From political self-deception to self-deception in political theory

In Political Self-Deception, Galeotti carves out valuable space for the analysis of behaviour on the part of political leaders that lies between straightforward deception and honest mistakes. In these comments I consider whether the concept of self-deception can travel from the political to the academic arena, to illuminate problems in how political theorists treat empirical data in the course of their normative work. Drawing on examples from the literature on the social bases of selfrespect, I show that political theorists too are vulnerable to the motivationally biased treatment of data. I suggest that this problem can helpfully be located on the same broad terrain Galeotti outlines, between lying and mistakes. I also identify some potential analogues, for the academic sphere, of Galeotti's proposed remedies for political SD. The paper goes on to reflect on how Galeotti herself employs empirical evidence in developing her account of self-deception. In particular, I challenge the empirical basis of her assumption that political self-deception is significantly more predictable, and therefore preventable, than political lying. My discussion seeks to show that, in addition to its intended contribution to the study of political deception, Political Self-Deception offers a valuable perspective on recent debates about the place of empirical evidence in political theory. However, approaching the book from this methodological angle reveals, in turn, some weaknesses in the empirical foundations of one of Galeotti's own key normative claims. ARTICLE HISTORY Received 15 June 2020 Revised 8 September 2020 Accepted 12 October 2020


Introduction
In Political Self-Deception (PSD), Galeotti carves out valuable space for the analysis of behaviour on the part of political leaders that lies between straightforward deception and honest mistakes. In particular, she argues that three major cases of US foreign policy failure are best understood as instances of political self-deception (SD): as cases of decision-makers believing that P, under the influence of the desire that P be the case, but against the evidence (Galeotti 2018, 19). 1 In these comments I consider whether the concept of SD can travel from the political to the academic arena, to illuminate problems in how political theorists treat empirical data in the course of their normative work. Drawing on examples from the literature on the social bases of self-respect, I show that political theorists too are vulnerable to the motivationally biased treatment of data. I identify some respects in which motivated empirical reasoning in political theory typically differs from political SD, on Galeotti's specific understanding of that concept. However, I suggest that the problem can helpfully be located on the same broad terrain she outlines, between lying and mistakes. I also suggest some potential analogues, for the academic sphere, of Galeotti's proposed remedies for political SD.
The paper goes on to reflect on how Galeotti herself employs empirical evidence in developing her account of SD. In particular, I challenge the empirical basis of her assumption that political SD is significantly more predictable, and therefore preventable, than political lying. My discussion seeks to show that, in addition to its intended contribution to the study of political deception, PSD offers a valuable perspective on recent debates about the place of empirical evidence in political theory. However, approaching the book from this methodological angle reveals, in turn, some weaknesses in the empirical foundations of one of Galeotti's own key normative claims.

Motivated empirical reasoning in political theory: applying the self-deception lens
At the heart of PSD is the application of the notion of individual SD to the political sphere, to generate new insights into the behaviour of political leaders and their relationship with democratic citizens. The book analyses three episodes in US foreign policy, in which political leaders misled the public in the course of backing, initiating or escalating military action. Rather than simply lying about the available evidence, or making honest mistakes, Galeotti argues that the deception of the public on the part of key political actors was, at least in part, the result of leaders' self-deception. What we see in these events, she argues, is the political analogue of a relatively widespread phenomenon in everyday life: 'the distortion of reality against the available evidence and according to one's wishes' (1).
Galeotti identifies four advantages -two explanatory and two normative -of adopting SD as a distinct analytical category in the study of democratic politics. First, it highlights the motivated and self-serving nature of the beliefs at work; something that is lost if we view these decisions as the product of straight mistakes on the part of political leaders. Second, unlike an account that reduces the behaviour to pure lies, an SD analysis also has a built-in explanation of political failure. If the deception of the public is built on distorted processing of data by leaders, it is unsurprising that poor decisions so often result. Third, the SD account allows us to hold political leaders properly responsible for their deception of the public. 2 Finally, Galeotti argues that political SD is at least somewhat predictable ex ante, given the characteristic conditions under which it arises. Thus being alert to the phenomenon of SD, amongst the lies and mistakes of political leaders, opens up possibilities of prevention that would otherwise be missed.
My aim here is to consider whether, despite the contextual differences, the SD lens might also have payoffs for thinking about the role of empirical evidence in political theory. My discussion starts from the broad observation that theorists often tell stories about the likely empirical preconditions for the realization of our normative ideals. In particular, we often construct pictures of how morally significant aspects of citizens' self-concept, or their other-regarding attitudes and behaviours, will respond to particular social and economic conditions. 3 For example, some relational egalitarians claim that distributive inequalities undermine our capacity to regard and relate to one another as equals (O'Neill 2010, 403). Others maintain that only when distributive inequality leaves some without enough will equal relationships be corroded (Anderson 1999, 326). These social psychological accounts are sometimes explicit, sometimes more implicit, but they often rest on shaky empirical foundations. 4 The tendency for political theorists to engage in speculative social psychology is well illustrated in the literature on the social bases of self-respect. Following Rawls, many liberal theorists have maintained that self-respect is a fundamental value; where selfrespect means having a secure sense of our own worth and the worth of our plans, and feeling confident in our ability to hold ourselves to our standards and to pursue our plans (Rawls 1999, 386). Since self-respect is of such importance, it is argued, we should seek to secure conditions conducive to its development and maintenance among citizens. However, when we turn to the philosophical literature on the social bases of self-respect, we find a wide variety of answers to the question of which social, political and economic arrangements offer citizens opportunities for self-respect. For example, consider the following accounts of how economic conditions shape our access to selfrespect: The basis of self-respect is not 'one's income share but the publicly affirmed distribution of fundamental rights and liberties' (Rawls 1999, 477) 'Only by making sure that the structure of the economy is such as to broadly disperse control over productive resources […] can we ensure that all citizens are able to have [self-respect]' (O'Neill 2012, 89) 'the experience of [economic] risk seems to be an essential precondition of the sort of selfrespect that liberals value' (Tomasi 2012, 80) Rawls famously stresses the role of the equal basic liberties in upholding self-respect and deemphasizes the importance of relative economic position, including such income inequality as would be permitted under the difference principle. When we turn to more recent Rawlsian literature on property owning democracy, we find greater focus on how the organization of economic life shapes opportunities for self-respect. For example, O'Neill suggests that widespread distribution of control over productive assets underpins self-respect. In a very different view of how the shape of the economy affects selfrespect, John Tomasi claims that economic risk is an essential precondition of selfrespect. Therefore, he argues, welfare and labour market policies that protect people against economic insecurity serve to undermine self-respect.
Self-respect is a complex and contested concept. Thus debate about its social bases will sometimes be rooted in disagreement at the conceptual level. However, what is striking about the passages cited above is that the authors offer divergent pictures of the conditions under which individuals enjoy opportunities for self-respect, despite the fact that they broadly agree on the Rawlsian account of what self-respect amounts to. Thus the disagreement we see here is best understood as empirical in nature. 5 The significant empirical component of disputes over the social bases of self-respect is often unacknowledged; or, if recognized, is typically not pursued. For example, Rawls's claim that self-respect should, within the terms of a Rawlsian society, be robust against economic inequality has been the subject of much criticism (for example Miller 1978, 18;Penny 2013;Zaino 1998). Yet little in the way of countervailing empirical evidence has been brought forth to support these critiques. 6 How does Galeotti's SD lens illuminate these dynamics in the literature on the social bases of self-respect? There is reason to think that, as in Galeotti's foreign policy cases, there is a form of motivationally biased empirical reasoning at work. In other words, the problem is not simply that theorists are too quick to speculate about the likely empirical preconditions for individual self-respect. Rather, they appear to tailor their accounts of the social bases of self-respect to fit the ideals to which they are already committed. If, like Tomasi, we prefer a system in which economic risk is individualized, we claim that bearing risk promotes individual self-respect. If we are drawn to a regime of property owning democracy that disperses control over productive assets, we claim that control over productive assets supports self-respect. In Galeotti's terms, theorists seem to be in the 'grip of a wish' that the economic or political conditions they independently believe to be desirable will also give rise to self-respect among citizens.
A large body of research into motivated reasoning suggests that our empirical judgements are driven by both 'accuracy goals' and 'directional goals'. Whilst we do not simply choose to believe whatever we want, our factual judgements involve 'a compromise between what the individual wishes were true and what can plausibly be believed based on the available data' (Ditto and Liu 2016, 105). In the moral domain, this results in a tendency to 'harness factual beliefs to support moral commitments' (Liu and Ditto 2012, 321). For example, individuals who are exposed to nonconsequentialist arguments against capital punishment judge capital punishment to 5 We might be tempted to defuse the empirical challenge by treating the proposed social bases of self-respect (for example, exposure to economic risk or control over productive assets) as partly constitutive of selfrespect. However, we must resist such moves if we seek, as these theorists do, to establish the justice of our proposed economic or political arrangements in part in reference to their capacity to uphold individual selfrespect. This justificatory move becomes unavailable, on pain of circularity, once we build the relevant conditions into our conception of self-respect. See Baderin and Barnes 2018, 7. 6 Interestingly Galeotti too makes passing reference to self-respect, claiming that SD has a negative long-run effect on the self-respect of self-deceivers (58, 70). Again, no empirical evidence is brought forward to support this claim.
be more immoral. But they also weaken their belief in its deterrent effects (Liu and Ditto 2012, 320). One underlying explanation for these, and similar, findings is that humans have a fundamental motivation to seek moral coherence. Research suggests that we often work from both directions to achieve this coherence: we 'reverse engineer' our purportedly factual beliefs about outcomes to fit our moral convictions, as well as adjusting our moral commitments in response to the facts (Clark, Chen, and Ditto 2015, 123). The pattern of reasoning I have identified in the literature on the social bases of selfrespect can usefully be viewed in the context of these broader psychological frameworks. What we see exemplified among philosophers of self-respect is a tendency to 'factualize' (Ditto and Liu 2016, 103) our normative ideals; to believe that good things will also bring good consequences. My aim here is to identify a pattern in the theoretical literature that is suggestive of motivational bias, rather than to firmly establish that such bias is present in the case of individual theorists. As Galeotti notes in relation to SD, it is very difficult to make such a diagnosis in the absence of direct access to an individual's psychological processes (82-83). However, when we observe a pattern in which theorists' empirical claims differ, in ways that systematically favour their broader normative convictions, motivated reasoning is a plausible hypothesis. Reliance on alternative bodies of evidence available at different times, for example, would not explain this pattern. Moreover, political theorists exhibit characteristics that, in other contexts, have been found to exacerbate the tendency towards the motivated treatment of data, namely holding stronger convictions and having greater knowledge in a given domain (for example, Druckman 2012, 202).

Is it SD proper?
I have argued that there is a pattern of empirical reasoning in political theory that is usefully understood to occupy the same broad space Galeotti identifies in her analysis of foreign policymaking, between lying and mistakes. However, a significant part of the work of PSD is devoted to carving out SD as a specific problem, distinct from broader phenomena of wishful believing, ideological bias or stubborn thinking. In particular, Galeotti emphasizes that SD proper involves an emotional reaction to an encounter with negative evidence relative to some wish: 'In SD, S is not simply motivated to believe that P, but is motivated to believe that P under the threat of contrary evidence which makes his wish that P emotionally loaded with fear and anxiety' (48, emphasis added). Thus the self-deceived individual encounters, and emotionally reacts to, evidence contrary to their wish, resulting in the formation of a clearly counter-evidential belief. This analysis points to a gap between the notion of SD and the problem of motivated empirical reasoning I have identified in work on the social bases of selfrespect, since much of this literature has simply floated free of relevant empirical evidence. In Galeotti's terms, philosophers of self-respect are engaged in 'wishful thinking' or 'believing beyond the evidence', rather than 'believing against the evidence' (32). In other words, their empirical claims appear to be directly induced by their normative wishes, rather than formed via an emotionally loaded appraisal of contrary evidence.
Even when political theorists do actively seek out empirical evidence pertinent to our normative concerns, the data we find will often be limited or inconclusive. This is both because empirical research is typically not designed to test the precise claims of interest to normative theorists, and because the social and political phenomena that concern us are often highly complex. 7 In such cases we have two options: We can acknowledge the inconclusive nature of the evidence and make explicit where our normative arguments are conditional on, as yet unproven, empirical claims. Alternatively, we can design our own empirical studies, tailored to generate data that speaks directly to our particular normative concerns. 8 It is worth pausing here to reflect on why such interdisciplinary efforts are not more commonplace. Probably much of the explanation lies in the inherent challenges of doing this kind of work. However, we also see, among some political theorists, more principled opposition to bridging the normative-empirical divide. For example, complex philosophical ideals such as self-respect are viewed as inherently resistant to measurement: 'Consider the case of a relational good that is difficult, if not impossible, to measure: self-respect. Now, even if we cannot measure how much self-respect a person has, we can tell whether she has access to the social bases of self-respect, and it is these social bases that ought to count as distribuenda' (Cordelli 2015, 100). This seems to me to get things the wrong way round. We cannot determine the general social bases of self-respect, and judge whether individuals have access to those conditions, if we cannot measure self-respect and how it behaves in varying social and economic contexts. In the absence of such evidence, our accounts of the social bases of self-respect must remain speculative, and susceptible to the tendency towards motivated empirical reasoning.
The discussion thus far has taken the literature on the social bases of self-respect as an example, in order to ground an exploration of SD in political theory. I have noted that theorists' lack of engagement with empirical evidence on the sources of self-respect points to a diagnosis of wishful thinking, rather than SD, according to Galeotti's account. However, there are other disciplinary subfields where the links between normative and empirical inquiry are better developed and there is therefore an emerging risk of SD proper. Most notably, theorists of deliberative democracy have engaged in a sustained way with data about real-world deliberative practice, and empirical researchers have taken their agendas from deliberative democratic theory (for overviews, see Ryfe 2005;Thompson 2008). Thus, where empirical evidence is largely missing in debate over the social bases of self-respect, we see the interpretation of data becoming a site of contestation in normative theorizing about deliberative democracy. For example, a significant part of the critical reaction to Brennan's recent book Against Democracy (2017) has focussed on his treatment of data on the effects of deliberation. In particular, some reviewers have objected that he 'cherrypicks' findings that support his case, and ignores opposing evidence (Chambers 2018, 504; see also Christiano 2017). 9 Interestingly, Brennan's critics suggest that his empirical account is motivated not (only) by a particular substantive 7 Thus if SD proper involves the formation of clearly counter-evidential beliefs, it may be fairly unusual in political theory.
However, we should not overstate the contrast with the foreign policy domain here. Political decision-makers too must typically deal with 'blurred data' (48), and this fact, Galeotti acknowledges, limits the role of SD in politics. 8 For an example of this approach, utilizing survey data in relation to philosophical debate about the social bases of selfrespect, see Baderin and Barnes 2018. 9 My intention here is not to evaluate these critiques of Brennan, but rather to suggest that they can usefully be viewed through Galeotti's SD framework.
normative agenda, but also by a desire to carve out a controversial position within democratic theory. This observation, I think, points to an important broader point about the nature of SD in political theory. Much contemporary philosophical inquiry involves a drive to uncover and to foreground disagreement (Wolff 2011, 3, 193-4), and we should be alert to a tendency to recruit empirical claims to this cause; as well as to the risk that we tailor empirical beliefs to fit our substantive normative convictions. To apply Galeotti's language, my suggestion is that there may be a distinctively philosophical form of 'anxious wish' (92) for disagreement that can bias our treatment of empirical data.

What can be done?
Whether we are dealing with wishful thinking, or with SD proper, how should we guard against motivated empirical reasoning in political theory? Here we can usefully reflect on Galeotti's proposals for combatting political SD (108-113, 245, 246). Galeotti argues that SD may arise when the costs of inaccuracy in data processing are low, or can be discounted (88). Thus one important potential response to the problem of SD is to increase the perceived costs of empirical inaccuracy. It is suggested that including information accuracy as part of 'moral training' (113) for political leaders might play some role here. However, greater emphasis is placed on the potential for new institutional measures to function as 'precommitment' devices against SD. We need, Galeotti argues, institutional devices to disrupt the 'groupthink' dynamics of collective political decision-making that, under specific contexts, can lead to SD (110). In particular, we need independent referees who do not share the anxious wish that drives the SD. Galeotti suggests that this role might be performed by citizens selected by lottery, and empowered to oversee the treatment of empirical evidence in key foreign policy decisions (112). Each of Galeotti's proposed 'prophylactic measures' (76) against political SD has potential analogues in the academic sphere. First, training in the identification and assessment of empirical claims, and discussion of the value of doing so, could be given a more prominent role in graduate political theory programmes. Second, more interdisciplinary refereeing practices could be employed as an incentive, and a check, against motivated empirical reasoning. For example, journals could commit to sending a random sample of political theory papers for empirical review, by specialists charged with identifying and challenging the factual presuppositions of our normative theories. 10

The empirical underpinnings of Political Self-Deception
Political theorists are increasingly engaged with questions about the place of empirical evidence in our discipline: When it comes to inquiry about how we ought to organize our political lives, what kind of role should we give to data about how the political world actually is? As well as a framework that illuminates problems in how we treat (or fail to treat) empirical evidence in political theory, PSD also offers a positive alternative approach.
Specifically, Galeotti's account begins with psychological evidence of the common experience of believing against the evidence, in the grip of some desire. She then goes on to show how this phenomenon also applies to the behaviour of political leaders, with potentially significant normative implications for democratic politics. Whilst affirming this overall approach, in which normative theorizing is grounded in a rich social psychology, I now want to challenge the empirical basis of one key move in Galeotti's argument.

Predictability and preventability: the contrast between political self-deception and political lying
One of the most normatively important ideas in PSD is the claim that political SD, unlike straight lying or honest mistakes, is ex ante predictable, and therefore potentially preventable. This contrast is emphasized at various points in the book 11 : 'Deception and mistakes, being ubiquitous, are normally detected by hindsight and can hardly be predicted and prevented. So far, no general circumstances for telling a lie or for making a mistake have been provided, and I think this is not by chance. By contrast, SD is in principle open to prevention for SD is triggered in specific circumstances ' (70-71).
'contrary to lies and mistakes, SD can be foreseen, when the circumstances likely to give rise to the process obtains' (243).
Specifically, political SD typically arises in the context of high stakes, time-pressured decisions taken in response to external events. These pressures, Galeotti argues, create emotional pressure, which in turn distorts political actors' processing of data. Paradigm cases involve decisions about international military intervention or reprisals for attacks: 'when dramatic threats arise, when something so important and apparently beyond rational control happens so as to trigger great anxiety' (83). 12 'momentous decisions with far-reaching consequences, made under the pressure of time and in difficult circumstances' (246).
Under these circumstances, if a decision-maker encounters evidence contrary to their wish, and they perceive the costs of empirical inaccuracy to be low, then SD may well be the result.
Whilst Galeotti acknowledges that these 'favourable' conditions are not sufficient for political SD to occur, here I want to question whether they are plausibly seen as necessary, or perhaps even characteristic, for SD in politics. First, the pressure of making a decision under a limited time frame posed by external events is not central to paradigm personal examples of SD, such as believing in my good health against the evidence, or believing in my partner's faithfulness despite evidence to the contrary. Thus we need a clearer explanation of why these pressures are necessary contextual conditions for SD in politics. 13 It is also unclear how the emphasis on these contextual features fits with Galeotti's own discussion of ex-post political SD, in which political 11 See also 4-5, 91, 109. This contrast is also prominent in Galeotti 2015, 888. 12 Anxiety is central to Galeotti's account of how, and why, political SD occurs. Interestingly, at the level of public opinion, there is some evidence that increased anxiety can reduce motivated political reasoning. See Atkeson and Maestas 2012. 13 Another challenge here is that time pressure is also more likely to provoke honest errors. This makes it more difficult to pick out cases of genuine SD from honest mistakes in situations of crisis. actors redescribe their goals or priorities in order to avoid acknowledging previous failures (94-95). Here we seem to have a case of genuine political SD, absent the pressure of responding to externally driven events under time constraints. Whilst Galeotti's analysis privileges the foreign policy context, she notes that SD may also occur in domestic policy-making, given 'similar circumstances of time pressure and a clear perception of it being a momentous, exceptional decision' (14). But again, it is insufficiently clear how this restriction on the scope of the account is justified. Why should we not expect SD to arise within more extended policy-making in the domestic sphere? Consider, for example, the ongoing Universal Credit (UC) programme of benefit reform in the UK. The UC programme has been the subject of a series of negative news stories since its inception in 2010, with a significant amount of criticism centring on how the responsible team within the Department for Work and Pensions has reacted to data challenging the programme. For example, a recent Public Accounts Committee report concluded that 'The Department's systematic culture of denial and defensiveness in the face of any adverse evidence presented by others is a significant risk to the programme' (House of Commons Committee of Public Accounts 2018, 5). 14 Critical reports about UC resonate closely with the descriptions in PSD of foreign policy-makers' reactions to negative evidence about their plans. In each case, we see political actors determined to press on and see things in a positive light, despite accumulating negative evidence.
The point here is not to reach any firm conclusion about the dynamics of UC policymaking. More tentatively, I want to suggest that, if we are interested in offering a 'speculative' reconstruction of events (14), it seems plausible that SD is playing some role here, as it is in the foreign policy crises analysed in PSD. Crucially, though, the key contextual conditions identified by Galeotti -time pressure and the need to react to dramatic external events -are absent. 15 Decision-makers are likely operating under other significant pressures, including ever increasing weight of sunk costs and reputational investment.
Galeotti argues that political SD is relatively predictable, since it arises in a specific kind of decision-making context. In response, I have suggested that it is insufficiently clear why we should expect SD to be confined to these circumstances. However, if we broaden the scope of the contextual conditions for political SD, the claim of ex ante predictability (and associated contrast with political lying) is weakened. There is an underlying tension here between the picture of political SD as a response to a highly unusual set of circumstances 'of exceptional political threat' (112), which underpins the claim of SD's foreseeability; and the acknowledgement that SD is a 'widespread and common phenomenon in daily life' (80).
We can also put pressure on the distinction between SD and lying from the other direction, by questioning the picture of lying as wholly unpredictable ex ante, and detectable only if it goes wrong. There are several bodies of empirical research that potentially cast light on when political lying is most likely to occur, how we might spot it, and what we might do to reduce it. For example, psychological research into the characteristic speech patterns of liars is now being applied to political speech (Bond et al. 2019). Outside of the political sphere, there is a growing body of work in behavioural economics and experimental psychology exploring when, why and how much we lie (for example, Abeler et al. 2014;Cappelen, Sørensen, and Tungodden 2013). A recent overview of this literature identifies a range of non-economic conditions under which lying tends to increase, including: with tiredness and cognitive depletion; when the payoff is closer to the decision-point; with awareness of future altruistic opportunities; with a feeling of anonymity at the decision-point; when seeking to cover a loss, rather than to make a gain; and when people with whom you identify are lying (Jacobsen, Fosgaard, and Pascual-Ezama 2018).
The findings of this research are complex, with important remaining areas of uncertainty (Gerlach and Teodorescu 2019). For example, there are plausible theoretical stories linking reliance on intuition with both increased and decreased lying; and empirical studies have produced mixed results (Köbis et al. 2019). There is also much work to do to explore how far these insights, which are typically derived from experiments involving individual lies for financial gain, can be applied to the realm of political decision-making, with its more complex group decision dynamics and payoffs. However, the broad direction of work in behavioural economics and psychology is to show that conditions that are conducive to lying are at least somewhat predictable and might also be open to prevention. Thus Galeotti's claim that 'So far, no general circumstances for telling a lie … have been provided' (70-71) is too quick, at least in the absence of critical engagement with the extensive body of empirical work that is seeking to establish such conditions. My general conclusion here is that the line that is drawn in PSD between lying as ex ante unpredictable, and SD as relatively predictable, is too sharp. This is both because of some uncertainty in the scope of the contextual conditions for political SD, and because Galeotti is too quick to dismiss political lying as wholly unpredictable. This challenge is significant, because it threatens to undermine one of the key normative advantages claimed for the SD lens: that by being alert to instances of SD amongst the political lies and mistakes, we open up otherwise unavailable opportunities for preventing the associated public deception and policy-making failures. We might wonder if Galeotti herself is falling prey to a form of wishful thinking here: perhaps she is too quick to commit to an empirical account that supports her drive to pick out SD as a distinctive, and normatively significant, phenomenon.

Conclusion
This paper has used Galeotti's concept of political SD as a tool to reflect on the treatment of empirical evidence in political theory. I have argued that, whilst the academic realm is far from the high stakes foreign policy decisions that are the focus of PSD, Galeotti's broader agenda is illuminating. In drawing our attention to the treatment of evidence that sits between lying and honest mistakes, the SD lens highlights the risk of advancing motivated empirical stories in political theory. I suggested that motivated empirical reasoning can be driven both by our particular normative convictions, and by a more general philosophical 'anxious wish' to establish disagreement. The remedies Galeotti outlines for political SD also have potential counterparts in the academic arena. I went on to critically reflect on the empirical underpinnings of a central claim in Galeotti's own argument: that the category of political SD has 'normative payoffs' (17), because it opens up possibilities of ex ante prediction and prevention that do not exist in cases of political lying. A brief consideration of the empirical evidence, I suggested, does not support such a sharp contrast between political SD and political lying.

Disclosure statement
No potential conflict of interest was reported by the author.