Sage Journals: Discover world-class research

Abstract

The ‘Robodebt’ scheme was an initiative pursued by the Australian Department of Human Services between 2016 and 2019 to increase the amount of money recovered from supposed ‘overpayments’ to recipients of welfare benefits. Drawing on the rich body of empirical material generated by the Royal Commission into the Robodebt Scheme as well as journalists and academic observers, this paper develops an understanding of the affair from the perspective of the sociology of organizations. Particular use is made of a growing body of research in the organizational sociology of ignorance. Following an outline of the main features of Robodebt, the paper explains the significance of the conception of ignorance as more complex than the mere absence of knowledge in organizational life. It then examines the specifics of the way in which Robodebt casts light on the role played by systemic, wilful ignorance in the relationship between law, bureaucracy and politics. The paper concludes with some reflections on the senses in which Robodebt was a manifestation not only of a crisis, fiasco or scandal, but also of the normal operation of the ‘will to ignorance’ (Nietzsche) in organizational life.

Keywords

ignorance organization rule of law politics bureaucracy Robodebt welfare

The question is, then, whether there are circumstances that change the relationship between knowing and ignorance, perhaps to a point in which ignorance becomes the most important resource of action. (Luhmann 1998: 94)

Introduction

What became known more broadly in Australia as the ‘Robodebt’ scheme, and within the relevant government bodies as the Online Compliance Intervention (OCI),¹ was an attempt by the Australian Department of Human Services (DHS) between 2016 and 2019 radically to increase the amount of money recovered from supposed overpayments to welfare benefits recipients² between 2010 and 2018. The designers and promoters of the scheme promised that instead of dealing with 20,000 ‘compliance interventions’ a year, it would be possible to complete 20,000 a week, generating around AUD$1.7 billion in savings over 5 years.

There are similarities between the Robodebt affair³ and similar scenarios characterized by a shift to automation in other settings, such as the Dutch toeslagenaffaire – the childcare benefits affair – between 2013 and 2019 (Besselink, 2021a, 2021b; Damen, 2021; Frederick, 2021; Pasman, 2021; Van den Berge, 2021; Venema, 2023) and the Horizon IT Post Office scandal in the United Kingdom (Marshall, 2022). There is considerable scope, then, for a comparative analysis (Hadwick and Lan, 2021), but also many important differences in the ways politics, bureaucracy and law intersected, as well as in the particular challenges posed for the rule of law. Even within the Australian context, there are also other examples of how automation has been mobilized by state agencies (Barrett, 2024), also with important differences as well as similarities, so here the focus is on the specifics of the Robodebt affair.

The problem with Robodebt was that it was fundamentally illegal and by any rational consideration it should never have seen the light of day. It was accompanied by constant public criticism, join-the-dots explanations of how it was contrary to the legislation by legal and policy scholars (Carney, 2018a, 2019; Hanks, 2017) and journalists (Henriques-Gomes, 2019; Knaus, 2019; Murphy, 2019), two Senate inquiries (Senate Community Affairs References Committee, 2017, 2020), a number of critical decisions in the Administrative Appeals Tribunal (AAT) and two 2019 court cases ruling against the scheme (Amato, Masterton). It came to a crashing halt in 2020, with the government having to refund AUD$746 million, and write off debts totalling AUD$1.71 billion.

A rich and varied body of terminology has emerged to define the affair, as a crisis, a scandal, a stunning failure of competence, institutional failure and malfeasance, extortionate, negligent, unconscionable, a scandalous example of state dishonesty and collusion, an elaborate shame, a manifestation of venality, incompetence and cowardice and ‘a catastrophic, multisystem failure within the Australian Public Service’ (Morton, 2022b). The affair, wrote Justice Murphy, was ‘a shameful chapter in the administration of the Commonwealth social security system and a massive failure of public administration’ (Prygodicz: para 5). It was, declared Catherine Holmes, a former Chief Justice of the Queensland Supreme Court, and head of the Royal Commission (RC) examining Robodebt:

a crude and cruel mechanism, neither fair nor legal, and it made many people feel like criminals. In essence, people were traumatised on the off-chance they might owe money. It was a costly failure of public administration, in both human and economic terms. (RC: xxix)

All of these characterizations are accurate, around which are organized important and useful accounts of the affair itself and also its wider implications (Braithwaite, 2020; Carney, 2018a, 2019; Hanks, 2017; Middleton, 2023; Morton, 2022a; Naylor, 2023; Whiteford, 2020, 2021).

An invaluable resource in coming to grips with the Robodebt affair and its implications has been the proceedings and final report of the Royal Commission into the Robodebt Scheme (Commonwealth of Australia, 2023).⁴ The report is a rare treasure trove of empirical detail about the functioning of both the DHS and the allied Department of Social Services (DSS), generating unique data not usually available to social science researchers because of the Royal Commission’s legal powers to extract normally inaccessible information. In addition to the knowledge about Robodebt in the public domain – academic articles, journalism, testimony from affected individuals, reports of the Commonwealth Ombudsman (Glenn, 2017) and two Senate inquiries – the Royal Commission was able to compel the production of internal documents such as email correspondence as well as internal policy minutes and proposals at various stages of drafting, and then forensically to cross-examine the authors and recipients of all these documents. These two bodies of data, in addition to the commentary by legal academics and journalists, drew out in painstaking detail the exact organizational mechanics of the establishment and maintenance of Robodebt over 4 years that tell us a great deal about the relationship between public bureaucracy, politics and law. The Royal Commission powerfully addressed all four of the ‘layers of meaning’ of policy fiascos identified by Bovens and ‘Hart (2017 [1996]): what unacceptable harm was done, who made or allowed it to happen, what were the causes and rationales for what happened and where should individual responsibility be allocated?⁵

The scheme did not meet the legal requirements of a conspiracy (Prygodicz: para 6), but its skirting of the relevant law was nonetheless organized and systemic, and not reducible to a ‘stuff up’. Throughout the Royal Commission report there is a body of strong evidence that key players were perfectly aware of the illegality of the scheme, but that their practised exercise of the mechanisms of organized ignorance made it possible to sustain a ‘plausible deniability’ (Dinges and Zakkou, 2023; Strayhorn, 2019; Walton, 1996) that was only cracked open under the forensic scrutiny of a skilful barrister and Commissioner backed by the authority of law. One important focus has been on attributing individual responsibility to the key players in the affair, but although it is true that different choices could have been made by many of those central actors, it is important to see their decision-making processes as located within a particular organizational setting.

As Dennis Thompson (2005) has put it in relation to the question of organizational ethics:

we should stop thinking about ethics so much in terms of individual vices (bribery, extortion, greed, personal gain, sexual misconduct) and start thinking about it more in terms of institutional vices (abuse of power, improper disclosure, excessive secrecy, lack of accountability). We have been paying too much attention to individual, and too little to institutional, vice. (p. 4)

This article contributes to the understanding of Robodebt as an example of ‘institutional vice’ by developing an analysis of the affair from the perspective of organizational sociology, highlighting the ways in which it brought to surface a number of important aspects of the relationship between law, bureaucracy and politics that deserve particular attention. Robodebt was certainly both a ‘scandal’ and a ‘fiasco’ in many respects (Whiteford, 2021) – in the foreseeable financial, emotional and psychological harm it inflicted on a large number of people, some driven to suicide (RC: 181-4; 325-46), and in being doomed to fail to achieve its objectives. In other respects, though, it was – disturbingly – a manifestation of the relatively normal operation of the contemporary state, the dynamics of which are worth analysing with the conceptual tools of organizational sociology. The machine had indeed malfunctioned, but perhaps not in the way we first think of when we use terms like ‘fiasco’ or ‘failure’. If it can be understood as the tip of a policy fiasco iceberg, its particular processes, mechanisms and dynamics become that much more broadly significant.

Among the many examples of policy failures and fiascos (Bovens & ‘t Hart, 2016, 2017 [1996]; McConnell, 2015; McConnell and Tormey, 2019), Robodebt stands out as a distinctive case in that in addition to the harms it inflicted, it involved the development of policy that was specifically unlawful, only brought to a halt with decisions of the Federal Court of Australia. In that respect, it deserves specific attention as a case study in the interaction between failures of both policy and the rule of law (Carney, 2018a, 2019).

A central, defining feature of the Robodebt scheme made especially clear in the proceedings of the Royal Commission, where individual actors were pressed to explain their actions, was its anchorage in ignorance, in key actors deliberately not knowing and wilfully avoiding knowing the relevant legal constraints. As Commissioner Holmes summed it up:

It is remarkable how little interest there seems to have been in ensuring the Scheme’s legality, how rushed its implementation was, how little thought was given to how it would affect welfare recipients and the lengths to which public servants were prepared to go to oblige ministers on a quest for savings. Truly dismaying was the revelation of dishonesty and collusion to prevent the Scheme’s lack of legal foundation coming to light. Equally disheartening was the ineffectiveness of what one might consider institutional checks and balances – the Commonwealth Ombudsman’s Office, the Office of Legal Services Coordination, the Office of the Australian Information Commissioner and the Administrative Appeals Tribunal – in presenting any hindrance to the Scheme’s continuance. (RC: iii)

For this reason, although the affair has many more implications for questions of legal and institutional design, as well as bureaucratic ethics and responsibility, my particular focus will be the dynamics of what Nietzsche (1966 [1886]: 35; 1968 [1884]: 328) termed the ‘will to ignorance’ and how that systemic organization of wilful ignorance operated in the Robodebt affair.

For Nietzsche (1968 [1884]: 328), life itself requires being encompassed by ‘a great, firm dome of ignorance’. My argument will be, then, that although quite a lot is known about the alliances between the will to power and the will to knowledge (Clegg, 2014; Foucault, 1972, 1980), it is also important, building on Nietzsche’s observations, to grasp the importance of the will to ignorance, which forms a triangular relationship with power and knowledge, a central ‘circuit of power’ (Clegg, 2023) posing particular challenges for the rule of law and ethical conduct in complex organizations. The value of looking closely at the Robodebt affair is that, as a case study in the structure and dynamics of organizational ignorance, it is a revealing example of this power–knowledge–ignorance configuration with particular relevance to the ever-shifting operation of the rule of law.

What Was Robodebt? ‘Mountains of gold’

In every country that has ever had a developed welfare state, the expenditure on unemployment and disability benefits, child support and other social services has long been a favourite target of enthusiasts for neoliberal austerity, ‘balancing the budget’, and so on. A frequently used approach has been to argue that the system needs ‘tightening’, and ‘loopholes’ need ‘closing’, all self-evident problems if one assumes that a significant proportion of people receiving a welfare benefit simply have to be cheats and frauds. In Australia, this enthusiasm has operated across the political spectrum. In relation to Robodebt, the Liberal-National Party coalition has been in the spotlight, but the Labor Party has also been very keen to present itself as ‘tough on welfare fraud’. In August 1990, for example, the incoming Minister for Social Security, Senator Graeme Richardson, announced that the Labor Government was going to close two such loopholes: identifying those recipients who had received benefits to which they were not entitled (because they had understated their income), and claims from more than one income support agency. Explaining the philosophy underpinning the initiative, he said:

It means our income-support system is further protected against overpayments. This is only fair since the welfare system is funded by the taxpayers of Australia who expect it to be run fairly, efficiently and at the least possible cost. We should not be in the business of protecting cheats. (RC 11, RC Doc 1)

This was to be achieved through ‘data-matching’: instead of a manual check of the recipient’s income tax data, recipients were required to provide their tax file number (TFN), so that it could be done mechanically, and also to facilitate data-exchange between all the various income support agencies.

All this was being made possible, noted Richardson, by ‘developments in computer technology’, that made it possible to match the identity and income records of different welfare agencies and to identify cases where the same individual has different profiles in different agencies. However, the costs of confirming the debt, as requiring by the Social Security Act, in the form of departmental labour – mainly tracking down employer and bank records – often exceeded the potential debt itself, so that only a small proportion of the data matches generated automatically could be pursued and turned into recoverable debts. All of this unprocessed data seemed to promise, as Commissioner Holmes put it, ‘mountains of gold’ (RC: 655).

When Scott Morrison⁶ was appointed the new Minister for Social Services in December 2014, the slogans he placed at the centre of his politics included ‘ensuring welfare integrity’ and being the ‘welfare cop on the beat’. In many respects, however, this was merely continuing an established tradition in Australian society and politics of constructing welfare recipients as the ‘undeserving poor’, combining the conflation of fraud and inadvertent non-compliance with the promise of huge savings from reductions in social welfare expenditure.

A little before Morrison’s appointment as Minister, in June 2014, a number of DHS staff had become inspired by a bright idea of how to solve the problem of the mismatch between possible data matches indicating overpayment and the department’s ability to follow them up. Drawing on recent developments in Web 2.0 technology as well as ‘nudge theory’ and theories of ‘behavioural economics’ (Thaler and Sunstein, 2008), they turned to the concept of ‘self-service’ via an online portal to dissolve the high labour costs associated with confirming debts with the recipient, past employers and banks.

In a way the fact that the scheme was made possible by automation is misleading, because data-matching had been pursued for some time, so that was not really the driving factor. The core changes being proposed were to treat the data-matching process based on the averaging of income over a financial year (the data from the Australian Tax Office – ATO) as definitive, and shifting the onus onto the recipient of clarifying that the income earned was in fact uneven across the year. One was guilty of welfare fraud until one could prove one’s innocence. Recipients themselves were now required to contact former employers and banks and dredge out their financial records going back 7 years, more than the 5 years required under tax law (Carney, 2018b). As the Commonwealth Ombudsman explained:

The OCI matches the earnings recorded on a customer’s Centrelink record with historical employer-reported income data from the Australian Taxation Office (ATO). Parts of the debt raising process previously done manually by compliance officers within DHS are now done using this automated process. Customers are asked to confirm or update their income using the online system. If the customer does not engage with DHS either online or in person, or if there are gaps in the information provided by the customer, the system will fill the gaps with a fortnightly income figure derived from the ATO income data for the relevant employment period (‘averaged’ data). . . . DHS no longer using its information gathering powers [under sections 63, 192 and 195 Social Security (Administration) Act 1999] to request information directly from third parties, such as employers. Under the OCI, it is now the customer’s responsibility to provide this information. (Glenn, 2017: 1, 5)

If your employer was no longer in existence, or declined to provide records that far back, or provide them in the correct format: tough, you had a debt to pay. When DHS staff did their sums, it looked like more than AUD$1b could be reclaimed from welfare recipients over 4 years.

This idea was a dream come true for a politician like Scott Morrison, keen to demonstrate his ability to trim the budget (Murphy, 2019), as well as every senior public servant hoping for a bright future under a new Minister, generating enormous pressure to deliver on that desire across the two government departments, and to bulldoze through inconvenient obstacles like illegality, immorality or unfairness (RC: 107).

But Was It Legal?

In principle, the relevant legislation barred the core logic of Robodebt from the very beginning in relation to three issues. First, the reversal of the onus of establishing/disproving a debt was contrary to the provisions of the Social Security Act (Carney, 2018a). As Peter Hanks explains, ‘the existence of a debt to the Commonwealth is something to be established by the Commonwealth’, and ‘the Social Security Act cannot be read as requiring that a person who has received a social security payment establish that there is no debt – indeed, s 1222A(a) and s 1223(1) deny any such possibility’ (Hanks, 2017: 7). Second, entitlements were determined for income each fortnight as actually earned, not as averaged over the year (Carney, 2018b). Another way of putting this can be regarded as a third fatal flaw in the legality of the scheme: one internal DHS legal advice prepared in 2017 pointed out that if a recipient’s entitlement to a payment was established on one basis – their fortnightly income, and not averaged over 12 months – then any recalculation of that entitlement and that income has to proceed on the same basis: as Fiveash put it, ‘Where there is no “income averaging” mechanism in the first instance of rate calculation, there can also be no “income averaging” in the second instance’ (RC: 187, RC Doc 2).

The scheme was subjected to a constant barrage of criticism from all sides: its victims, relevant non-governmental organizations (NGOs), legal scholars, journalists, opposition politicians, the Senate, the AAT and the courts. But the monster refused to die, and it took one more final stake thrust into its heart in the form of a class action for unjust enrichment (Prygodicz v Commonwealth of Australia), commenced in November 2019. The Solicitor-General advised the government that it was bound to fail in defending the claim, the government announced refunds on all debts raised using averaging, amounting to AUD$746 million, and writing off debts totalling AUD$1.71 billion. By June 2020 the scheme was dead.

In May 2022 the governing Liberal-National Party, led then by none other than Scott Morrison, lost power to the Australian Labor Party (ALP), led by Anthony Albanese. The ALP’s election platform had included the establishment of a Royal Commission into Robodebt, as recommended by the second Senate Inquiry (Senate Community Affairs References Committee, 2022), to clarify how such a public policy catastrophe could have happened and to ensure these do not repeat in the future. The Royal Commission, led by Commissioner Catherine Holmes, a former Queensland Supreme Court Judge, was established on 18 August 2022, delivering its report on 7 July 2023.

The affair itself revealed both the strengths and weaknesses of the use of law to temper the arbitrary exercise of power (Krygier, 2011, 2016a, 2016b). In relation to strength, the final outcome was testimony to the power of civil society, various affected individuals, the #notmydebt campaign and a number of journalists, although in the end it was only the force of court judgements that finally gave the rule of law the edge needed to end to the scheme. At the same time, it was also an example of the limits to the capacity of legislative and judicial branches of government to temper the power of the executive branch because of a ‘limited adherence to the doctrine of separation of powers’ (Carney, 2019: 9), raising important questions about exactly how that aspect of the rule of law is currently structured and being reconfigured.

Running through all of the responses to the Royal Commission’s challenges to the official narrative about Robodebt was a persistent thread of ‘unknowingness’. The more senior and more highly paid public servants and politicians were, the worse their memories and powers of analysis appeared to be, the greater their inability to read and understand the relevant legislation, the worse their ability to recall significant conversations and emails or grasp the consequences of their actions, and the greater their incapacity to read or understand any document, let alone legal advice. Wilful, organized ignorance was central to the affair, requiring some more considered examination.

Organized Ignorance

Although knowledge and ignorance are most often understood as opposites, human beings’ relationship to both has always been complex. In the Old Testament, it was eating from the tree of knowledge of good and evil that had Adam and Eve cast out from the Garden of Eden, and since then have been innumerable examples of the positive functions of ignorance and the negative effects of ‘too much’ knowledge. As well as a ‘will to power’ and a ‘will to truth, Nietzsche (1966 [1886]) also wrote of the ‘will to ignorance’, on which the first two are based:

And only on this now solid, granite foundation of ignorance could knowledge rise so far – he will to knowledge on the foundation of a far more powerful will: the will to ignorance, to the uncertain, to the untrue! Not as its opposite, but – as its refinement! (p. 35; see also Landy, 2002)

Given that the world is infinite, all knowledge is by definition focused on the exclusion of other aspects of the world, about which the knower can only remain ignorant (Ungar, 2008). Knowledge is indeed power, but so, too, is ignorance.

Whether in relation to close interpersonal relations, small-scale social groups, or large-scale societies governed by complex organizations, knowledge and non-knowledge are interwoven with each other and are mutually constitutive. Ignorance, as Luhmann (1998) put it is ‘the other side of knowing’ (p. 81). Pascal used the analogy of an expanding balloon to explain the relationship: if the contents of the balloon constitute knowledge, and ignorance consists of the surface area of the balloon, ignorance expands alongside increasing knowledge (Bakken and Wiik, 2018). Sooner or later every scholar comes to the realization that the more they know, the more they realize remains to be known. As Michael Smithson (1985) argues, the sociology of knowledge requires the companionship of a sociology of ignorance (p. 151). He also highlights the various social norms against knowledge, organized around questions of privacy, civility and avoidance of conflict. There are always layers of meaning to social interactions to which individuals in different social positions have different types of access, depending on the type of knowledge/ignorance typical of their social position.

Ernst Renan (1990 [1882]: 11, 16) declared the ability to forget essential to nation-formation, and today it would be hard to find a single example of national identity – or indeed, national memory – that did not have at least one ‘dirty secret’ hidden away somewhere and conveniently forgotten. Stanley Cohen (2001) also explains the centrality of denial of atrocities and suffering to state-formation,⁷ and other scholars have outlined the ways in which mechanisms of denial, obfuscation, the generation of doubt and the over-production of information function across non-state actors such as corporations as well as states (Barton and Davis, 2018; Schwarzkopf, 2020). As Moore and Tumin argued in 1949, ‘ignorance must be viewed not simply as a passive or dysfunctional condition, but as an active and often positive element in operating structures and relations’ (p. 795).

Furthermore, what is experienced as knowledge is often in fact an artefact of meta-ignorance – ignorance of one’s ignorance (Smithson, 1985: 156) or the Dunning–Kruger effect (Dunning, 2011). As R.D. Laing (2005 [1970]) put it, ‘If I don’t know I don’t know, I think I know’ (p. 58). Here, there are connections between systemic ignorance and what Alvesson and Spicer (2012, 2016) term ‘functional stupidity’ in organizations, a refusal to engage adequately with the body of information and knowledge relevant to an organization’s long-term concerns for the sake of short-term advantage. The 2008 financial crisis, they point out, was ‘a testament to the stupidity lurking at the heart of knowledge-based societies’ (Alvesson and Spicer, 2016: n.p.).

There is also a family connection with the problem of hypocrisy, in the sense of a mismatch between what is said to be taking place and what is in reality taking place (Brunsson, 1989, 2003). However, this discrepancy at the heart of the concept of hypocrisy captures only part of the dynamics of the organization of ignorance. Equally important is the dimension of deliberate, systemic ‘unknowingness’ which becomes most evident when hypocrisy is subjected to any sort of scrutiny and challenge.

It will take some time for studies of ignorance to catch up with the volume of scholarship and writing on knowledge; Moore and Tumin’s (1949) early observations on the social functions of ignorance were never followed up systematically. Nonetheless, there has been a steadily growing field of ‘ignorance studies’ or ‘agnotology’ (Burke, 2023; Dossey, 2014; Gross and McGoey, 2022; Smithson, 1985, 1989; Ungar, 2008; Croissant, 2014; Proctor, 2008; Stehr, 2012; Verburgt, 2020; Zerubavel, 2006; Zwierlein, 2016), treating ignorance as much more than simply the absence of knowledge, across a wide variety of fields, including in relation to organizations and management (Jalonen, 2023; Stel, 2019). Even expertise can be constituted in terms of its own denial or self-negation, as Deval Desai (2023) argues in relation to the expertise concerning rule of law reform. In general terms, ‘agnotology’ is the scientific study of ‘agnosis’ – the generation of ignorance – aiming to ‘explore how ignorance is produced or maintained in diverse settings, through mechanisms such as deliberate or inadvertent neglect, secrecy, and suppression, document destruction, unquestioned tradition, and myriad forms of inherent (or avoidable) culture-political selectivity’ (Proctor and Schiebinger, 2008: vii). A distinction is generally made between unintended ignorance – unknown unknowns, known unknowns, absence of knowledge – and intended ignorance – keeping information secret within an organization, keeping the outside world ignorant or different parts of the organization ignorant of the others.

Recent work in organization studies that moves beyond regarding ignorance simply as something to be eliminated, to examine it as central to organizational structure and dynamics, includes the work of Alvesson et al. (2022), Bakken and Wiik (2018), Dana (2006), Jalonen (2023), Knudsen et al. (2023), McGoey (2007, 2012a, 2012b, 2019). Roberts (2012, 2022), Schaefer (2019), Stel (2019) and Van Portfliet and Fanchini (2023). As Tore Bakken and Eric Wik (2018: 1119) put it, the management of ignorance is of equal importance to the management of knowledge, arguing that ‘the basic problem of organization theory is ignorance’, making the core task of organization theory ‘to explore and describe ignorance in its many facets and to recreate displaced ignorance in order to open up new spaces of possibilities’. The concern is then not just who and why individual actors may be ignorant of important concerns, but how ignorance is institutionalized, ‘baked’ into the organization’s form, structure and dynamics.

As Linsey McGoey observes, there are a variety of ‘ways that cultivating ignorance is often more advantageous, both institutionally and personally, than cultivating knowledge’. She uses the term ‘strategic ignorance’ to capture ‘such practices of obfuscation and deliberate insulation from unsettling information, encompassing “the mobilization of the unknowns in a situation in order to command resources, deny liability in the aftermath of disaster, and to assert expert control in the face of both foreseeable unpredictable outcomes”’ (McGoey, 2012a: 555).

The various outlines of the many forms taken by the social construction of ignorance include Michael Smithson’s (1985, 1989, 2008) overview, Robert Proctor’s (2008) identification of the elements of error, incompleteness, distortion, ambiguity, inconsistency, irrelevance, omission, vagueness or politeness – and Ralph Hertwig and Christoph Engel’s (2016) taxonomy of the ‘deliberate ignorance’ of ‘Homo Ignorans’, capturing both its possible productive effects and more problematic aspects such as avoidance of responsibility and liability.

Against that background, Julia Roberts (2012) distinguishes between the absence of knowledge, and ignoring as the deliberate avoidance of knowledge (p. 217), and outlines three types of ignorance specific to organizational life (Roberts, 2012: 216): first, the absence of knowledge, or known unknowns and unknown unknowns. Second, not knowing existing knowledge, so knowable known unknowns, unknown knowns and errors. Third, the suppression of knowledge: taboos, denials, secrecy and privacy. In relation more specifically to organizations, she distinguishes the following four elements of organizational ignorance, although they can at times overlap:

What is unknown across the whole organization, and what parts of the organization do not know about other parts.

What is unknown in relation to external actors: other organizations, competitors, customers and suppliers.

The focus on pursuit of knowledge in some areas and not others.

Unknowns among members of an organization ‘when such ignorance is actively sought and deployed by the organization’ (Roberts, 2012: 219).

There is, then, a variety of strategies of containing knowledge/information within the organization, and also compartmentalizing it, making it hard to see how policies and actions are interconnected. Secrecy (Costas & Grey, 2018) is also clearly central to the management the organization’s relationship with its environment – creating and sustaining environment ignorance of the organization’s functioning – and making it integral to its ‘operational closure’ and self-referential self-reproduction, or autopoiesis, to use Luhmann’s (1990) terms.

Roberts’ (2022, 2013) typology refers to ‘taboos, denials, secrecy and privacy’, where ‘denial’ is defined as refusal to acknowledge knowledge that ‘does not correspond with one’s existing cognitive frameworks’ (Roberts, 2022: 368), generating cognitive dissonance, for which the denial is a solution. Tolerating such cognitive dissonance through denial is a common response and is sometimes referred to as wilful ignorance or wilful blindness (Heffernan, 2011; McGoey, 2007; Schaefer, 2019). Developing the dimension of ignorance as actively constructed, a number of commentators will place greater emphasis on the deliberate construction of ‘unknowingness’ for particular strategic purposes (Van Portfliet and Fanchini, 2023). Dana (2006) refers to the ‘active avoidance of knowing’, and other terms used include ‘systematic inattentiveness’, ‘unseeing’ and the ‘deliberate ignorance’ of ‘homo ignorans’ (Hertwig and Engel, 2016). The analysis of ‘‘active ignoring’ (Alvesson et al., 2022: 842) is, as Alvesson and his colleagues point out, not entirely new (e.g., Jackall, 2009 [1988]), but ‘its study has until fairly recently remained in the margins of scholarly inquiry’ (p. 841).

A particularly important variation on denial is the psychoanalytical term ‘disavowal’, referring to a type of ‘unknowing knowledge’, where something is known, sort of, but people proceed as if it is unknown. Nadia Abu El-Haj (2023) gives the example of the Palestinian counternarrative to the prevailing story of the founding of the state of Israel, which is rendered ‘unknown’ through a logic that – if were explicitly articulated – would read: ‘yes, we know about the 1948 Nakba, but so what?’ The psychological mechanism at work here is one of, as Octave Mannoni (2003) put it, ‘I know well, but all the same . . .’ (p. 68), a psychic splitting at both the individual level of key actors within the organization (sometimes allied with others outside) and the collective level of the organization as a whole, a defence mechanism against a reality that threatens the self-perception and self-presentation of the organization and its members.

All of this literature on the organization of ignorance has clear applications to the Robodebt affair: key actors were wholly ignorant of the relevant legal provisions as well as being determined to remain so, actively and passively colluding with each other in maintaining that ignorance. More than that, the Robodebt affair provides important distinctive detail on the precise mechanisms and drivers of strategic ignorance in relation to the particular example of legal ignorance, which is significant because of the specific role that law plays in regulating organizational action. Especially relevant for the Robodebt affair is what Alvesson et al. (2022) refer to as ‘the will to try to not know’, encompassing ‘how organizational members actively seek to refrain from being informed and arriving at a qualified judgment about a relevant subject in their professional roles’ (p. 840, original emphasis).

One of the emphases in the critical commentary has been on the decisions and actions (or inaction) of key individuals in the two government departments, which is entirely appropriate for the purposes of identifying the responsibility of the sequence of events. Another has been the structure and processes which facilitated the introduction and maintenance of the scheme, with many of the Royal Commission’s recommendations addressing the structural changes which could make a repeat less likely. It is here that the literature on organizational ignorance can make a distinctive contribution, identifying the deeper organizational dynamics and processes that ran through the affair, effectively compelling all the individuals actors to move in a particular direction. As Alvesson et al. (2022) put it:

It is thus not necessarily a matter of free, stable individual (or group) will but a processual outcome, where a particular trajectory takes over and ‘pushes’ the will – for instance, when a group starts to move in one direction, and individuals feel forced to go along. For individuals, incremental moves toward either increasing or weakening wilful ignorance can be motivated by the strength or weakness of their own volition and by the perceived knowability of the object . . . (p. 842)

The focus of the critical commentary has been on the ways in which the affair has been a ‘failure’ and thus dysfunctional, but it is equally important to grasp the ways in which it was, at the same time, entirely functional for the organizational dynamics of the contemporary state. In some sense what ‘failed’ in Robodebt was overreach – the dissonance with the relevant legal provisions was too extreme, the number of people negatively affected too large and disparate. It remains important, then, to grasp the ways in which more modest versions of its mechanisms and processes remain central to normal organizational life.

The Force of Law: The Royal Commission into the Robodebt Scheme

The Royal Commission made it possible to crack open the strategic, organized ignorance surrounding Robodebt with the ‘force of law’, with its powers to compel production of internal documents, to require all participants to make statements explaining what they did and did not know and why they took the actions they did, and to have skilful barristers – aided by the Commissioner herself – interrogate the key players. When someone said, in response to being asked what they knew about the legal issues, ‘I don’t recall’ or ‘I didn’t turn my mind to that’, the follow-up question could be ‘but wasn’t it your job to know?’, to which there would be no answer or denial: ‘that’s not my understanding’, or ‘I was very busy at the time, you’ve no idea the volume and complexity of the many issues I was dealing with’. The Royal Commission highlights very effectively both the individual strategies and the characteristics of organizational culture underpinning the Robodebt scheme. Commissioner Holmes observes that a central problem was the picture drawn of welfare recipients in politics and the media, which worked as a central driving force behind the affair. She commented that ‘politicians need to lead a change in social attitudes to people receiving welfare payments’, and that it is politicians ‘who need to abandon for good (in every sense) the narrative of taxpayer versus welfare recipient’ (RC ii).

The centrality of the organization of ignorance to the affair is made clear by the fate of knowledge about the scheme’s essential illegality. There was a remarkable consistency, given how varied interpretations of legislation can be, between the very first legal advice prepared within DSS in 2014, and the final assessments of the Solicitor-General in 2020 as well as Justice Murphy’s judgement in the Prygodicz case. In November 2014, David Mason⁸ wrote:

It is flawed as the suggested calculation method (averaging employment income over an extended period) does not accord with legislation, which specifies that employment income is assessed fortnightly. It follows that the debt amount calculated could be incorrect according to law. . . .we can’t see how such decisions could be defended in a tribunal or court, particularly when DHS have the legislative authority to seek employment income information from employers. (RC: xxix)

Five years later, the Solicitor-General said more or less the same thing, constituting ‘an authoritative opinion that the Commonwealth did not have a proper legal basis to raise, demand or recover asserted debts solely on the basis of income averaging, a practice fundamental to the Scheme’. Looking back over the scheme, the Solicitor-General’s assessment made it clear that ‘the Commonwealth had unlawfully been raising asserted debts against current and former income support recipients’ (RC: 542), as predicted by David Mason and others in DSS back in 2014 and 2015.

Yet, this relatively straightforward knowledge had no impact on the introduction and maintenance of the scheme for 4 years. This was where the division between the two departments was especially significant. Strictly speaking DSS was responsible for the legal aspects of welfare provision, and from the outset the DSS lawyers all declared the idea dead in the water without the appropriate change to the legislation. However, somehow this advice never filtered through with any effect to the DHS (RC: 43). The closest the Royal Commission could get to tracking whether it was conveyed to DHS was a cryptic set of dot points in an email that seemed to suggest awareness of the advice (RC: 66-67). Instead of absorbing the advice against the scheme, then, various quasi-legal constructions were formulated sustaining the view that the change was only one of policy,⁹ based on wholly erroneous readings of the legislation, cherry-picking those provisions that seemed to give the department the power to proceed with the scheme, claiming that ‘the best available evidence’ was good enough, and that debts were only raised as a ‘last resort’.

Throughout the life of the scheme there were numerous appeals to the Administrative Appeals Tribunal (AAT), with the decisions in many cases explaining how the scheme was contrary to the legislation, but these were never appealed – as is normally the case when there are differences of legal interpretation – to the second tier of administrative appeal (Carney, 2018a, 2019). Within DHS, the assertion would be made that there was no need to appeal because there was no error of law and that the decision involved no important legal principle – when the legal reasoning in the decisions indicated precisely the opposite (RC: 558). This of course suggests that the leading decision-makers in DHS did in fact know full well that the whole Robodebt scheme was unlawful (Whiteford, 2021: 355), despite their repeated declarations that nothing was amiss. The Royal Commission proceedings brought to the surface the particular strategies adopted by key players like Kathryn Campbell,¹⁰ Malissa Golightly¹¹ and Annette Musolino¹² – acting at a distance, so to speak, from the politicians like Morrison, followed by Alan Tudge¹³ and Stuart Robert,¹⁴ in the interests of plausible deniability – to avoid the question of legality, side-stepping any possibility of seeking definitive advice, no doubt because they were perfectly aware that this would lead to the end of the scheme.

There were a number of key switching points or watersheds in the progression of the scheme, but one that will serve as a useful example was the month that Campbell and numerous other key players went on leave between mid-December 2016 and early January 2017. By this time the public criticism had become intense, with the scheme described by Peter Martin in the Sydney Morning Herald as a ‘weapon of math destruction’,¹⁵ and an investigation by the Commonwealth Ombudsman was underway. The acting DHS Secretary, Barry Jackson, responding to the advice of his DHS colleagues, decided it was high time that advice was sought from the Australian Government Solicitor (AGS), and issued a request that the advice be sought. As the Royal Commission report puts it:

. . . it is clear that by 10 January 2017 there was an internal view held by a number of lawyers at DHS that the arguments in favour of averaging were weak and unconvincing, and that external advice should be sought on that question. Steps had been taken to obtain that advice, including confirming that there was an AGS solicitor available to give it, and drafting the questions to be answered by the external advice. (RC: 188)

On Cambell’s return from leave on 10 January, however, the request for AGS advice just melted away. Jackson told the Royal Commission that this could only have happened if Campbell had issued an instruction that the request for advice be withdrawn. Campbell denied this when questioned in the Royal Commission, claiming that she was ‘unaware that a request for legal advice in the terms described by Mr Gladman had been made and [was] unaware that any such advice, draft or otherwise, had been prepared’ (RC 188). Her lawyer suggested that the request had simply, somehow ‘fallen through the cracks’ of the period when various DHS staff were on leave. Commissioner Holmes, however, did not accept this interpretation, concluding that Campbell was in fact fully ‘aware of Mr Jackson’s request for advice and its progress’. The only possible interpretation was that Campbell has ‘instructed DHS officers to cease the process of responding to Mr Jackson’s request for advice, motivated by a concern that the unlawfulness of the Scheme might be exposed to the Ombudsman in the course of its investigation’ (RC: 189).

Underpinning this and other switching points lay a variety of mechanisms, strategies and practices of systemic ignorance. These mechanisms and practices constituted a combination of (a) ensuring ignorance (I didn’t read the email, the report), (b) creating an impression of ignorance (I can’t recall) and (c) ensuring and maintaining ignorance outside the organization of policies, practices and strategies. The division of labour between two departments created numerous possibilities for the eliciting of organized ignorance, particularly in relation to the legal dimensions of the scheme. Advice would be generated in DSS that recognized the limits imposed by the legislation, but it would somehow fail to filter through to DHS, whose members would then invent their own interpretation of the legal issues. It is hard to know if DSS emails were just ignored and deleted, or if they genuinely remained within DSS, but either way there was a systemic blockage of essential communication, a core aspect of what Alvesson and Spicer (2012: 1204) term ‘stupidity management’.

The clear division that exists between written and oral organizational communication functions as fertile ground for the production of various forms of ignorance. The attribution of responsibility relies heavily on written documents, and what has only been said can only be inferred, imperfectly. Even in relation to the communication that is written, the practice is never to acknowledge receipt of a report or email, or only to use ambiguous language. So even when emails have been sent from a person’s address, there is no need to let it influence one’s decisions or actions. When challenged – in a Royal Commission, by a journalist, or in another way – one’s memory can simply fail.

A number of characteristics of the organizational setting contributed to the introduction and development and maintenance of an unlawful government policy. First, it was a complex, multilayered organizational field, involved two separate government departments, with responsibility for legal questions lying technically with DSS, but DHS staff also willing to claim legal knowledge. The possibilities for accurate knowledge to go ‘missing in action’ – and for individuals to evade responsibility for that – were correspondingly high, as was the potential for the competition for correct interpretation to push things to the lowest common denominator: that is, what would please the Minister.

Second, there was a particular relationship between the realms of politics and bureaucracy organized around a combination of, first, irresistible pressure to respond to the political imperatives of a significant boost to the government’s budget as well as being seen to be a successful ‘welfare cop’, and second, the mechanisms of ‘plausible deniability’, in which public servants simply knew how to enact the wishes of the relevant Minister(s) without ever having to be instructed explicitly, as well as how to avoid passing on bad news, especially in writing.

An important aspect of the systemic production of plausible deniability is the use of promotion as a reward, or demotion or shifting to another, lower-status position as a punishment, for the ability to ‘read the Minister’s mind’, simply to know what policies and practices are expected without having to be explicitly told.¹⁶ There are at least two layers of meaning at play in any complex organization – first, what is explicitly said and written, which can work to establish responsibility and accountability, but which can also in any case underpin plausible deniability if formulated ambiguously, with multiple possible readings (Hodges, 2020; Poznansky, 2022; Davenport and Leitch, 2005),¹⁷ often referred to as ‘dog whistling’. Second, what is known but left unstated, or ‘known’ unknown knowns. As Commissioner Holmes observed, the ‘only rational explanation’ of the failure of senior DHS counsel Annette Musolino to pass on the arguments concerning the legality of the scheme ‘is that she knew DHS executives, including Ms Campbell, did not want advice of that nature’ (RC 242). In this sense, most members of a complex organization intuitively understand this multilayered character of meaning and communication; they simply know that bearers of bad news are not welcome, and those individuals that feel conscience-driven to persist with the difficult news inevitably pay some sort of price as problematic whistleblowers (Kenny, 2023).

This means that such ‘bad news’ can often only be conveyed with decisive effect by powerful external actors – in this case, the courts and the Solicitor-General. There is a silent, invisible undercurrent of ‘knowledge’ that can be presented as ignorance to suit particular political interests as and when required. In this respect the desire to meet ideological, financial and political aims outweighs concerns with legality, fairness or correct procedure, let alone justice and the avoidance of harm to vulnerable citizens. Language plays an important role – in the Robodent affair, figures like Campbell, Golightly as well as Ministers Morrison and Tudge, repeatedly asserted that the method of calculating debt had not changed, in the face of clear evidence that it had (RC: 84). The ethos seems to be that declaring black to white often and loudly enough will make it true, a characteristic of political life that is now very familiar. This approach to language included simply deleting difficult points from the key texts on which decision-making processes rely (RC 72).¹⁸

Another aspect of plausible deniability at the senior levels is to claim to be relying on ‘advice’, either without specifying the source of the advice, but if that fails in subsequent contestation, to make lower-placed members of the organization the scapegoats. With respect to the relationship with recipients, the bureaucratic rhetorical strategies included framing the scheme in entirely positive terms, as acting on the obligation to ensure good management of tax dollars, and the pretence that something actually quite difficult – recovering information about sources and amounts of income, and indeed, that is why DHS wanted to stop doing it, because it required staff resources – is easy for recipients.

Third, the legal sub-field was layered in a way that was fertile ground for very creative legalism – there were internal lawyers as well as individuals claiming to possess legal knowledge, the Administrative Appeals Tribunal (AAT), the Commonwealth Ombudsman, the Senate, the Solicitor-General and the Federal Court, with only the latter having a decisive and conclusive interpretation of the legal issue. Running through this organizational complexity was of course the ‘problem of many hands’ (Thompson, 1980, 2005: 11–32; see also van Krieken, 2006) in the decision-making process – so many individuals were involved in the process that it took on a life and will of its own, making it very difficult to change course once a certain direction had been taken. As the Royal Commission proceedings revealed, those few individuals who did attempt to put a brake on any aspect of the scheme found themselves ignored, abused, moved to another position, their appointment not renewed and so on.

One set of strategies related specifically to the legal dimensions of the scheme. When legal advice was unfavourable, it was ignored. In general, one produces a culture of doubt by turning simplicity into complexity, in much the same way as the tobacco and fossil fuel industries (Oreskes & Conway, 2008; Proctor, 2008). Again, when challenged after the event, one hadn’t read it. If found to be lying when confronted with evidence, that is redefined as ‘sub-optimal phrasing’. The external advice from PwC and Clayton Utz disappeared in a puff of smoke, and heaven and earth were moved to avoid hearing what the Solicitor-General had to say. The legal arguments in the AAT decisions were never appealed, restricting the impact to those particular cases, and their principles certainly not applied, academic commentary by experienced legal scholars dismissed as ‘misunderstanding’ the scheme, in favour of quasi-legal argumentation that would be embarrassing to an undergraduate law student. The correct reading of the relevant legislation by some in-house lawyers was countered by its complete misreading by others, incorrectly applying other parts of the Social Security Act, ignoring the legal reasoning in AAT decisions and academic commentary, and making up their own quasi-legal arguments supporting the scheme. All of this supposedly legal reasoning in support of the scheme was yet one more example of what Alvesson and Spicer (2012, 2016) term the prevalence of functional stupidity in organizational life, as well as the organization of ignorance.

Conclusion: It’s Not a Bug, It’s a Feature

The Royal Commission makes a number of recommendations about the ways to avoid a repeat of similar formations of state policy escaping the constraints of legal provisions intended to a large extent to prevent precisely that sort of exercise of state power, imposing illegal, unjust and morally repugnant practices on the most vulnerable in society. They include a requirement that detailed and authoritative legal advice accompany policy proposals being considered by the Cabinet, publication of first-instance AAT decisions and a strengthened focus on customer service. Those changes, if they are effected, will certainly help prevent the occurrence of a Robodebt-type affair, but my concern is that the affair revealed characteristics of the ways in which politics, bureaucracy and law intersect that will not be especially easy to counteract, even with these sorts of changes. Exactly how the bureaucratic culture that made Robodebt possible could be changed remains, as Laura Tingle (2023) notes, ‘one of the biggest questions arising from the royal commission’.

There is already an extensive literature on institutional design, ethics and responsibility that makes it clear how significant a task it is (Adams and Balfour, 1998, 2014; Bakan, 2004; Bovens, 1998; Brunsson et al., 2022; Jackall, 2009 [1988]; Thompson, 1980, 2005). I have argued elsewhere, for example, that a central problem is the consistency between the forms of behaviour within organizations that are understood as unethical, and those extra-organizational behavioural principles that have in fact become normalized in social life today (van Krieken, 2006). In this sense the Robodebt affair says as much about the nature and dynamics of society, politics and law more broadly as it does about the treatment of recipients of welfare benefits, in relation to the negative construction of welfare recipients as the ‘undeserving poor’, and the idea that the popular will, as represented by elected politicians, should not be constrained by something as undemocratic as law, despite that fact that law does in fact emerge through democratic processes of deliberation.

The problem is a broad one – states are constantly acting illegally, the rule of law seems only to have effect when such action is confronted by the judicial system when court cases can be funded and run. More than that, governments also systematically rework law to align with their political objectives. In Australia, the field of refugee and asylum seeker law is just the most obvious example (Saul, 2010; Triggs, 2017). Should other government departments also be subjected to the critical gaze and forensic analysis of a Royal Commission, it seems reasonable to guess that very similar pictures would emerge, making the Robodebt affair really just the tip of the policy failure iceberg.

The mechanisms and processes of the rule of law, in constraining the arbitrary exercise of power, were painfully slow to bring Robodebt to a halt, by which time enormous harm had been inflicting on a large number of citizens least able to absorb it. The affair clearly constituted a variety of failures in the functioning of administrative law (Carney, 2019), while also revealing some important aspects of the normal functioning of the realm lying at the intersection of politics, bureaucracy and law. What the recent scholarship on the organization of ignorance makes clear is that if future institutional design is to address the systematic avoidance of responsibility and accountability in complex organizations like government departments, the particular role of wilful ignorance requires distinct attention, given its apparent effectiveness in creating an escape hatch for individuals and indeed whole organizations from those institutional mechanisms for legal and ethical accountability tending to focus on the alliance between knowledge and power. In relation to the exercise of state power, it clearly is not possible to mobilize a Royal Commission to examine every government department’s relationship to its legal context, but some significant steps can be taken towards achieving similar effects in relation to other examples of policy failures and fiascos if the will to ignorance is recognized as central to organizational life.

Footnotes

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Robert van Krieken

Notes

References

Abu El-Haj

(2023) ‘We know well, but all the same . . . ’ factual truths, historical narratives, and the work of disavowal. History of the Present 13: 245–264.

Adams

Balfour

(1998) Unmasking Administrative Evil. London: Sage.

Adams

Balfour

(2014) The dynamics of administrative evil in organizations. In: Jurkiewicz

(ed.) The Foundations of Organizational Evil. London: Routledge, pp. 16–30.

Alvesson

Spicer

(2012) A stupidity-based theory of organizations. Journal of Management Studies 49: 1194–1220.

Alvesson

Spicer

(2016) The Stupidity Paradox: The Power and Pitfalls of Functional Stupidity at Work. London: Profile Books.

Alvesson

Einola

Schaefer

(2022) Dynamics of wilful ignorance in organizations. British Journal of Sociology 73: 839–859.

Bakan

(2004) The Corporation: The Pathological Pursuit of Profit and Power. New York: Free Press.

Bakken

Wiik

(2018) Ignorance and organization studies. Organization Studies 39: 1109–1120.

Barrett

(2024) ‘ATO eyeing ramp-up of controversial robotax scheme in bid to recoup $15bn in ‘on-hold’ debts’. The Guardian, 20 February. Available at: https://www.theguardian.com/australia-news/2024/feb/20/ato-eyeing-ramp-up-of-controversial-robotax-scheme-in-bid-to-recoup-15bn-in-on-hold-debts

10.

Barton

Davis

(2018) Introduction. In: Barton

Davis

(eds) Ignorance, Power and Harm, Critical Criminological Perspectives. New York: Springer, pp. 1–12.

11.

Besselink

(2021a) De Afdeling Bestuursrechtspraak en de rechtsstatelijke crisis van de Toeslagenaffaire: Over een onverstandige en onvoldoende reactie. NJB – Nederlands Juristenblad 201: 194–198.

12.

Besselink

(2021b) Hoogste rechter, hoogste onrecht: over de rechter en de wet in de toeslagenaffaire. Nederland Rechtstaat, 8 January. Available at: https://www.nederlandrechtsstaat.nl/hoogste-rechter-hoogste-onrecht-over-de-rechter-en-de-wet-in-de-toeslagenaffaire/

13.

Bovens

(1998) The Quest for Responsibility: Accountability and Citizenship in Complex Organisations. Cambridge: Cambridge University Press.

14.

Bovens

‘t Hart

(2016) Revisiting the study of policy failures. Journal of European Public Policy 23: 654–666.

15.

Bovens

‘t Hart

(2017 [1996]) Understanding Policy Fiascoes, Fiascos. London: Routledge.

16.

Braithwaite

(2020) Beyond the bubble that is Robodebt: how governments that lose integrity threaten democracy. Australian Journal of Social Issues 5: 242–259.

17.

Brunsson

(1989) The Organization of Hypocrisy. Chichester: John Wiley & Sons.

18.

Brunsson

(2003) Organized hypocrisy. In: Czarniawska

Sevón

(eds) The Northern Lights: Organization Theory in Scandinavia. Copenhagen: Copenhagen Business School Press, pp. 201–222.

19.

Brunsson

Gustafsson Nordin

Tamm Hallström

(2022) ‘Un-responsible’ organization: how more organization produces less responsibility. Organization Theory 3: 1–20.

20.

Burke

(2023) Ignorance: A Global History. New Haven, CT: Yale University Press.

21.

Carney

(2018a) Robo-debt illegality: a failure of rule of law protections? Australian Public Law. Available at: https://www.auspublaw.org/blog/2018/04/robo-debt-illegality

22.

Carney

(2018b) The new digital future for welfare: debts without legal proofs or moral authority? UNSW Law Journal Forum 2018: 1–16.

23.

Carney

(2019) Robo-debt illegality: the seven veils of failed guarantees of the rule of law? Alternative Law Journal 44: 4–10.

24.

Clegg

(2014) Circuits of power/knowledge. Journal of Political Power 7: 383–392.

25.

Clegg

(2023) Frameworks of Power. 2nd ed. London: Sage.

26.

Cohen

(2001) States of Denial: Knowing about Atrocities and Suffering. Cambridge: Polity.

27.

Commonwealth of Australia (2023) Royal Commission into the Robodebt Scheme Report. Canberra, ACT, Australia: Australian Government Printer.

28.

Costas

Grey

(2018) Violence and organization studies. Organization Studies 40: 1573–1586.

29.

Croissant

(2014) Agnotology: ignorance and absence or towards a sociology of things that aren’t there. Social Epistemology 28: 4–25.

30.

Damen

(2021) Ik was het niet, ik was het niet, het was de wetgever!’ NJB – Nederlands Juristenblad 354: 371–374.

31.

Dana

(2006) Strategic ignorance and ethical behavior in organizations. In: Tenbrunsel

(ed.) Research on Managing Groups and Teams: Ethics in groups. Vol. 8. Amsterdam: Elsevier, pp. 39–57.

32.

Davenport

Leitch

(2005) Circuits of power in practice: strategic ambiguity as delegation of authority. Organization Studies 26: 1603–1623.

33.

Desai

(2023) Expert Ignorance: The Law and Politics of Rule of Law Reform. Cambridge: Cambridge University Press.

34.

Dinges

Zakkou

(2023) On deniability. Mind 132: 372–401.

35.

Dossey

(2014) Agnotology: on the varieties of ignorance, criminal negligence, and crimes against humanity. Explorations 10: 331–344.

36.

Dunning

(2011) The Dunning-Kruger effect: on being Ignorant of one’s own ignorance. Advances in Experimental Social Psychology 44: 247–296.

37.

Foucault

(1972 [1969]) The Archaeology of Knowledge and the Discourse on Language. New York: Pantheon.

38.

Foucault

(1980) Power/Knowledge: Selected Interviews and Other Writings, 1972-1977. Brighton: Harvester.

39.

Frederick

(2021) Zo hadden we het niet bedoeld: De tragedie achter de toeslagenaffaire. Amsterdam: De Correspondent.

40.

Gearty

(2024) Versions of denial. London Review of Books 46: 24–25. Available at: https://www.lrb.co.uk/the-paper/v46/n02/conor-gearty/short-cuts

41.

Glenn

(2017) Centrelink’s Automated Debt Raising and Recovery System. A Report about the Department of Human Services’ Online Compliance Intervention System for Debt Raising and Recovery. Report by the Acting Commonwealth Ombudsman, Richard Glenn, under the Ombudsman Act 1976, report no.02/ 2017. Canberra, ACT, Australia: Commonwealth Ombudsman.

42.

Gross

McGoey

(2022) Routledge International Yearbook of Ignorance Studies. London: Routledge.

43.

Hadwick

Lan

(2021) Lessons to be learned from the Dutch Childcare Allowance Scandal: a comparative review of algorithmic governance by Tax Administrations in the Netherlands, France and Germany. World Tax Journal 13: 609–645.

44.

Hanks

(2017) Administrative law and welfare rights: a 40-year story from Green v Daniels to Robot debt recovery. AI Administrative Law Forum 89: 1–15.

45.

Heffernan

(2011) Willful Blindness: Why We Ignore the Obvious at Our Peril. New York: Walker.

46.

Henriques-Gomes

(2019) Labor calls on government to scrap ‘malfunctioning’ robodebt scheme. The Guardian, 20 July. Available at: https://www.theguardian.com/australia-news/2019/jul/30/labor-calls-on-government-to-scrap-malfunctioning-robodebt-scheme

47.

Hertwig

Engel

(2016) Homo ignorans: deliberately choosing not to know. Perspectives on Psychological Science 11: 359–372.

48.

Hodges

(2020) Plausible deniability. In: McIntosh

Mendoza-Denton

(eds) Language in the Trump Era: Scandals and Emergencies. Cambridge: Cambridge University Press, pp. 137–148.

49.

Jackall

(2009 [1988]) Moral Mazes: The World of Corporate Managers. New York: Oxford University Press.

50.

Jalonen

(2023) Ignorance in organisations – a systematic literature review. Management Review Quarterly. Epub ahead of print 27 January. DOI: 10.1007/s11301-023-00321-z.

51.

Kenny

(2023) Constructing unknowers, destroying whistleblowers. Ephemera: Theory & Politics in Organization 23: 49–74. Available at: https://ephemerajournal.org/sites/default/files/2023-04/23%281%29Kenny.pdf

52.

Knaus

(2019) Senate inquiry calls for Centrelink robo-debt system to be suspended until fixed. The Guardian, 22 June 2017. Available at: https://www.theguardian.com/australia-news/2017/jun/21/senate-inquiry-calls-for-centrelink-robo-debt-system-to-be-suspended-until-fixed

53.

Knudsen

Grønbæk Pors

Bakken

(2023) Organised ignorance. Ephemera: Theory & Politics in Organization 23: 1–18. Available at: https://ephemerajournal.org/content/organised-ignorance-practices-and-politics-organisation-ignorance

54.

Krygier

(2011) Four puzzles about the rule of law: why, what, where? And who cares? In: Fleming

(ed.) Getting to the Rule of Law. New York: New York University Press, pp. 64–104.

55.

Krygier

(2016a) The rule of law: pasts, presents, and two possible futures. Annual Review of Law & Social Science 12: 199–129.

56.

Krygier

(2016b) Tempering power. In: Adams

Meuwese

Ballin

(eds) Constitutionalism and the Rule of Law: Bridging Idealism and Realism. Cambridge: Cambridge University Press, pp. 34–59.

57.

Laing

(2005 [1970]) Knots. London: Routledge.

58.

Landy

(2002) Nietzsche, Proust, and will-to-ignorance. Philosophy & Literature 26: 1–13.

59.

Luhmann

(1990) The autopoeisis of social systems. Essays on Self-Reference. New York: Columbia University Press, pp. 1–20.

60.

Luhmann

(1998) The ecology of ignorance. In: Luhmann

(ed.) Observations on Modernity. Stanford, CA: Stanford University Press, pp. 75–112.

61.

Marshall

(2022) Scandal at the post office: the intersection of law, ethics and politics. Digital Evidence and Electronic Signature Law Review 19: 12–28. Available at: https://journals.sas.ac.uk/deeslr/article/view/5395

62.

McConnell

(2015) What is policy failure? A primer to help navigate the maze. Public Policy & Administration 30: 221–242.

63.

McConnell

Tormey

(2019) Explanations for the Brexit policy fiasco: near impossible challenge, leadership failure or Westminster pathology? Journal of European Public Policy 27: 685–702.

64.

McGoey

(2007) On the will to ignorance in bureaucracy. Economy & Society 36: 212–235.

65.

McGoey

(2012a) The logic of strategic ignorance. British Journal of Sociology 63: 533–576.

66.

McGoey

(2012b) Strategic unknowns: towards a sociology of ignorance. Economy & Society 41: 1–16.

67.

McGoey

(2019) The Unknowers: How Strategic Ignorance Rules the World. New York: Zed Books.

68.

McGovern

(2021) The origin of the phrase ‘will no one rid me of this turbulent priest? Notes & Queries 68: 266–266.

69.

Mannoni

(2003) ‘I know well, but all the same . . .’. In: Rothenberg

Foster

Žižek

(eds) Perversion and the Social Relation. Durham, NC: Duke University Press, pp. 68–92.

70.

Middleton

(2023) What robo-debt reveals about the public service. The Saturday Paper. Available at: https://www.thesaturdaypaper.com.au/opinion/topic/2023/03/11/what-robo-debt-reveals-about-the-public-service#hrd

71.

Moore

Tumin

(1949) Some social functions of ignorance. American Sociological Review 14: 787–795.

72.

Morton

(2022a) How bureaucrats kept robo-debt alive. The Saturday Paper. Available at: https://www.thesaturdaypaper.com.au/news/politics/2022/12/10/how-bureaucrats-kept-robo-debt-alive#hrd

73.

Morton

(2022b) Inside Morrison’s push for robo-debt. The Saturday Paper. Available at: https://www.thesaturdaypaper.com.au/news/politics/2022/11/12/inside-morrisons-push-robo-debt#hrd

74.

Murphy

(2019) The robodebt horror was all about boosting the budget. That’s the brutal truth. The Guardian. Available at: https://www.theguardian.com/australia-news/2019/nov/30/the-robodebt-horror-was-all-about-boosting-the-budget-thats-the-brutal-truth

75.

Naylor

(2023) Robodebt, Kafka and institutional absurdism in Australia. Alternative Law Journal 48: 299–302.

76.

Nietzsche

(1966 [1886]) Beyond Good and Evil: Prelude to a Philosophy of the Future. New York: Vintage.

77.

Nietzsche

(1968 [1884]) The Will to Power (trans. Kaufmann

Hollingdale

). New York: Vintage.

78.

Oreskes

Conway

(2008) Challenging knowledge: how climate science became a victim of the cold war. In: Proctor

Schiebinger

(eds). Agnotology: The Making and Unmaking of Ignorance. Stanford, CA: Stanford University Press, pp. 55–89.

79.

Pasman

(2021) Kafka in de rechtsstaat. De gevolgen van een leesfout: de toeslagenaffaire ontleed. Amsterdam: Prometheus.

80.

Poznansky

(2022) Revisiting plausible deniability. Journal of Strategic Studies 45: 511–533.

81.

Proctor

(2008) Agnotology: a missing term to describe the cultural production of ignorance (and its study). In: Proctor

Schiebinger

(eds) Agnotology: The Making and Unmaking of Ignorance. Stanford, CA: Stanford University Press, pp. 1–33.

82.

Proctor

Schiebinger

(2008) Preface. In: Proctor

Schiebinger

(eds) Agnotology: The Making and Unmaking of Ignorance. Stanford, CA: Stanford University Press, pp. vii–viii.

83.

Renan

(1990 [1882]) What is a nation? In: Bhabha

(ed.) Nation and Narration. London: Routledge, pp. 8–22.

84.

Roberts

(2012) Organizational ignorance: towards a managerial perspective on the unknown. Management Learning 44: 215–236.

85.

Roberts

(2022) Organizational ignorance. In: Gross

McGoey

(eds) Routledge International Yearbook of Ignorance Studies. London: Routledge, pp. 367–376.

86.

Saul

(2010) The Kafka-esque case of Sheikh Mansour Leghaei: The denial of the International Human Right to a Fair Hearing in National Security Assessments and Migration Proceedings in Australia. UNSW Law Journal 33: 629–661.

87.

Schaefer

(2019) Wilful managerial ignorance and symbolic work: a socio-phenomenological study of managing creativity. Organization Studies 40: 1387–1407.

88.

Schwarzkopf

(2020) Sacred excess: organizational ignorance in an age of toxic data. Organization Studies 41: 197–217.

89.

Senate Community Affairs References Committee (2017) Design, Scope, Cost-Benefit Analysis, Contracts Awarded and Implementation Associated with the Better Management of the Social Welfare System Initiative. Canberra, ACT, Australia: Australian Government Printer.

90.

Senate Community Affairs References Committee (2020) Centrelink’s Compliance Program: Second Interim Report, 978-1-76093-096-70930967. Canberra, ACT, Australia: Australian Government Printer.

91.

Senate Community Affairs References Committee (2022) Accountability and Justice: Why We Need a Royal Commission into Robodebt. Canberra, ACT, Australia: Australian Government Printer.

92.

Smithson

(1985) Toward a social theory of ignorance. Journal for the Theory of Social Behaviour 15: 151–172.

93.

Smithson

(1989) Ignorance and Uncertainty: Emerging Paradigms. New York: Springer.

94.

Smithson

(2008) Social theories of ignorance. In: Proctor

Schiebinger

(eds) Agnotology: The Making and Unmaking of Ignorance, Stanford, CA: Stanford University Press, pp. 209–229.

95.

Stehr

(2012) Knowledge and non-knowledge. Science, Technology & Innovation Studies 8: 3–13.

96.

Stel

(2019) Ignorance. In: Atkinson

Delamont

Cernat

, et al. (eds). Contemporary Perspectives in Qualitative Research – SAGE Research Methods Foundations. Thousand Oaks, CA: Sage. Available at: https://doi.org/10.4135/9781526421036795484

97.

Strayhorn

(2019) Plausible deniability. Journal of Theoretical Politics 31: 600–625.

98.

Thaler

Sunstein

(2008) Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press.

99.

Thompson

(1980) Moral responsibility of public officials: the problem of many hands. American Political Science Review 74: 905–916.

100.

Thompson

(2005) Restoring Responsibility: Ethics in Government, Business, and Healthcare. Cambridge: Cambridge University Press.

101.

Tingle

(2023) Cover-up culture that created robo-debt must also go. Australian Financial Review, 4 February. Available at: https://www.afr.com/politics/federal/the-cover-up-culture-that-created-robo-debt-must-also-go-20230202-p5ched

102.

Triggs

(2017) Overreach of executive and ministerial discretion: a threat to Australian democracy. Victoria University Law & Justice Journal 71: 8–13.

103.

Ungar

(2008) Ignorance as an under-identified Social Problem. British Journal of Sociology 59: 301–326.

104.

van den Berge

(2021) Bestuursrecht na de toeslagenaffaire: hoe nu verder? Over het rechtscharakter van het bestuursrecht. Ars Aequi 11: 987–995.

105.

van Krieken

(2006) The ethics of corporate legal personality. In: Clegg

Rhodes

(eds). Management Ethics: Contemporary Contexts. London: Routledge, pp. 77–96.

106.

Van Portfliet

Fanchini

(2023) How to study strategic ignorance in organizations: a material approach. Ephemera: Theory & Politics in Organization 23: 217–230.

107.

Venema

(2023) De mythe van het dilemma tussen rechtszekerheid en rechtvaardigheid in de reflectie van de Afdeling Bestuursrechtspraak van de Raad van State op het kinderopvangtoeslagschandaal. Tijdschrift voor bestuurswetenschappen en publiekrecht 2023: 34–45.

108.

Verburgt

(2020) The history of knowledge and the future history of ignorance. KNOW: A Journal on the Formation of Knowledge 4: 1–24.

109.

Walton

(1996) Plausible deniability and evasion of burden of proof. Argumentation 10: 47–58.

110.

Whiteford

(2020) Robodebt was a fiasco with a cost we have yet to fully appreciate. The Conversation, 16 November. Available at: https://theconversation.com/robodebt-was-a-fiasco-with-a-cost-we-have-yet-to-fully-appreciate-150169

111.

Whiteford

(2021) Debt by design: the anatomy of a social policy fiasco – or was it something worse? Australian Journal of Public Administration 80: 340–360.

112.

Zerubavel

(2006) The Elephant in the Room. Silence and Denial in Everyday Life. Oxford: Oxford University Press.

113.

Zwierlein

(2016) The Dark Side of Knowledge: Histories of Ignorance, 1400 to 1800. Leiden: Brill.

114.

Deanna Amato v Commonwealth of Australia (2019) VID611/2019 634. https://www.comcourts.gov.au/file/Federal/P/VID611/2019/3859485/event/30114114/document/1513665

115.

Masterton Madeleine v Secretary, DHS (2019) VID73/2019 634.

116.

Prygodicz v Commonwealth of Australia (No 2) (2021) FCA 634. https://www.judgments.fedcourt.gov.au/judgments/Judgments/fca/single/2021/2021fca0634

117.

Royal Commission Documents (RC Docs). https://robodebt.royalcommission.gov.au/

118.

RC Doc 1: Exhibit 2–2685 – SMO.9999.0001.0105 – Media Release of Senator Richardson dated 21 August 1990.

119.

RC Doc 2: Exhibit 4–5107 – CTH.3007.0004.3602_R – FW- debts: Glyn Fiveash email to Tracy Tozer, 11 January 2017.

120.

RC Doc 3: Exhibit 1-0850;Exhibit 1–1207 – CTH.3716.0001.0001_R – NM signed brief Concepts for future compliance activity. June 2014 Minute.

121.

RC Doc 4: Exhibit 3–4321 – RBD.9999.0001.0369: Peter Martin, Sydney Morning Herald, 7 January 2017, ‘How Centrelink unleashed a weapon of math destruction’.

The Organization of Ignorance: The Australian ‘Robodebt’ Affair,Bureaucracy,Law and Politics

Abstract

Keywords

Introduction

What Was Robodebt? ‘Mountains of gold’

But Was It Legal?

Organized Ignorance

The Force of Law: The Royal Commission into the Robodebt Scheme

Conclusion: It’s Not a Bug, It’s a Feature

Footnotes

Funding

ORCID iD

Notes

References