The weaponisation of free speech under the Morrison government

ABSTRACT

In recent years, the concept of ‘free speech’ appears to have reached a position of significant prominence in public debate. In this context, we ask how the concept of free speech was used discursively by the Morrison-led government. Utilising a political discourse analytic approach, and examining texts from 2019 to 2021, we argue that, and show how, the government engaged in multiple discourses around free speech that were incompatible, inconsistent and incoherent. These inconsistencies and discontinuities were so significant as to render this a ‘weaponisation’ of free speech discourse. Thus, the article both reveals the complexities of the Morrison government’s discursive positioning of free speech, and renders sensible the otherwise nonsensical way in which ‘free speech’ was used to garner support for a range of public policies.

近年来，“自由发言”似乎在公共辩论中到达了一个很高的位置。我们要问，莫里森政府是如何使用自由发言论这一概念的。本文采用政治话语分析方法，研究了2019到2021年的文档，指出并说明政府参与有关自由发言的多重讨论既不系统、又不统一、也不连贯。这种支离破碎是造成自由发言 武器化的重要原因。本文指出了莫里森政府对“自由发言”混乱定位有其复杂性。用该说法收获对一系列公共政策的支持，只是用得没什么道理，本文则把它整理得有道理一些。

KEYWORDS:

• Free speech
• weaponisation
• public debate
• political discourse analysis
• Australia

1. Introduction

In recent years, ‘free speech’ has featured prominently in political debate in many countries including the UK (Smith 2020) and the United States (Liptak 2018). While politicians routinely use the concept of free speech to justify their policy positions, it also appears that the centrality of free speech in public debate has reached new heights in Australia. In this context, it is important to examine whether this is the case, and if it is, how the concept of free speech is being used discursively. What do government politicians mean when they invoke the concept of free speech?

We take a political discourse analytic approach to answering this question because it allows us to explore the ways in which the government sought to legitimise, and build support for, policy. After justifying our framework, we provide the context within which to understand current debates, including evidence that the topic of free speech has increased significantly as a component of public debate in the last five years and the key issues in that debate. We then engage in a political discourse analysis of the Prime Minister’s and Ministers’ press releases and media commentary on this topic during the term of Prime Minister Scott Morrison.

We find that the government discursively constructed free speech in ways that demonstrated significant inconsistencies and discontinuities. We find that the scope of those inconsistencies and discontinuities was so large as to render the discourse as a whole incoherent. While it is not unusual for discourses to compete with one another in public debate, in this case the discourse was incoherent to a degree that rendered it impossible for the public to understand what was meant when the government said it was concerned with protecting free speech. In this context, we describe this behaviour as having amounted to a ‘weaponisation’ of free speech by the Morrison government, and we connect weaponisation to the political discourse analysis literature, thus augmenting understandings of dissimulation as an art of political discourse. This also renders sensible the otherwise nonsensical use of free speech by the government.

2. Discourse analysis

Analysing this discourse raises the possibility of applying a wide range of theoretical perspectives and methodological approaches (Dunn and Neumann 2016, 9). There is an enormous literature on Discourse Analysis (see, eg, Fairclough 1992; Renkema 2004), an interdisciplinary approach drawing from fields as diverse as linguistics, pragmatics, philosophy, and political science that seeks to analyse and understand the functions of discourse (Renkema 2004, 2). It seeks to explain how ‘specific systems of meaning-production have been generated, circulated, internalized and/or resisted’ (Dunn and Neumann 2016, 4) through language use. Critical Discourse Analysis is similarly interdisciplinary, and incorporates a diversity of approaches including the socio-cognitive associated with van Dijk and the ‘dialectical-relational’ associated with Fairclough (Wodak 2009, 311). What proponents of CDA distinctively share as an intrinsic component of their approach is a commitment to examining and addressing social and political injustice, mediated through the power of discourse in specific contexts (Titscher et al. 2003, 146). This is a commitment to examining how societies operate to produce insecurity and inequality, or might be changed to sustain wellbeing (Van Dijk 1997a, 22; Fairclough 2003, 202). CDA seeks to make visible the ‘reciprocal influences of language and social structure’ in which language is both constituted by, and constitutes, social practice (Titscher et al. 2003, 147).

In part as a result of its embrace of a diversity of disciplinary perspectives, a wide range of methods, and a plurality of defining features, Discourse Analysis and Critical Discourse Analysis have been critiqued for lacking a systemic, objective approach to data collection and analysis, and permitting those engaged in this work to pick from an eclectic mix of principles and procedures (Widdowson 2004; Titscher et al. 2003, 163–164; Blommaert 2005). Adherents, however, defend their perspectives and approaches (e.g. Wodak 2009, 313–315).

While there is no room to parse out these debates here, our approach is informed by a political discourse analytic approach (Chilton and Schäffner 2002, 199; Van Dijk 1997b), which emphasises the importance of studying political discourse as the use of persuasive language ‘of which the public is only half aware’ (Chilton and Schäffner 1997, 207). In our study, the speakers are formally positioned as authoritative by virtue of being a ‘discourse community’ (Chilton and Schäffner 2002, 21) of government MPs who engage in a ‘constitutive part of the political process’ in publicly debating this topic (Van Dijk 1997b, 21). In such a context, one of the ‘strategic functions’ that discourse can achieve is dissimulation, defined in the Oxford English Dictionary as concealing, feigning and hypocrisy. Dissimulation can be achieved through the dissemination of inadequate, inaccurate or confusing information (Chilton and Schäffner 1997, 212–213). A political discourse analytic approach is particularly germane to our question, because it allows us to focus on the language use of government MPs and ask how they have constructed the concept of free speech in relation to their policy choices, including whether and the extent to which dissimulation occurs.

We analyse and identify ‘discourses’, defined as social practices that seek to construct a particular meaning and reality (Titscher et al. 2003, 26-27; Van Dijk 1997a, 14–15). They are made up of ‘texts’; the analytical units which reveal discrete manifestations of the discourses we seek to identify and the ‘evidence to be described empirically’ to uncover their features (Titscher et al. 2003, 26). Texts are examined through dominant ‘representations’ (Dunn and Neumann 2016, 5), a concept common to Discourse Analysis and Critical Discourse Analysis. Representations are ‘things as they appear to us’; understandings of social phenomena (Dunn and Neumann 2016, 33, 37, 60). The ‘span’ of a representation constitutes a ‘discourse’ that establishes ‘preconditions and parameters for [political] action’ (Dunn and Neumann 2016, 34, 61). Identifying discourses therefore ‘entails showing the affinities and differences between representations in order to demonstrate whether they belong to the same discourse’ (Dunn and Neumann 2016, 7). Below, we examine political debate around free speech by government MPs, with a view to uncovering instability and discontinuity (Dunn and Neumann 2016, 104) in how appropriate parameters for free speech are discursively represented. This is because underlying our question of how government MPs have constructed the meaning of free speech in the study period is a hypothesis that this has become sufficiently unstable so as to lack coherence. Thus, we also posit discourses of free speech as being ‘weaponised’.

The idea that some discourse has been ‘weaponised’¹ is widespread, although the concept of weaponisation is loose and arises within a wide variety of theoretical perspectives and approaches. For example, it has been used in rhetorical studies to identify strategic language use as an ‘instrumental tool and aggressive means to gain compliance and avoid accountability’ by demagogues (Mercieca 2019, 266). It has been used in a Bourdieuian analysis to identify rapidly laying the ‘essential cornerstones of material violence’ to underpin authoritarian politics (Pascale 2019). It has been described, absent a theoretical frame, as composed of trolling, and the deliberate use of deception, fraud and harassment to ‘attack, harass, and silence’ others (Wu 2018).

In the explicitly political sphere of public policy disputes, weaponisation has been used to describe when the language of ‘fake news’ has been used by parliamentarians to ‘deligitimize negative reporting’ about them (Farhall et al. 2019, 4354, 4364; Farkas and Schou 2018); when scientific language was mobilised to reduce support for gender-affirming procedures for transgender youth (Ashley 2020); and when far-right groups used feminist discourses to legitimise Islamophobia (Cockbain and Tufail 2020). It has been used to describe when ‘woke’ came to mean a negation of critiques of the far-right, and ‘free speech’ came to denote those opposed to discrimination as themselves discriminating against their opponents (Cammaerts 2022, 735, 738–740). In these analyses, words have been mobilised in a bid to win public and parliamentary support for policy by using terms in a way that is inconsistent with the available evidence, and inconsistent with their prior usage. One of the contributions of this paper is to incorporate this latter idea of ‘weaponisation’ into political discourse analysis to assist in revealing a particular type of dissimulation by authoritative government speakers.

3. Data collection

As evidenced below, there has been a significant increase in public debate around free speech in Australia since 2017. We therefore choose to focus our analysis on the period from August 2018 when Scott Morrison became Prime Minister to the end of December 2021.

In this period, we searched all Prime Ministerial media releases for those that mentioned ‘free speech’ or ‘freedom of speech’ (n = 27). Where we identified relevant media releases, we triangulated from them to media reports and other Ministers’ media releases on the same policy topics in the same time frame. This enabled us to collate a comprehensive set of texts relevant to the government’s use of the idea of free speech.

We analysed the texts with a view to uncovering how Morrison and his ministers were constructing the meaning and parameters of free speech. Specifically, we were concerned to uncover evidence of dissimulation through representations; to discern affinities and differences, inconsistencies and discontinuities; and what the discourse was. This included considering whether there was one free speech discourse or more than one at play.

4. The context for free speech in Australia

Before proceeding to analyse the data, a political discourse-analytic approach requires elaboration of the context. We focus on the most salient aspects first, before outlining the specificities of our period of study.

Free speech has had a strong, but mediated, place in Australian political legal and culture. Unusually among liberal democracies, Australia does not possess a federal bill of rights and free speech is not explicitly protected in federal constitutional or statutory law. Rather, Australia has a complex, multilayered protection for free speech derived in part from a common law tradition, and augmented since 1992 by a doctrine developed by the High Court of Australia of an implied constitutional freedom of political communication (McNamara 2007, 231–232; National Human Rights Consultation 2009, 120–122).

While research has shown that Australians value free speech (Human Rights and Equal Opportunity Commission 2005; Wilson et al. 2005, 86), Australia’s historical approach to free speech has been premised on utilitarianism (Kildea 2003, 65), which has both granted leeway to governments to restrict free speech in overbroad ways (Australian Human Rights Commission 2022, 55) and allowed the engendering of public support for limitations on free speech that support a public good. The paradigmatic example is vilification laws, which are understood to protect vulnerable target communities from harm.

In this context, free speech has been a consistently held value, relatively free from the vicissitudes of public debate. However, intuitively it seems that in recent years this has changed. To investigate whether or not this is the case, we searched all capital newspapers² around Australia, and The Australian, for news articles³ containing the term ‘free speech’ or ‘freedom of speech’ in the period 2001–2021. We then calculated those articles as a proportion of the total number of news articles published in those outlets in the same period. Figure 1 shows the results. They indicate a significant increase in the prominence of the topic of free speech since 2011, especially from 2017 onwards.

Figure 1. Proportion of news articles concerned with ‘free speech’ 2001–2021.

In 2017 the federal government requested the Australian Bureau of Statistics to initiate a postal survey asking Australian voters if they supported the right of same-sex couples to marry (ABS 2017), and the survey was conducted in September-November. Research has shown that freedom of speech was a ‘dominant discussion topic’ on the Facebook pages of those who opposed marriage equality (Chen 2016, 293) and that those advocating a ‘no’ vote came to identify their position with freedom of speech (Poulos 2020a, 1; 2020b). It is notable, however, that in the period since 2017 the prominence of free speech has remained consistently higher than was previously the case, which justifies our choice of time period.

In the years 2017-2021, our analysis showed that five issues raised the prominence of debate around free speech in Australia. These are all concerned with policy implementation and change, rendering them appropriate topics on which to conduct political discourse analysis.

The first was the 2018 Federal Government inquiry into free speech in universities. Education Minister Dan Tehan announced a review into ‘the rules and regulations protecting freedom of speech on university campuses’ (Tehan 2018), purportedly in response to ‘the influence of left-wing activists’ on campus (Koziol 2018), and protests against controversial speakers like Bettina Arndt and Quentin Van Meter (Dodd 2018). The inquiry’s report noted that the available evidence did not substantiate suggestions of a ‘freedom of speech crisis’ (French and Robert 2019, 13, 18) but still provided a model voluntary code on free speech. In 2020 the Federal Government commissioned a review of universities’ compliance with the adoption of the model code, and in 2021 it legislated⁴ to define academic freedom in the Higher Education Support Act 2003.

The second was Australian Federal Police (AFP) raids on journalists. On 4 and 5 June 2019, the AFP raided journalist Annika Smethurst’s home and car, and the ABC’s Sydney offices. The raids related to Daily Telegraph reports about plans ‘to grant domestic surveillance powers to the Australian Signals Directorate (ASD)’, and stories by ABC News about ‘alleged war crimes [committed] by Australian Special Forces in Afghanistan’ (Ananian-Welsh 2019), known as ‘The Afghan Files’ (Oakes and Clark 2017). The AFP accessed and copied Smethurst’s phone data⁵ and documents from the ABC’s computers (Lyons 2019). The High Court of Australia later ruled that the warrant used by the AFP to carry out the raid on Annika Smethurst’s home was invalid,⁶ but that the AFP was able to keep the data it had obtained.⁷ No prosecution was pursued (Galloway 2020; McCartney 2020). The board of the National Press Club of Australia described the raids as indicative of ‘a worrying attitude on the part of law enforcement, security agencies and the federal government … contrary to the public’s right to know’ (National Press Club 2019), while Digital Rights Watch called the raid on Smethurst’s home ‘a gross abuse of national security powers’ (Singleton Norton 2019).

The third was religious freedom. The Morrison-led government made several attempts to introduce a religious discrimination bill. This commenced in October 2019 with the release of first exposure drafts of religious discrimination bills,⁸ proposed amendments to those drafts in December 2019,⁹ and then second exposure drafts in January 2020.¹⁰ On 25 November 2021, PM Morrison introduced a package of amended bills.¹¹

The fourth was a policy initiative claimed to prevent online trolling. On 28 November 2021, PM Morrison announced plans to introduce legislation that would require ‘global social media giants to unmask anonymous online trolls’, to ensure that victims can ‘identify and commence defamation proceedings’ (Morrison and Cash 2021). Communications Minister Paul Fletcher MP introduced the Social Media (Anti-Trolling) Bill 2022 to parliament in February 2022.¹² The Bill would have required social media companies to reveal the identity and contact details of social media users who make defamatory comments online.¹³ However, it was widely criticised as inhibiting freedom of speech in significant ways. Commentators criticised it as likely to inhibit important political dissent (Payne 2022), and ‘fuel the current trend of government MPs suing their social media critics’ (Klein 2021). The Attorney-General’s office clarified that the bill is ‘really about “only defamation”’ and not ‘intended to address broader issues of harm and abuse online’ (Butler 2022). Thus, the Bill was unlikely to help those who suffer the effects of online trolling (Beckett 2021) as ‘most online abuse isn’t defamatory’ but simply offensive or upsetting (Gregoire 2022). Media experts suggested that the Bill could threaten vulnerable persons, such as ‘corporate whistleblowers, political activists, people fleeing domestic violence, or people in jobs with strict social media policies’ (Butler 2021) and encourage social media companies to increase the amount of personal data they collect (Beckett 2021). The Bill lapsed in April 2022.

The fifth was anti-vaccination discourse during the COVID-19 pandemic. In January 2022, Immigration Minister Alex Hawke deported Serbian tennis player, Novak Djokovic, ‘on health and good order grounds’, stating that ‘it was in the public interest to do so’ (Karp 2022). Minister Hawke argued that if Djokovic was allowed to stay in Australia, this may increase anti-vaccination attitudes among Australian residents and encourage further demonstrations against vaccination mandates and COVID-19 restrictions,¹⁴ thus locating the decision within a public debate about anti-vaccination discourse.

5. Empirical data: key representations

In this section we describe empirically the evidence we collected to uncover features of the discourses (Titscher et al. 2003, 26), providing extensive quotes to allow insights into the constructions we examined (Sengul 2023, 203). We found five key representations, which necessarily overlap. They are: (1) that free speech should be more strongly protected even if it makes people feel uncomfortable; (2) that free speech should be more strongly protected even if it is harmful; (3) that free speech should be limited if it makes people uncomfortable or is harmful; (4) that existing laws that limit free speech should be respected and upheld; and (5) that free speech should not apply equally to everybody.

The first representation was that free speech needs to be more strongly protected, even if it makes people feel uncomfortable. PM Morrison suggested that ‘there’s free speech for some and not for others’ at universities, and that ‘if you have a different view [to the Left] then apparently you’re a bigot’ (Morrison 2018). Tehan suggested that ‘a university education should involve dealing with ideas and concepts that are challenging’ (Brook 2018), and that universities ought to ‘protect “all free speech – even where what is being said may be unpopular or challenging”’ (Merritt 2018). Senator James Paterson called for universities to have funding withheld if they failed to uphold free speech, and accused some of being ‘hell-bent on enforcing [their] ideological hegemony’ (Paterson 2018). The November 2021 Religious Freedom Bill was also promoted as protecting free speech; in this case for people of faith. The PM claimed it was designed to ‘protect people of religion, or faith’, as well as ‘those who choose not to have a faith or religion’, from discrimination,¹⁵ thus positing those who wanted to express religious views as having insufficient protection for their freedom of speech.

The second representation was that free speech needed to be more strongly protected, even if the speech harms and even if that means overriding existing laws. The Religious Freedom Bill included a provision which would protect persons from potential discrimination or vilification claims under federal, state, or territory laws if they expressed a genuinely held statement of belief,¹⁶ thus seeking to override existing legal limits¹⁷ on freedom of speech that restrict vilification in order to protect vulnerable targets. Morrison stated that people should have an untrammelled right to religious expression:

What I’m keen to ensure is that for your religious views in Australia, that you can get about them peacefully, that you shouldn’t feel intimidated about them. I think what we saw in the [general] election was … people having a sense of discomfort even just expressing their views when it comes to their religion. (Morrison 2019a)

He also capaciously posited freedom of speech as encompassing not being made even to feel uncomfortable.

Later that year, Morrison referenced this purpose of the Bill as ‘an expression of liberty … which we all hold dear in countries such as Australia’ (Morrison 2019b), thereby placing it on a plane with liberty generally and unassailably. In February 2022, Morrison reaffirmed that the Bill was designed to ensure ‘that people can speak their beliefs in this country … which you can’t do in some other countries … and that should be protected like other … important protections that we have’ (Morrison 2022). He also commented that ‘Australians shouldn’t have to worry about looking over their shoulder, fearful of offending an anonymous person on Twitter’, and that we ought to ‘veer away from the artificial and phoney conflicts, boycotts, controversies and cancelling created by anonymous and cowardly bots, bigots and bullies’.¹⁸ He posited those opposed to his (on this occasion) absolutist version of freedom of religious expression as themselves being not only cowardly and bullying – clearly negative behaviours – but also as ‘bigots’. The use of the term ‘bigot’ in this context arguably did two things. First, it recalled a historic moment in 2014 when then Attorney General George Brandis argued for the repeal of federal vilification laws and declared in parliament that, ‘people have the rights to be bigots’ (Chan 2014). Second, it posited a section of the community who do not suffer from systemic discrimination – Christians opposed to same sex marriage – as being themselves subjected to bigotry and being victims of discriminatory behaviour.

The third representation was in contradistinction to the first two; that freedom of speech should be limited to prevent harm and ensure accountability. PM Morrison introduced the online trolling Bill by stating that freedom of speech entails a responsibility to be accountable for what you say. He went further, saying that ensuring responsibility and accountability by limiting speech that makes people uncomfortable or that harms them was the only way to be ‘free’:

In a free society such as Australia where we value our free speech, it is only free when that is balanced with the responsibility for what you say. Free speech is not being allowed to cowardly [sic] hide in your basement and sledge and slur and harass people anonymously and seek to destroy their lives … In this country, we value freedom of speech, and freedom of speech means you take responsibility for what you say. (Morrison and Cash 2021)

The fourth representation coheres with the third; that existing laws that limit free speech should be respected and upheld. Morrison confirmed that ‘the Government is committed to press freedom’ (Morrison 2019c). At the same time, he suggested that press freedom can be abrogated because existing laws ought to be upheld:

[W]e have clear rules and protections for the freedom of the press. There are also clear rules protecting Australia’s national security and everybody should operate in accordance with all of those laws passed by our parliament. (Morrison 2019d)

Home Affairs Minister Dutton commented that ‘[w]e have clear rules and protections for that freedom of the press and we also have clear rules and laws protecting Australia’s national security’ (Martin and Lyons 2019). Morrison affirmed this approach, claiming it was shared collectively by other Australians:

I think Australians carry these two things closely, they believe strongly in press freedoms – as do I – but they also believe no one’s bigger than the law. I’m not, you’re not, no one is. (Morrison 2019c)

Here, the ‘law’ was presented as the supreme arbiter of a dispute between two competing national priorities, as though this were simple to resolve.

Morrison asserted the legitimacy of the AFP actions based on following their rules for operation, irrespective of whether these actions impinged upon press freedom:

[T]he raids … occurred in accordance with Australia’s laws and in accordance with the Australian Federal Police that acts independently of government ministers doing their job. (Morrison 2019e)

Two days later, Morrison again justified the raids on the basis that the AFP had followed the correct legal procedures prior to the raids:

the raids were undertaken using a warrant. You know, to get a warrant, you’ve got to see a judge and so they were made consistent with the law. (Morrison 2019c)

When asked by a journalist if it ‘bothered’ him that the AFP raided Smethurst’s home, Morrison replied, ‘It never troubles me that our laws are being upheld’ (Morrison 2019d).

The fifth representation was that free speech ought to be granted only to legal citizens. Minister Hawke stated he denied tennis player Novak Djokovic a visa to play in the Australian Open because his presence 'may lead to an increase in anti-vaccination sentiment … potentially leading to an increase in civil unrest of the kind previously experienced in Australia with rallies and protests which may themselves be a source of community transmission'.¹⁹ His positing of ‘rallies and protests’ as transmission sites, as opposed to fora for the exercise of freedom of speech, was germane. The Federal Government was accused of hypocrisy because, while it had deported Djokovic based on concerns that his presence in Australia could incite anti-vaccine attitudes, it had taken no action against Liberal MPs who had expressed anti-vaccine sentiment that could have the same outcome. When challenged on this, PM Morrison responded,

If you’re an Australian, you’re a citizen, you’re a resident, you can be here and you can express your views. (Morrison 2022)

While Home Affairs Minister Karen Andrews expressed her disapproval of former Liberal MP George Christensen’s attendance at an anti-vaccination rally in Canberra, she merely stated that, as ‘a member of Parliament’, Christensen was able to ‘make his own decisions’ about attending (Gillespie and Tamer 2022). Neither Morrison nor Andrews expressed concern that Liberal MPs would contribute to anti-vaccination attitudes in Australia.

6. Discussion of findings and analysis: multiple and mutually incompatible discourses

These representations were constitutive of both a basic discourse around free speech per se, and specific discourses, which notably conflicted with one another in many ways. Taken together, these discourses are centred around a core human right – free speech – the parameters of which are intimately connected with justice (Van Dijk 1997a, 21). Language about free speech is both constituted by its pre-existing parameters, and constitutive of (Titscher et al. 2003, 147) the free speech the government wished to value and protect.

The government posited free speech as important and as a value to which it is committed; this was an overarching discourse. They claimed that there was free speech ‘for some and not for others’, that ‘you shouldn’t feel intimidated’ to express your views, that ‘we all hold dear’ the principle of liberty in Australia, that we should not be ‘fearful of offending’ others, that we ‘value our free speech’, that ‘the government is committed to press freedom’ and Australians ‘believe strongly in press freedoms’, that Australians ‘can express your views’, and that ‘our laws are being upheld’. They frequently used collective first person pronouns – ‘we’, ‘our’ – to posit their position as that of the polity as a whole. They communicated a strong presupposition of the importance of freedom of speech as a value, which the government and its Ministers explicitly committed themselves to. This conveyed that their policy approach to relevant issues was protective of freedom of speech, consistently with prior usage.

Yet simultaneously we identified two other discourses that posited free speech differently. One was a discourse about the need to expand free speech. The other was a discourse about the need to limit free speech. These two discourses had ‘affinities’ (Dunn and Neumann 2016, 7). Both suggested that the current situation for free speech is inadequate and needs to change; and both embodied a clear call for policy change. Examples included that universities ought to protect free speech better, that university funding should be withheld if they do not, that people of faith needed better speech protection, that policy change is needed for people to feel comfortable, that people’s ability to express their beliefs needed to be protected, and that we did not want civil unrest as a result of expressing different views. Importantly, actors couched some of these claims in the normative language of ‘should’; what the government and the country should do to protect free speech. This positioned these discourses as legitimation discourses for their policy change. The government sought public support for policy change commensurate with their positioning of free speech as a value that warranted protection.

However, these two other discourses also had significant differences; indeed the differences were so great that they were demonstrably inconsistent with one another. One called for greater protection for free speech while the other called for greater restrictions on free speech. They harnessed the idea that the harms of some speech, and the capacity of some speech to make people feel uncomfortable, justified either greater protection for free speech, or greater restrictions on free speech, depending on the policy under discussion. These discourses meant that the government’s authoritative (Chilton and Schäffner 1997, 207) claims on policy positions directly contradicted their own authoritative claims on other policy positions.

On the one hand, Morrison and his ministers claimed that greater protections for free speech were necessary in relation to universities; they showed support for free speech in principle, and advocated a protective approach to free speech including when it might make people feel uncomfortable or challenged. In relation to religious freedom, Morrison advocated a more protective approach to freedom of speech than is currently accepted in Australian law. He privileged the freedom to express oneself as one wishes, regardless of potential harms to others. He did not differentiate between the harms of vilification and people feeling offended; he elided this distinction. Morrison seemed to advocate a position in which existing vilification laws should be nullified in favour of free speech. In relation to speech that may arouse anti-vaccination sentiment when expressed by Australian citizens, Morrison and his minister conceptualised free speech as the right to express one’s views regardless of the impact it may have on others’ wellbeing. This discourse therefore posited free speech as unlimited and unbridled.

On the other hand, the Morrison-led government simultaneously posited limitations on free speech as necessary in relation both to speech that makes people feel uncomfortable and speech that has the capacity to harm. This occurred in relation to freedom of the press, where Morrison and Dutton discursively positioned press freedom as straightforwardly able to be overridden by national security law. Indeed, Morrison’s comment that ‘no one’s bigger than the law’ implies the primacy of law in general, even where it disproportionately or negatively impacts upon freedom of speech or the press. Yet in other contexts he did not take this position. Although the online trolling bill was unlikely to deal with the actual harms of trolling, in this case the Morrison government suggested that the potential to cause harm ought to be an important consideration, and that free speech entails a responsibility for speakers to be accountable for harms. He called for a balance to be exercised between a person’s right to speak and how they speak to ensure they do not harm others. He emphasised that all speech can not be completely free, and that some harmful speech ought to be restricted. In the example of anti-vaccination discourse, the government viewed potential harmful impacts of speech as relevant to policy implementation. This discourse therefore posited free speech as having appropriate limits.

These two discourses were engaged in by one government in a single parliamentary period. The differences between these two discourses are so stark that they completely contradict one another and are mutually incompatible. This renders the government’s discourse unstable, inconsistent, discontinuous and incoherent; a clear example of dissimulation (Chilton and Schäffner 1997, 212). How is the public to understand what the government means by free speech in a context where authoritative government speakers engage simultaneously in such contradictory discourses?

This dissimulation was furthered through a conditional subject positioning around free speech. We have already noted that the government stressed the importance of free speech to Australians, and argued in its basic discourse that Australians possess this right. But there were two instances in which they distinguished between Australians and non-Australians. In discussing freedom of religion, the government contrasted Australians - who they posited as being able to speak freely about their beliefs - with people ‘in some other countries’. While this may have reassured Australians that free speech is protected domestically, in so doing it also cast other (unnamed) countries as insufficiently protective of free speech, thus positing the collective ‘we’ in Australia as a human rights protector (cf HRLC 2021). The other example was in relation to anti-vaccination discourse, in which they went further, positing that only legal Australian citizens possess free speech. This differentiation is not commensurate with any sensible reading of the legal protections that exist for free speech, nor with the premises underlying the basic discourse. This intervention sought to sever a section of the community from membership of the category of people who possess free speech, for the purposes of legitimating a policy decision that was inconsistent with the government’s basic discourse reifying freedom of speech.

In sum, we have shown inconsistencies, discontinuities and confusion in these discourses. We showed first that the differences between these discourses are at least as significant, if not more so, than the affinities between them. We have also shown instability and discontinuity in the discursive use of the idea of ‘free speech’. Returning to the precepts of political discourse analysis, this makes clear that dissimulation was occurring. Political discourse of government ministers and the Prime Minister is a ‘constitutive part of the political process’ (Van Dijk 1997b, 21). This of course does not require complete coherence or consistency. Yet, at the same time, political discourse involves persuasive language use of which the public is not fully aware (Chilton and Schäffner 1997, 207). In this case the instability and confusion rendered it impossible for the public to understand or get a sense of what the government meant by invoking free speech, or what kind of policy was important to the government. Their claims to hold free speech in high regard were meaningless and self-contradictory. Government speakers concealed what they meant by free speech at the same time as they claimed to support it. One of the strengths of political discourse analysis lies in its ability to reveal ‘covert’ language use (Van Dijk 1997b, 38) that has political functions for how the public understands the government’s policy positions. Our analysis has shown that there was a high level of language use around free speech, but that the covert messaging was incoherent. The discursive management of public opinion to seek legitimation for policy decisions (Van Dijk 1997b, 39–40) was therefore undermined.

At the same time, these discourses can be described as ‘weaponised’ in so far as they sought to build public and parliamentary support for policy in a manner that was inconsistent with prior usage, and inconsistent with the available evidence (Ashley 2020; Cockbain and Tufail 2020). This understanding of weaponisation is usefully connected with a political discourse analysis. It augments the utility of political discourse analysis by rendering visible a manifestation of incoherent and mutually inconsistent, multiple, simultaneous discourses as a form of dissimulation. In this sense, finding that the Morrison government discursively weaponised free speech in the period under study renders sensible the otherwise nonsensical use of language to garner support for a range of public policies by a government.

7. Conclusion

In this article we have undertaken an original analysis of the political discourse of the Morrison government ministers and the Prime Minister during the period August 2018 to Dec 2021. We sought to uncover the ways in which language was being deployed to legitimate policy positions and calls for policy change. In doing so, we evidenced first that free speech as a concept in public discourse has grown in prominence. Second, we uncovered a basic discourse in which free speech was posited as a value the government sought to protect, which was also consistent with the prior place of free speech in Australian political culture. Third, we also identified other discourses that took place in the same time period. These discourses advocated policy change that was claimed superficially to be consistent with the protection of free speech. However, we identified deep inconsistencies and contradictions in those discourses, which were so profound as to render them mutually incompatible. This meant it was not possible for the public to understand what the Morrison-led government meant when it sought to legitimise its policy directions by drawing on the value of free speech.

Representations within discourses establish the preconditions and parameters for political action (Dunn and Neumann 2016, 34, 61). But when those representations are too confused, they are unable to establish those preconditions and parameters; unless the precondition is, in fact, a deliberate attempts to sow confusion. While the government engaged in discourse around free speech in ways that sought to legitimise their approach on a specific topic, as is the purpose of political discourse, this process was impaired and forlorn. We are forced to conclude either that they did not care about the messages they were sending, or that they were attempting to confuse the public.

This study is, of course, limited to less than one full parliamentary term. We have also been unable to uncover whether, and the extent to which, these discourses in Australia resonate with debates around free speech in other countries, and are therefore indicative of a larger global trend. We do acknowledge that the volume of public discourse around free speech also has increased in the United States in a similar time period (Bowman and Gelber 2021, 263). It is likely that this study fits within a broader, global phenomenon of a changing discourse around free speech commensurate with a broader discursive politics that is seeking to cast dominant groups as victims of human rights abrogations (eg Cammaerts 2022; Sengul 2023).

We are able to conclude that the Morrison-led government weaponised free speech. They did this by claiming to hold this concept as a core value, identifying it as a value held in the collective first person ‘we’, and claiming they would engage in policy change to protect it while simultaneously engaging in policy change that put free speech at risk, using the same discursive approach. This study has made a further contribution by drawing the concept of weaponisation explicitly into a political discourse analytic framework. This helps to clarify how weaponisation of discourse can be characterised, while simultaneously strengthening frameworks for political discourse analysis of the phenomenon of weaponisation for future work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Australian Research Council [grant number DP190100266].

Notes on contributors

Katharine Gelber

Katharine Gelber is Head of the School of Political Science and International Studies at the University of Queensland, a Fellow of the Academy of Social Sciences Australia, a former Australian Research Council Future Fellow (2012-2015), and a former President of the Australian Political Studies Association. Her expertise is in freedom of speech and speech regulation, with research projects into the regulation of hate speech, especially online, and other allegedly harmful speech. Her recent publications include the jointly edited Free Speech in the Digital Age (Oxford Uni Press, 2019) with Susan Brison, and Free Speech After 9/11 (Oxford Uni Press, 2016) as well as articles in Journal of Public Policy, Parliamentary Affairs, Law and Society Review, Political Studies, Contemporary Political Theory, Melbourne University Law Review, Review of International Studies, and the Australian Journal of Human Rights.

Molly Murphy

Molly Murphy is a Sessional Lecturer at the School of Political Science and International Studies at the University of Queensland. Their research interests include free speech, counter speech, public discourse, and speech-based harms.

Notes

1 The Oxford English Dictionary notes several definitions for weaponise, both literal and figurative. Figuratively: ‘to use or repurpose (something) in order to undermine, criticise, or oppose others, or in order to spread discord’. The Macquarie Dictionary: ‘to use (something, such as an issue, gender, race, religion, etc.) as a means of attack or to cause injury or harm’.

2 The Advertiser (Adelaide), the Age (Melbourne), the Australian, the Courier Mail (Brisbane), the Herald-Sun (Melbourne), the Hobart Mercury (Hobart), the Northern Territory News (Darwin), the Sydney Morning Herald (Sydney), the Sun Herald (Sydney), and the West Australian (Perth).

3 We used Factiva to search for articles, letters to the editor, and editorials. We excluded identical duplicates and extraneous newspaper items, like sports, obituaries, and market data.

4 Higher Education Support Amendment (Freedom of Speech) Act 2021 (Cth).

5 Smethurst v Commissioner of Police [2020] HCA 14, 9.

6 Smethurst v Commissioner of Police, 44.

7 Smethurst v Commissioner of Police, 104.

8 Exposure Draft, Religious Discrimination Bill 2019 (Cth).

9 Exposure Draft, Religious Discrimination (Consequential Amendments) Bill 2019 (Cth).

10 Second Exposure Draft, Religious Discrimination Bill 2019 (Cth).

11 Religious Discrimination Bill 2021 (Cth), Religious Discrimination (Consequential Amendments) Bill 2021, Human Rights Legislation Amendment Bill 2021.

12 Cth, Parliamentary Debates, House of Representatives, 10 February 2022, 310 (Paul Fletcher, Minister for Communications, Urban Infrastructure, Cities and the Arts).

13 Explanatory Memorandum, Social Media (Anti-Trolling) Bill 2022 (Cth) 2.

14 Cited in Djokovic v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs [2022] FCAFC 3, 64.

15 Commonwealth, Parliamentary Debates, House of Representatives, 25 November 2021, 10811 (Scott Morrison, Prime Minister and Member for Cook).

16 Religious Discrimination Bill 2022 (Cth) s 12.

17 Anti-Discrimination Act 1977 (NSW), ss20C, 38S, 49ZT, 49ZXB; Anti-Discrimination Act 1991 (Qld), ss124A, 131A; Crimes Act 1900 (NSW), s 93Z; Criminal Code 1913 (WA), ss 76-80H; Discrimination Act 1991 (ACT), s67A, Racial and Religious Tolerance Act 2001 (Vic), ss24, 25; Racial Discrimination Act 1975 (Cth), s18C; Racial Vilification Act 1996 (SA), s4.

18 Cth, Parliamentary Debates, House of Representatives, 25 November 2021, 10814 (Scott Morrison, Prime Minister).

19 Cited in Djokovic v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs [2022] FCAFC 3, 64.