(McGregor
2006: 503–04). The White Australia policy, which ended in 1966, and demographic popu-
lation momentum ensured this composite nationalism continued. Gulmanlli (2014: 592–
93) argues Howard reframed multiculturalism to be ‘acceptable within an Anglospherist
cultural fame work, which foresees assimilation’. This accommodates the ethnic national-
ism element McGregor describes. The finding reflects qualitative scholarship that
18 L. MANSILLO
Howard’s construction of others –‘non-Australians’against brethren ‘ordinary Austra-
lians’–tapped into ethnic nationalism implicit in his populism (Snow and Mottiff
2012: 285). Migrant assimilation following September 2001 was encouraged and key to
the construction of others (Tate 2009). The implicit promotion of semi-cultural assimila-
tion could to an extent explain the recent weakening of migrant hostilities towards the
monarchy (Table 2). Howard’s role in influencing national identity and values provides
one explanation for migrant opinion change.
The monarchy’s resilience beyond the referendum, particularly its rapid rise in support
since 2007, posits a predicament of explaining the phenomenon of improved public
opinion if banal royalty is rejected. It is plausible in Billig’s(2002: xii) thesis that ‘[c]
onstant news about royalty provides [ …] banal reminders of nationhood’operates in
Australia. Monarchism’s resurgence is all the more puzzling if the notion that the
Queen embodies some form of Australianness is rejected –as some scholars (Warhurst
1993) have –and ignore Anglo–Celtic identification with royalty (Black and Smith
1999), and the reinforcement ethnic nationalism provides to public opinion.
Conclusion
Australian public opinion on the monarchy has been fluid and not inexorably downhill as
expected in a better educated, affluent and less religious Australia. In 1999, Australians
were reluctant to abolish its monarchy with the alternative presented, but Australians
have become more loyal to the Crown. Previous expectations of continual erosion of
support for the ancient institution were folly. Monarchy has become more popular in a
country that is better educated and less religious. Support is based on political socialisation
around national events and banal nationalism which rebuilds the public’s faith in monar-
chy. When monarchy is an issue, its salience appears make it unpopular. When monarchy
is not problematised or has low salience opinion is more favourable.
Distant from the referendum, public opinion and Australia is at a crossroads. The
issue’s continuing relevance was marked by the Opposition Leader Bill Shorten’s
attempt to revive the issue on Australia Day 2015 when Tony Abbott gave Prince
Phillip a knighthood to public and backbench chagrin. Malcolm Turnbull, the former
Australian Republican Movement campaigner, replaced Tony Abbott, the former Austra-
lians for Constitutional Monarchy activist, as prime minister in September 2015. It
remains to be seen if the public opinion improvement is a temporary function due to
the arrival of new royal family members or campaigning will resume on the republic. Aus-
tralia’s renewed loyalty to the Crown makes the topic remains fertile for academic discus-
sion on public opinion and its consequences.
Notes
1. It is important to note that responses vary to a small degree on the exact question asked. Bean
(1993: 192) noted that ANPA, NSSS and AES all have comparable questions which have only
‘minor differences’.
2. The first question has minor differences with the number of response categories. Second there
are minor wording changes across the NSSS to the AES and ANPA questions. The author
agrees with Bean’s opinion (1993: 192) ‘[t]here is no reason to assume that [ …] these small
variations would have any bearing on the comparisons’. The ANPA and AES have three
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 19
response categories: ‘very important’,‘fairly important’and ‘not very important’. The NSSS has
the additional category: ‘not important at all’. The 1979 ANPA the question ‘should Australia
retain the Queen’had into two categories. In 1967 only the importance question was asked.
3. Curiously, before 1993, there were significantly more people who wished to retain the monar-
chy but did not think the Queen was important. This changed by 1993 and probably due to the
1992 royal scandals.
4. Billig (2002:2–15) wrote of the popular press preoccupation and nothingness of academic
press on the issue because sociologists and social psychologists failed to identity monarchy
itself as a ‘problem’.
5. Across the 13 surveys with combined measures, the correlations of the two variables are
between .7 and .8 and Cronbach’s alpha reliability coefficient ranges between .75 and .9.
The variables combined make a highly reliable scale.
6. The variables are coded as Bean (1993: 197–98) described his variables.
7. Due to generational replacement the reference category in the reproduction of Bean’s cohort
analysis was altered.
8. Following the 1992 annus horribilis, the effect for partisanship is comparatively smaller.
9. The finding is reflected in Jackman’s(1998: 174) analysis where measures of ethnocentric racial
attitudes strongly correlated with support for monarchy.
10. In 1999, a recorded low of 25 per cent of Australians were monarchists if the Australian Consti-
tutional Referendum Study is used, however, respondents were primed by the referendum (McAll-
ister 2002: 256).
11. One occasion typifies this sort of activity: in November 1994 Keating detailed royal visit
expenses with the aid of a Dorothy Dixer (Hansard 1994).
12. The last data point in Figures 2 and 3is a Newspoll from June 2014, which was 11 weeks after
Tony Abbott reintroduced Order of Australia Knights and Dames. The ‘dignified’part of the
state made inroads in opinion, but the ‘efficient’part could have marred opinion.
13. The Union Flag affixed in Australian flag’s corner would impact collective imagining of
Australia.
14. Principal component analysis with Varimax rotation used. First three eigenvalues for 1996,
2001 and 2013, respectively, are 2.45, 1.19 and .85; 2.74, 1.26 and .77 and 2.89, 1.21, and .79.
15. Percentages do not always sum exactly to subtotals because of rounding.
16. Dependent variable scaled from 0 (most pro-republic) to 1 (most pro-monarchy).
17. + significant at p≤.10; *p≤.05; ** p≤.01; ***p≤.001. All figures have been rounded to two
decimal places, with the exception of age and years of education which are to three decimal
places.
18. The means and standard deviations of the dependent variables at each time point show con-
sistency. There are minor measurement differences between the ANPA, NSSS and AES,
meaning that the means are not a good guide of the level of public support for the monarchy.
19. The 1990 Election panel file held by the Australian Data Archive no longer has all independent
variables; the country of birth is missing from this analysis unlike Bean (1993).
Acknowledgements
The author is indebted to many who he thanks but especially Rodney Smith, Steven McEachern,
Edith Gray, David Smith, Clive Bean, Charles Pattie, Adele Lausberg, Shaun Ratcliff and Marian
Simms for their comments, patience and assistance through the article’s different stages of devel-
opment. A previous version of this article was presented at the 2015 Political Studies Association
of the United Kingdom annual meeting.
Disclosure statement
No potential conflict of interest was reported by the author.
20 L. MANSILLO
Notes on contributor
Luke Mansillo is a PhD Candidate in the Department of Government and International
Relations at the University of Sydney. He has come to the University of Sydney from the
Australian National University, Canberra, where he completed a BA (Hons.) in political
science and Masters of Social Research at the School of Politics and International Relations
and Australian Demographic and Social Research Institute (ADSRI) respectively. He is the
2015–2016 Australian Political Studies Association Postgraduate Representative.
ORCID
Luke Mansillo http://orcid.org/0000-0002-4366-3993
References
Abrams, D. 1997. Uncertainty and social psychological responses. Psychologist 10: 500–01.
Aitkin, D. 2004. Macquarie University Australian Political Attitudes Survey 1979 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.edu.au/
ada/00009>. Consulted 2 December 2014.
Aitkin, D., Kahan, M. and Stokes, D. 2011. Australian National Political Attitudes 1967 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.
au/ada/01059>. Consulted 11 September 2014.
Blunsdon, B. 2015.Australian Survey of Social Attitudes 2013. Canberra: Australian Data Archive,
The Australian National University. URL: <https://www.ada.edu.au/social-science/01272>.
Consulted 11 December 2015.
Bagehot, W. [1867] 2009.The english constitution. Reprint. Oxford: Oxford University Press.
Bean, C. 1993. Public attitudes on the monarchy–republic issue. Australian Journal of Political
Science 28: 190–206.
Bean, C., Gow, D. and McAllister, I. 1998. Australian Election Study 1998 [data file]. Canberra:
Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.au/social-
science/01001>. Consulted 11 September 2014.
Bean, C., Gow, D. and McAllister, I. 2001. Australian Election Study 2001 [data file]. Canberra:
Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.au/social-
science/01048>. Consulted 11 September 2014.
Bean, C., McAllister, I., Gibson, R. and Gow, D. 2004. Australian Election Study 2004 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.
au/social-science/01079>. Consulted 11 September 2014.
Bean, C., McAllister, I., Gow, D. and Gibson, R. 2007. Australian Election Study 2007 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.
au/social-science/01120>. Consulted 11 September 2014.
Billig, M. [1992] 2002.Talking about the royal family. Reprint. London: Routledge.
Billig, M. 1995.Banal nationalism. London: Sage.
Black, E. and Smith, P. 1999. Princess Diana’s meanings for women: Results of a focus group study.
Journal of Sociology 35: 263–78.
Bourdieu, P. 1990.In other words. Cambridge: Polity.
Commonwealth Parliament of Australia [Hansard]. 1994.House of Representatives Parliamentary
Debates, official Hansard, 17 November, 3831.
Davidson, S., Fry, T.R.L. and Jarvis, K. 2006. Direct democracy in Australia: Voter behaviour in the
choice between constitutional monarchy and a republic. European Journal of Political Economy
22: 862–73.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 21
Davies, C. 2012. How the royal family bounced back from its ‘annus horribilis’.The Guardian 24
May. URL: <www.theguardian.com/uk/2012/may/24/royal-family-bounced-back-annus-
horribilis>. Consulted 4 November 2014.
Day, G. and Thompson, A. 2004.Theorizing nationalism. Basingstoke: Palgrave Macmillan.
Dennis, J. and Webster, C. 1975. Children’s images of the president and of government in 1962 and
1974. American Politics Research 3: 386–405.
Duruz, J. and Johnson, C. 1999. Mourning at a distance: Australians and the death of a British prin-
cess. In A. Kear and D. Stienburg, eds. Mourning Diana: Nation, culture and the performance of
grief. London: Routledge.
Evans, J. 1980. Women and politics: A re-appraisal. Political Studies 28: 210–21.
Goot, M. 1994. Contingent inevitability: Reflections on the prognosis for Republicanism. In G.
Winterton, ed. We, the people. St Leonards, NSW: Allen & Unwin.
Gulmanelli, S. 2014. John Howard and the ‘Anglospherist’reshaping of Australia. Australian
Journal of Political Science 49: 581–95.
Huynh, K. 2009a. If there is hope: It lies with the Bogans. In J. George and K. Huynh, eds. The
culture wars: Australian and American politics in the 21st century. Melbourne: Palgrave
Macmillan.
Huynh, K. 2009b. Us and them: National identity and the questions of ‘belonging’. In J. George and
K. Huynh, eds. The culture wars: Australian and American politics in the 21st century.
Melbourne: Palgrave Macmillan.
Jackman, S. 1998. Pauline Hanson, the mainstream and political elites: The place of race in
Australian political ideology. Australian Journal of Political Science 33: 167–86.
Jennings, M.K. and Zhang, N. 2005. Generations, political status and collective memories in the
Chinese countryside. The Journal of Politics 67: 1164–89.
Johnson, C. 2007. John Howard’s‘values’and Australian identity. Australian Journal of Political
Science 42: 195–09.
Johnson, C., Ahluwalia, P. and McCarthy, G. 2010. Australia’s ambivalent re-imagining of Asia.
Australian Journal of Political Science 45: 59–74.
Jones, R., McAllister, I., Denemark, D. and Gow, D. 1993. Australian Election Study 1993 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.
au/social-science/00763>. Consulted 11 September 2014.
Jones, R., McAllister, I. and Gow, D. 1996. Australian Election Study 1996 [data file]. Canberra:
Australian Data Archives, Australian National University. URL: <http://ada.anu.edu.au/social-
science/00943>. Consulted 11 September 2014.
Kelley, J., Bean, C. and Evans, M. 2004a. National Social Science Survey Integrated Data 1984–1988
[data file]. Canberra: Australian Data Archive, Australian National University. URL: <http://ada.
edu.au/ada/00594>. Consulted 27 November 2014.
Kelley, J., Bean, C. and Evans, M. 2004b. National Social Science Survey 1990 Election Panel [data
file]. Canberra: Australian Data Archive, Australian National University. URL: <http://ada.edu.
au/ada/00630>. Consulted 27 November 2014.
Kullmann, C. 2008. Attitudes towards the monarchy in Australian and New Zealand compared.
Commonwealth & Comparative Politics 46: 442–63.
McAllister, I. 1992.Political behaviour: Citizens, parties and elites in Australia. Melbourne:
Longman Cheshire.
McAllister, I. 2002. Elections without cues: The 1999 Australian Republic referendum. Australian
Journal of Political Science 36: 247–69.
McAllister, I. 2011.The Australian voter. Sydney: UNSW Press.
McAllister, I., Bean, C., Gibson, R. and Pietsch, J. 2010. Australian Election Study 2010, v. 2 [data
file]. Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.
edu.au/social-science/01228>. Consulted 11 September 2014.
McAllister, I., Bean, C., Pietsch, J. and Gibson, R. 2013. Australian Election Study 2013 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.
au/social-science/01259>. Consulted 11 September 2014.
22 L. MANSILLO
McAllister, I., Jones, R., Papadakis, E. and Gow, D. 1990. Australian Election Study 1990 [data file].
Canberra: Australian Data Archive, Australian National University. URL: <http://ada.anu.edu.
au/social-science/00570>. Consulted 11 September 2014.
McAllister, I. and Mughan, A. 1987. Australian Election Study 1987 [data file]. Canberra: Australian
Data Archive, Australian National University. URL: <http://www.ada.edu.au/social-science/
00445>. Consulted 11 September 2014.
McGregor, R. 2006. The necessity of Britishness: Ethno-cultural roots of Australian nationalism.
Nations and Nationalism 12: 493–511.
Özkirimli, U. 2005.Contemporary debates on nationalism: A critical engagement. Basingstoke:
Palgrave Macmillan.
Sapiro, V. 1994. Political socialization during adulthood: Clarifying the political time of our lives.
Research in Micropolitics 4: 197–223.
Schuman, H., Akiyama, H. and Knäuper, B. 1998. Collective memories of Germans and Japanese
about the past half-century. Memory 6: 427–54.
Schuman, H. and Rogers, W. 2004. Cohorts, chronology and collective memories. Public Opinion
Quarterly 68: 217–54.
Sears, D.O. and Brown, C. 2013. Childhood and adult political development. In L. Huddy, D.O.
Sears and J.S. Levy, eds. The Oxford handbook of political psychology. 2nd edn. Oxford:
Oxford University Press.
Sears, D.O. and Valentino, N.A. 1997. Politics matters: Political events as catalysts for preadult
socialization. The American Political Science Review 91: 45–65.
Snow, D. and Moffitt, B. 2012. Straddling the divide: Mainstream populism and conservatism in
Howard’s Australia and Harper’s Canada. Commonwealth & Comparative Politics 50: 271–92.
Tate, J. 2009. John Howard’s“nation”: Multiculturalism, citizenship, and identity. Australian
Journal of Politics and History 55: 97–120.
Torney-Purta, J. 2005. Adolescents’political socialization in changing contexts. Political Psychology
25: 465–78.
Warhurst, J. 1993. Nationalism and republicanism in Australia: The evolution of institutions, citi-
zenship and symbols. Australian Journal of Political Science 28: 100–120.
Wear, R. 2008. Permanent populism: The Howard government 1996–2007. Australian Journal of
Political Science 43: 617–34.
Wellings, B. 2003. Britishness and the failure of Australian Republicanism. Studies in Ethnicity and
Nationalism 3: 35–50.
Appendix 1
Table A1. Factor loadings for nationalism scale.
14
1996 2001 2013
Ethnic Civic Ethnic Civic Ethnic Civic
To be Truly Australian: born in Australia .845 −.024 .849 −.014 .865 .074
To be Truly Australian: live in Australia most of life .787 .160 .816 .155 .871 .080
To be Truly Australian: be Christian .601 .156 .608 .128 .561 .253
To be Truly Australian: speak English .423 .492 .498 .517 .466 .540
To be Truly Australian: Australia citizenship .336 .550 .411 .557 .417 .597
To be Truly Australian: feel Australian .089 .686 .063 .759 .325 .578
To be Truly Australian: respect Australian laws −.066 .762 −.006 .800 −.222 .826
Cronbach’s alpha .678 .728 .750
(n) (1693) (1896) (1465)
Source: Australian Election Study, 1996 and 2001; Australia Survey of Social Attitudes, 2013.
AUSTRALIAN JOURNAL OF POLITICAL SCIENCE 23
