The power of belief: religious freedom in Australian parliamentary debates on same-sex marriage

ABSTRACT

More than merely a theme of the Australian parliamentary debates on the bill to legalise same-sex marriage, ‘religious freedom’ appeared in the bill’s very title. This paper explores why and how this happened using a corpus-assisted analysis of the 663 parliamentary speeches made during the marriage legislation debates from 2004 to 2017. The analysis demonstrates that by 2017, the idea that marriage equality was a profound threat to religious freedom was well entrenched in the parliamentary discourse. The study finds that the potential offence of religious sensibilities came to be regarded by politicians as more significant than ongoing discrimination, thereby granting tremendous social power to religious institutions to practise discrimination in the face of changing values in society.

澳大利亚关于同性婚姻的法案既然在标题中写明了“宗教自由”的字样，国会所辩论的就不止于婚配这件事了。作者通过对2004到2017年婚姻立法辩论期间的663次国会发言做的语库辅助分析，探讨了这个问题。根据我们的分析，婚姻平等威胁宗教自由的观点到2017年都盘踞在国会的话语中。笔者发现，宗教情感的潜在伤害被政治家视为比现行的歧视更具意义，这样就赋予宗教机构以巨大的社会权力，在变化的社会价值观面前实施歧视。

KEYWORDS:

• Religion
• Australian politics
• sexuality
• parliamentary speeches
• text analysis

Acknowledgements

The author is grateful for the constant support and excellent feedback of Marion Maddox from Macquarie University. Clare Monagle also provided wise suggestions at critical points along the way and staff and doctoral researchers from the Department of Modern History, Politics and International Relations responded with helpful feedback to an early work-in-progress presentation. The author also acknowledges the work of the anonymous reviewers whose careful engagement and attention has made this a better paper than it otherwise would have been.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Elenie Poulos is a doctoral researcher in the Department of Modern History, Politics and International Relations at Macquarie University. She is a minister in the Uniting Church in Australia, and was, for 15 years, the National Director of UnitingJustice Australia, the Church’s national justice policy and advocacy unit.

ORCID

Elenie Poulos http://orcid.org/0000-0001-6804-8068

Notes

1 See https://www.dss.gov.au/our-responsibilities/families-and-children/programs-services/recognition-of-same-sex-relationships, accessed 23 August 2019.

2 At the time of writing, in response to religious groups and the right wing of the Liberal-National coalition, the conservative Morrison government was preparing a draft Religious Discrimination Bill for public consultation, to be modelled on existing anti-discrimination laws.

3 McKeown counts 23 bills after 2004, including two bills’ multiple appearances on the Notice Paper. Multiple appearances of individual bills have been consolidated for the purposes of this paper, giving a total of 20 unique bills after the 2004 amendment, and a total of 22 bills including the two government bills in 2004.

4 PDF to Text, available at http://pdftotext.com

5 Collocates were identified using a setting of 4 words to the left and right with a minimum frequency of 3 occurrences across the sub-corpus.

6 MI is an effect size measure representing the collocational strength of the relationship between two words (Baker, Gabrielatos, and McEnery 2013). The T-score is a significance measure that takes into account the size of a corpus and reflects the statistical level of confidence in an association between the two words. Potential collocates with an MI score less than 3 or a T-score less than 2 were excluded. See ‘A guide to statistics: t-score and mutual information’, Collins Wordbanks Online, https://wordbanks.harpercollins.co.uk/other_doc/statistics.html, accessed 28 March 2018.

7 Two measures were used to identify keywords. Log-likelihood establishes any significant difference in word frequency between the two corpora under comparison – the high threshold of p < 0.000001 was set to give 99.9999% confidence that an identified keyword did not appear on the list by chance (also, keywords had to appear in at least 10% of the texts). The ‘log ratio’ statistic provided an ‘effect size’ measure indicating the size of the difference in frequency in the corpora being compared (Hardie 2014).

8 No lexical words collocated with ‘freedom’ in a statistically significant way in the 2004 speeches. ‘Freedom’ appeared only six times and ‘freedoms’ three times in the 2004 speeches. The collocation analysis for ‘freedom’ was run on the exact word ‘freedom’ because the word-form includes ‘freedoms’ which appeared 99 times in the 2017 speeches within the name of the bill and would have skewed the results. The appearance of ‘freedom/s’ as a collocate for ‘freedom’ in 2017 indicates use of the word more than once within an 9-word span (4 words to the left and right) as in ‘the right of freedom of conscience and freedom of speech’.

9 ‘Equality’ appears 880 times in the supportive speeches of 2006–16, including 36 times within the bill names. The cluster ‘marriage equality’ appears 608 times, excluding its use in bill names. The cluster ‘same-sex marriage’, by comparison, appears 213 times.