Paternalism in Australian parliamentary debate: the case of drug testing social security recipients

Abstract

Across the globe, welfare conditionality and sanctioning increasingly permeate social welfare programs. Paternalism is one of the key normative rationales invoked when both scholars and politicians debate the legitimacy of this reform. With a view to bringing the scholarly and political debates into closer conversation with each other, this paper examines how paternalism manifests in political debate. We systematically analyse the paternalist arguments made by Australian federal parliamentarians in favour of the virtually identical 2017 and 2018 policy proposals to drug test welfare recipients, both of which resulted in a stalemate. We find that paternalistic arguments primarily employed soft, weak, and welfare paternalism, with heavy emphasis on the purported benefits of the intervention, limited emphasis on the issue of personal liberty, and noticeable silence about autonomy and consent. These findings shed light on the scholarly features of paternalism that are obscured in contemporary political discourse. This analysis can direct political philosophers to features of paternalism that need more attention as well as suggest ways that drug and welfare policy advocates may engage more effectively with paternalist arguments.

Introduction

In attempting to map the debate over the legitimacy of welfare conditionality, paternalism is one of a number of normative rationales identified by scholars (Molander & Torsvik, 2015, 2022; Paz-Fuchs, 2008; Watts & Fitzpatrick, 2018). Its importance as a key justification for the increasing attachment of behavioural conditions to welfare is well-established in the academic literature (Eriksen & Molander, 2019; Mead, 1997). And yet, despite a growing body of evidence of the harms experienced by people subject to welfare conditionality, the prevalence of paternalist justifications for welfare conditionality persists (Dwyer et al., 2019). The focus of our paper is not the impact of paternalist policies, but the nature of paternalist justification for welfare conditionality. More specifically, we seek to systematically analyse the ways in which paternalism manifests in Australian federal Parliamentary debate on the proposal to make the provision of welfare conditional on social security recipients passing drug tests. Our intention is to bring together the fine-grained distinctions made in political philosophy with empirical data about how paternalism appears ‘in the wild’ (in parliamentary debate).

Intervening into the lives of others, for their own good, is a familiar aspect of societies’ approaches to illicit drug use. Given that illicit drug use is seen to be problematic – for health, social, psychosocial, and economic reasons – state interventions to reduce or cease illicit drug consumption has by its very nature paternalistic overtones. There is an academic literature on the application of paternalism as a normative rationale for prohibition (Hirst, 2020; Husak, 1989; Smith, 2002) but here we are concerned with welfare conditionality. State interventions become particularly acute when we focus on those people who use drugs who are also marginalised, poor and receive social security payments (of all varieties). It is here that paternalism can be used to justify state interference where there is perceived to be a mutual obligation for people receiving assistance from the state. This is the context for this work, examining how paternalism features in the arguments for intervention towards people who use drugs and who are in receipt of government financial support.

In the academic literature, Dworkin (2020) identifies three conditions for something to be classed as paternalistic: that it interferes with a person’s liberty or autonomy (interference condition); that the person does not consent (non-consent condition); and that the intervention is premised on the belief by those introducing it that it will improve the person’s welfare (benefit condition).

Given liberal society’s presumption of respect for individual autonomy, paternalism in public policy is widely criticised and often treated with suspicion (Johnsen et al., 2021). However, carefully reasoned arguments can be made for the justifiability of paternalism in some situations where vulnerable people are at risk of serious harm (Parsell & Marston, 2016; Watts et al., 2018). These arguments must not only address the stringency of the liberal respect for autonomy, but also an epistemic issue inherent to paternalism known as the ‘knowledge problem’ (Rizzo & Whitman, 2009). For policymakers to justifiably implement a paternalist intervention, they must have the knowledge needed to evaluate whether such an intervention would genuinely improve the welfare of those being interfered with. This knowledge equates to an understanding of the ‘True Preferences’ of those being interfered with. These are preferences which have not been warped by the cognitive biases, difficulties in information processing, or insufficient willpower that paternalists allege are ‘tainting’ the preferences of those being interfered with. Rizzo and Whitman (2009) claim that as policymakers do not have sufficient information to judge true preferences, they are unjustified in legislating paternalism. Evidently, for paternalism to be justified, there is also a high epistemic standard that must be met. This longstanding debate on the ethics of paternalism within legal and political philosophy can be (and has been) used as a resource for clarifying the public debate about the ethics of welfare conditionality and drug policy (e.g. Anker, 2016; Bellomo, 2019; Hirst, 2020; Thomas & Buckmaster, 2011). And yet, there seems to be a significant gap between this literature and political practice.

To analyse paternalism, philosophers often employ carefully formulated thought experiments (e.g. Dworkin, 2020; Feinberg, 1986). These highly idealised case studies allow philosophers to make fine grained distinctions, and therefore systematically theorise the conditions under which paternalism is, or is not, acceptable. The distinctions between soft/hard, weak/strong, and moral/welfare paternalism relate to differing justifications for normative legitimacy and involve different kinds of burdens of justification. This theoretical variegation means that policies which exhibit paternalist logics can nonetheless differ quite substantially. Depending on the different empirical assumptions being made, different forms of paternalism can be evident in arguments being made in support of the same policy. Table 1 summarises and illustrates these three distinctions with reference to the field of drug policy. ¹

Table 1. Three paternalist theoretical distinctions

When analysing real world cases of paternalism, policy scholars usually construct what they consider the most plausible intention behind a policy intervention, and the most plausible empirical assumptions – for example, with respect to the agency of the targets of the intervention – underpinning it (Le Grand & New, 2015). This enables them to distinguish paternalist from non-paternalist policies and to determine that particular policies are instances of certain forms of paternalism.

While good philosophers aim to engage the best possible construction of their opponents’ argument, this magnanimity leads to a potential gap between paternalism as it exists in philosophy, and paternalism as it exists in political debate. Philosophers are able to suppose away the inconvenient gaps in empirical knowledge with which actual legislators must grapple. Our paper makes a novel contribution to the literature on paternalism by engaging closely with a corpus of statements made in support of a paternalist policy by legislators. We aim to provide an empirical picture of paternalistic lawmaking that will shed light on scholarly features of paternalism that are obscured in the public policy discourse. This will not only contribute to scholarly understanding of public ethics, but offer insights relevant to political actors – such as welfare rights groups – who seek to contest paternalist policies.

Methods

Case study

We take as our case study a paternalistic policy proposal by the Commonwealth government of Australia to subject unemployed social security recipients to drug testing. In an omnibus bill introduced in 2017 (Social Services Legislation Amendment (Welfare Reform) Bill 2017 (hereafter SSLA17)), the Australian federal government sought to introduce a drug testing trial. This trial was to involve the testing of 5000 randomly selected recipients of Newstart and Youth Allowance – two of the main social security payments for the unemployed – for specified illicit drugs in three distinct geographical locations. If those select recipients did not undertake the drug test, then they would have their social security payments sanctioned. Those recipients who tested positive were to be subject to income management for 24 months, this involving the quarantining of 80 per cent of their social security payments onto a ‘basics card’ which could only be used in select ways at select stores. For this 24-month period, these welfare recipients would be subject to further drug tests, and if they tested positive for a second time, they were to be referred to medical experts who would set out treatment activities (such as rehabilitation and counselling), which the recipient must undertake to fulfil their mutual obligation requirements. If these welfare recipients did not meet their expanded mutual obligation requirements after testing positive, they would be subject to benefits sanctions.

After significant opposition in the Senate, the government removed the drug testing trial schedule from the omnibus bill, and this revised bill passed in March 2018. Later in 2018, the federal government reintroduced a substantively identical drug testing proposal to the legislative process in the Social Services Legislation Amendment (Drug Testing Trial) Bill 2018 (hereafter SSLA18). However, due to the 2019 federal election, the bill lapsed at the end of forty-fifth parliament. The new federal government, formed by the same political party but with different leadership, reintroduced the proposal in the Social Services Legislation Amendment (Drug Testing Trial) Bill 2019 (hereafter SSLA19) but did not attempt to pass it through the Senate. This Bill lapsed following the change of government in 2022.

The proposal to drug test welfare recipients is the latest iteration of a welfare reform agenda developed and implemented over many decades termed ‘new paternalism’ (Mead, 1997). This agenda seeks to make welfare programmes more active in shaping the behaviour of recipients through greater conditionality and sanctioning on the grounds that people who are ‘dependent’ on welfare are not competent judges of their own best interests. We, therefore, expect the paternalist arguments in this context to be especially well-honed and, by extension, a good representation of paternalism ‘in the wild’. This expectation is buttressed by our previous work on this parliamentary debate, where we established that paternalism was one of the most prevalent moral frames in the speeches of supporters of the drug testing proposal (Curchin et al., 2021).

Data

In our previous analysis of this policy debate, we used the Parliamentary Hansard record to construct a dataset of all spoken political debate about SSLA17 and SSLA18 (Curchin et al., 2021). This dataset included both direct and indirect debate about these proposals, spanning across a wide array of parliamentary speeches: from second reading speeches to question time speeches. For the analysis at hand, we refined this dataset. First, we isolated all speeches, which our previous analysis had revealed to involve paternalist moral framing. Second, amongst this subset of the data, we determined which of these speeches involved parliamentarians actively employing paternalist arguments to promote a particular stance, and which of those speeches merely involved rebuttal of paternalist arguments. We excluded the latter from our dataset. Lastly, Thomas Weight re-examined the entire dataset from our previous analysis to ensure that all speeches wherein paternalism manifested were included. This refined dataset comprised approximately ninety-six pages, with Table 2 describing the distribution of speeches across political parties.

Table 2. Number of paternalist speeches by political party

Analytic framework

Our central research question is ‘How does paternalism manifest in parliamentary debate over drug testing of welfare recipients?’ Our analysis is oriented by the following sub-questions:

• In practice, how relevant are the different theoretical forms as made within the philosophical literature? Which form/s of paternalism are dominant in parliamentary debate on this policy?

• What empirical and moral premises do parliamentarians rely on to make paternalistic arguments?

• How do paternalists situate their support for the policy in relation to liberal values?

• How do parliamentarians overcome the knowledge problem?

Coding method

We coded the dataset at the level of the utterance. In practice most utterances contained a sentence or two. We left uncoded utterances that were not relevant, such as those in which the speaker simply introduced themselves, coding instead only utterances which contributed to the speaker’s framing of the drug testing proposal and their engagement with a paternalistic rationale for this policy. We employed a two-part coding strategy. Firstly to answer the question of which form of paternalism is dominant in this debate, Thomas Weight operationalised the forms of paternalism for analysis (see Table 1) by developing a theory-driven coding schema which included inclusion and exclusion criteria. This coding process took us some way to answering our second research question which concerned the moral and empirical premises parliamentarians relied upon. Each form of paternalism is underpinned by moral assumptions and has different empirical conditions of acceptability. For example, hard paternalists do not consider the volitional capabilities of the people they are interfering with as relevant to their position, whereas soft paternalists do – in their eyes those being interfered with must be acting involuntarily when they harm themselves. As such, coding for theoretical forms alerted us to the moral and empirical premises that parliamentarians are relying on.

Secondly to answer our remaining questions we employed a hybrid coding schema with theory and data-driven elements We operationalised Dworkin’s (2020) characterisation of the three essential conditions of paternalism by coding utterances that pertain to the interference with someone’s liberty or autonomy (the interference condition), the consent of the subjects of the intervention (the non-consent condition), or the purported benefits of the intervention (the benefit condition). This is not a mutually exclusive typology, as an utterance’s meaning can implicate more than one condition of paternalism. Within these three categories, we made further data-driven distinctions about their content. For example, for utterances about benefit, we distinguished between those which concerned the efficacy of welfare conditionality, and those which concerned the consequences of drug consumption for personal welfare. This hybrid schema enabled us to see which parts of paternalism politicians silence, obfuscate, or emphasise, and how they do so. It helped us focus on the premises explicit and implicit in their arguments and also revealed how they grapple with the knowledge problem and the critical trade-off between welfare on the one hand, and liberty and autonomy on the other.

To test the deductive codes and start developing the inductive codes, TW coded the Minister’s second readings speech for SSLA17 and SSLA18. We chose these speeches because they are particularly important in the drug testing proposal corpus. Each speech is delivered by the Social Services Minister – the parliamentarian who oversees federal social security policy – and therefore contains the most formal and authoritative justification of the proposal. Following further refinement to the coding schema, Thomas Weight then coded the parts of our dataset which concerned SSLA18. The authors discussed the coding results and made further refinements to the schema before TW re-coded the SSLA18 debate and coded the SSLA17 debate. Katherine Curchin then double coded 20 per cent of the speeches for quality control.

Results

Theoretical forms

Table 3 shows the number of instances and the number of speeches wherein parliamentarians making paternalist arguments employed particular theoretical forms (across both rounds of debate and in total).

Table 3. Summary statistics: Theoretical forms

As Table 3 shows, parliamentarians employed soft paternalism more than they did any other theoretical form. This led to a consistent and dominant characterisation of welfare recipients as being in a vulnerable state wherein they lack the autonomy to address their drug problems and unemployment. Drug testing was held to be a form of ‘support’ or ‘assistance’ aimed at helping people recognise their problems, break free of their addiction, find employment, and move off welfare. Proponents claimed that the measures are designed to ‘help people take control of their lives’ (MP O’Brien, 2018), implying that people who use drugs lack agency. However, despite the predominance of soft paternalism, hard paternalism still manifested in a significant way. On eleven occasions, parliamentarians endorsed interference with undesirable voluntary behaviour. For example: ‘There will be appropriate consequences for people who deliberately miss an appointment without a reasonable excuse or refuse a drug test in order to avoid a possible result’ (MP Tehan, 2018). It was also common to see soft and hard paternalism coincide with each other, either in the same sentence or in the same speech. For example, MP Tony Pasin claimed that ‘We also know that, without assistance, many people with substance abuse problems can’t or won’t take action to help themselves’ (2017, our italics). In this utterance, normative justification relies on the agency of the subject being both unwilling (‘won’t’ – hard paternalism) and unable (‘can’t’ – soft paternalism).

The next most common theoretical form was welfare paternalism. Interestingly, whilst the alleged benefits of drug testing was the most discussed element in our case study, it was seldom specified exactly how drug testing would benefit welfare recipients. However, when the benefits were explicitly specified, they were primarily characterised as a welfare, rather than moral, benefit. The discussed welfare benefits were fivefold: remedy of the harmful drug addictions many welfare recipients supposedly have; an increase in the welfare recipient’s possibility of gainful employment; an improvement in welfare recipients’ ability to be effective parents; a reduction in the incidences of violence amongst welfare recipients; and an increase in the self-esteem of welfare recipients. Whilst moral paternalism was the least common theoretical form in the case study, it was not absent altogether. It appeared most clearly in relation to the valorisation of paid employment. Work was held to not only be better for one’s physical and psychological wellbeing, but also for one’s moral wellbeing. Some speeches referred to the notion of a good citizen, alleging that a good citizen is in employment, for example:

at the end of the day, that’s all that matters: people getting off drugs, getting off alcohol, being restored, and then being sent into the workforce so that they can make a meaningful contribution as Australian citizens in their community (MP Hastie, 2018).

In this case study, the Weak/Strong distinction did not have as much purchase as the Soft/Hard and Welfare/Moral distinctions. This is because parliamentarians making paternalist arguments in this case study seldom discussed the wants and desires of welfare recipients. However, when they did, there was a general presumption that welfare recipients wanted to find work, but they just needed help in changing the ways in which they seek that end (i.e. they needed help to stop using drugs) – this is the language of weak paternalism. For example, MP Mark Coulton claimed that ‘The purpose of testing welfare recipients in the drug testing trial is to find people who may need help to address a barrier to employment that they may have not acknowledged or disclosed previously’ (2018). Whilst less common, there were some parliamentarians who used strong paternalism. They held that tougher welfare conditionality is necessary because some welfare recipients are deliberately choosing welfare receipt over paid employment. For example, MP Andrew Laming stated that:

For those on the margins that are just opting in and out, using ice as a convenient excuse not to work – we know there are around 85,000 persistent evaders and many of them are using drug and alcohol as an excuse – your time is up. The holiday is over. We will, through a series of reforms, eventually prevent you leading that life. (2018)

Analysis by the three conditions of paternalism

Table 4 provides summary statistics (across both rounds of the debate) indicating the prevalence in the dataset of the three conditions of paternalism. There was a vast over-representation of references to the measure’s benefits (80 per cent), more limited references to interference (19 per cent), and only three occasions where the consent or otherwise of the targets of intervention was spoken of.

Table 4. Summary statistics: Paternalist utterances

Condition 1 – Interference

Parliamentarians frequently described the content of the drug testing proposal, and in the process recognised that it would involve some degree of interference but offered no reflection on the moral significance of interfering with citizens’ liberty. Speeches typically stressed that the measure ‘is designed to restrict their [i.e. people who test positive for drugs] access to cash and limit their ability to use their payments to fund further harmful drug use, while not reducing the amount of payment they receive’ (Sen Scullion, 2018). Only one speech endorsed the value of liberty: MP Andrew Laming stated that ‘By breaking the status quo, there were three or four key considerations. That first consideration is that we must respect the privacy of the individual and adhere to the legal obligations of this nation’ (2018).

Condition 2 – Non-consent

The question of the consent or otherwise of the people affected by the proposed measure was rarely drawn attention to in these speeches. Of the three utterances we identified, none of them involved a recognition that welfare recipients did not consent to the proposed intervention. Instead, they involved an attempt to establish precisely the reverse – that in some way welfare recipients did, in fact, consent to this intervention. Noting the challenges that welfare recipients face, it was assumed that they were asking for help. For example, MP Bert Van Manen stated that ‘These are people living in very difficult circumstances, and these are people crying out for assistance’ (2017).

Condition 3 – Benefit

The purported benefits of the drug testing measure was the feature most commonly referred to by a wide margin. Parliamentarians spoke frequently of the imperative for action created by the harms welfare recipients were experiencing, due to their welfare dependency or drug consumption. Many speeches sought to establish that the proposed intervention would be effective in addressing these harms, thereby benefitting the welfare recipient. We identified utterances about the efficacy of specific elements of the proposed policy such as income management (n = 13), mandatory treatment (n = 8), and drug testing (n = 6), utterances about the efficacy of the proposed measures more generally (n = 28) and the efficacy of welfare conditionality with respect to behavioural reform (n = 2). It is noteworthy that some paternalist parliamentarians did not imagine that people experiencing problems were unable to seek and receive help, but rather that mandatory drug testing would speed up the process for people slow to recognise that they needed assistance. Interestingly, there were several utterances that concerned the lack of availability of current treatment options for people with drug abuse issues (n = 14), with parliamentarians making the pragmatic argument that implementing these measures (which included provision for more treatment in some locales) would address the shortfall in treatment and therefore benefit the welfare recipient.

Discussion

The results of this study suggest there is a significant gap between how scholars theorise and think through paternalism, and how politicians debate and legislate paternalism. That is, while paternalism in the philosophical literature is a principled stance that is thoroughly grounded in the liberal tradition, paternalism ‘in the wild’ is a more practical and variable position that is relatively unconstrained by liberal principles and epistemic difficulties. Philosophical discussion of the conditions under which state paternalism is justified is focused on questions of personal liberty, autonomy, well-being, and the role of the state in managing the trade-off between all three. These foci are not reflected in paternalist discourse in this debate. Engaging with the normative dimensions of paternalism would benefit the debate’s deliberative quality.

The Liberal party, from whom most paternalist arguments came, represents itself as the party of individual freedom, with a commitment to ‘a lean government that minimises interference in our daily lives’ (Liberal Party of Australia, 2022). And yet, for parliamentarians employing the paternalist frame, there was a systematic disregard for the significant violation of personal liberty that this intervention entailed. Whilst there were numerous utterances concerning the interference entailed by the paternalist measures, only one of them explicitly addresses the fact that this intervention constitutes a violation of the liberty of social security recipients. Alongside this, there was only one utterance which affirmed the importance of liberty considerations regarding this intervention. Relatedly, rather than deliberating about the value of the autonomy of social security recipients and the conditions under which it could legitimately be infringed upon, paternalists assumed that violations of autonomy are justified if the intervention creates benefit for those being interfered with, irrespective of how significant these benefits are. The non-consensual nature of the proposal was obscured by politicians who imagined that the desire for intervention could straightforwardly be deduced from poverty or drug use. As several submissions to the Senate Committee inquiries into these proposals demonstrate, social security recipients are strongly opposed to the measures (ACOSS, 2018; AUWU, 2019). This discursive tactic, and the general lack of parliamentary discussion of consent and autonomy, suggest that paternalists in this debate occupy a significantly different frame of reference than philosophers: one that is less liberal and more pragmatic.

Parliamentarians who employed the paternalist frame had a cavalier attitude to the serious epistemic difficulties that must be addressed for a paternalist policy to be justified. Though most paternalist utterances concerned the purported benefits that this intervention would have for the social security recipients, their engagement with empirical research was superficial. The discussion about the efficacy of the proposals consisted primarily in appeals to ‘common sense’ and generalisations from the supposed efficacy of the Cashless Debit Card (CDC) trials, even though an independent review into the CDC found that the evidence of its efficacy is ‘mixed and cannot offer any definite conclusion’ (Mavromaras et al., 2021, p. 2). There was limited discussion about the causes and nature of drug addiction, and the discussion that was had was not evidence-based. In the limited instances wherein the aetiology of drug addiction was discussed, it was generally asserted that unemployment was positively related to drug addiction. In an utterance characteristic of this, MP Llew O’Brien stated that:

Substance abuse has many contributing factors. Welfare dependency is just one factor, but it is one that government is able to influence but has neglected for some time. In society, drug addiction can contribute to a cultural dependence which, when passed on from generation to generation, becomes a difficult cycle to break. (2018)

But perhaps the most indicative example of the epistemic insouciance of most parliamentarians was in how they addressed the knowledge problem, applying across all forms of paternalism. The epistemic burden on paternalists to be certain of the benefits to the target group of an intervention which restricts liberty was deftly dealt with by stating that drug testing would only be introduced initially as a ‘trial’ aimed at improving the evidence base. Politicians who supported the measure held that in the absence of conclusive knowledge of what would work to move unemployed people who have drug problems into employment, there was an imperative to try something new. Whereas scholars who discuss the knowledge problem place the onus on paternalists to demonstrate the benefit of their intervention (Rizzo & Whitman, 2009), parliamentarians placed the onus on opponents of paternalism to show that there is no benefit They also set the standard of proof high. They held that paternalism is justifiable until the specific intervention is demonstrated not to benefit the targets. They asserted that the intervention must be trialled in Australia and that neither the failure of similar policies in other countries nor the opinion of experts in the drug treatment sector that this intervention was unlikely to benefit its targets could be considered sufficient evidence. At the same time, the trial the government proposed would have been of little use to opponents of the measure because its design was not sufficiently rigorous to be capable of conclusively establishing a lack of positive impact.

Feinberg (1986, p. 8) distinguishes between mixed and unmixed paternalism: between laws that are justified partly on paternalistic grounds, and laws that are justified only on paternalistic grounds. As our previous analysis of this debate shows (Curchin et al., 2021), this paternalist proposal clearly draws upon several other justifications, and is therefore a clear instance of mixed paternalism. The present analysis reveals a second way in which paternalistic laws can be ‘mixed’. We found that there was significant theoretical variegation in the paternalistic justifications of this proposal. Within the same speech politicians vacillated between different forms of paternalism, even dissonant ones (such as when an utterance contained soft and hard paternalism). The multiple forms of paternalism employed within speeches reflects blurred normative justifications, and suggests a pragmatic approach to policy argumentation. The variegation within the paternalism in this case study lends further support to our core finding: that in contrast to the liberal philosophical understanding, paternalism in the wild (at least, in this study) is a pragmatic position that is relatively unconstrained by liberal principles, clarity in forms of argumentation, and any epistemic difficulties.

We suggest that the dominance of certain forms of paternalism – soft over hard, weak over strong, and welfare over moral – reflects the dominant ways that drugs are currently framed in the Australian context. The cultural meaning attributed to drug use is highly culturally specific and varies greatly through time (Ritter, 2022, pp. 13–16). In the parliamentary speeches of Australian paternalists, drugs are understood to pose a threat to health and (financial) well-being rather than associated with immorality per se. People who use drugs are understood to share the same ultimate goals – such as employment – as other members of society/politicians, but as incapable of identifying their own needs and purposefully acting to address them. The relative dominance of these forms of paternalism also tells us much about the value attributed to paid labour in Australia. Employment is imbued with a moral quality as well as being considered crucial to well-being.

On the basis of our findings, we offer the following reflections that may apply to Australian political actors seeking to contest paternalist policies. Active engagement with the normative dimensions of paternalism, such as making clear the liberty violation entailed by the policy and the lack of consent on the part of the targets of the intervention may be useful. This is also likely true for those who wish to defend drug testing on more principled paternalistic grounds. After all, consent can effectively mediate the paternalist trade-off between liberty and welfare – even in, as Danny Scoccia argues (2013, p. 80), instances of hard paternalism. Opponents of the policies may also usefully seek to champion the value of autonomy and particularly the autonomy of members of marginalised social groups. Furthermore, our analysis affirms the importance of contesting the empirical assumption that the intervention will improve the well-being of those subject to it. Successful contest of policy proposals such as these will be facilitated by enabling the voices of people affected by the intervention to be heard in public debate and contesting the othering they have been subjected to. Paternalist interventions are often necessitated by a concern for treating others with respect, such as in the case of those who are medically incompetent (Howard, 2018). Indeed, it is likely that many Australians interpret the paternalist framing of the drug policy proposal this way. Yet when equal respect and concern is understood to encompass not only attention to citizens’ welfare but to their voice and agency, the paternalist framing of the drug testing proposal appears far less egalitarian. One strategy suggested by our analysis available to drug policy advocates seeking to counter the proposal would be to argue that the failure by paternalists to treat people who use drugs as possessing moral authority with regards to their own lives is a failure of concern and respect.

Conclusion

Our analysis shows that whilst the philosophical discussion of paternalism attends carefully to the target’s agency, there is a remarkable incuriosity shown towards the agency of the targets of drug testing by paternalist politicians. This collection of paternalist speeches marginalises the perspectives of people subject to interference with their liberty. Discussion of consent is absent. Rather than a careful trade-off being made between the values of wellbeing and autonomy, we see a cavalier attitude towards the autonomy of people who are suffering and stigmatised by virtue of their drug use and/or reliance on income support. Rather than treating their autonomy as outweighed by the potential well-being benefits of the intervention, paternalists treated their autonomy as nullified by their suffering, and the poor life choices they presumed this suffering resulted from. The potential benefits of drug testing to the target population were frequently asserted but politicians relied on common sense understanding of people who use drugs or receive income support instead of using empirical evidence to support their claims. In fact, the preoccupation with these benefits was in tension with the expert consensus that the policy would prove harmful to people who use drugs (Curchin et al., 2021). Politicians’ inattention to the perspectives of those who would be subject to the intervention, the disregard for expertise and evidence on how the policy would practically affect them, and the incuriosity towards their subjective experience all speak to a lack of care. This lack of care is ironic given that support and compassion are key themes in paternalist rhetoric, which appear in this data set. While the violation of liberty registers in philosophical discussion, the accompanying failure of care goes under the radar. Our study is therefore an important corrective to the prevailing abstract discussions about paternalism.