DOUBT AS A POLITICAL VIRTUE

DOUBT AS A POLITICAL VIRTUE Quassim Cassam University of Warwick, UK Abstract: This paper explicates the notion of doubt and the relationship of doubt to belief and conviction. It distinguishes three types of political virtue – leadership, systemic, and corrective – and considers whether doubt is a political virtue in any of these three senses. It is argued that while doubt is not a leadership virtue, it is a systemic and a corrective virtue. Specifically, it is potentially an antidote to methods, ideological, and psychological extremism. A distinction is drawn between extremism and forms of radicalism that have resulted in social progress. It is possible for doubt to play a role in countering extremism without thereby also countering progressive radicalism. The concluding section develops a theory of deradicalization and identifies the role of radical doubt in deradicalization programmes. The proposed empirically informed account of deradicalization highlights the role of narratives in radicalization and deradicalization. 1 In an account of the terrorist attacks on London on 7 July 2005, John O’Farrell reflects on the four men who set out that day to kill as many people as possible. They demonstrated that ‘absolute certainty is the most dangerous mind-set in politics. They were so very wrong and yet must have been so very sure they were right. What a precious political virtue is doubt. How civilized doubt is’ (2017: 133). The 7/7 bombers did not heed Russell’s advice ‘always to entertain our opinions with some measure of doubt’. Russell said this in a television interview. He continued: ‘I shouldn’t wish people dogmatically to believe any philosophy, not even mine’. If they had taken this to heart, they would surely not have acted as they did. One sense in which doubt is a political virtue is that it is an antidote to the excessive certainty and fanaticism to which at least some political actors are prone. Indeed, an unwillingness to entertain doubts about their policies, actions and ideals is often represented as one of the main distinguishing intellectual characteristics of fanatics. Fanatics are not the only political actors who are criticized for lacking doubt. In the third volume of his Bush at War series, Bob Woodward reports a conversation about the issue of doubt with President George W. Bush. He reports the President as saying: ‘“I know it is hard for you to believe, but I have not doubted what we’re doing…. There is no doubt in my mind we’re doing the right thing”’ (Woodward 2006: 325-6). In a later interview with Woodward, Bush expanded on this theme: “First of all”, he said, “a president has got to have calcium in his backbone. If I weaken, the whole team weakens. If I’m doubtful, I can assure you that there will be a lot of doubt. If my confidence level in our ability declines, it will send ripples throughout the whole organization. I mean, it’s essential that we be confident and determined and united. I don’t need people around me who are not steady” (2006: 326). Woodward contrasts Bush’s attitude with that of his Secretary of State, Condoleezza Rice, who maintained that ‘doubt was an essential ingredient in decision making because it forces careful consideration and readjustment’ (2006: 326). If the President had been more willing to engage in careful consideration and readjustment, he might have had second thoughts about his decision to invade Iraq in 2003. On the other hand, there are obvious difficulties with the notion that doubt is a political or, indeed, an intellectual virtue. 19th century abolitionists like William Lloyd Garrison did not doubt that they were doing the right thing in campaigning as they did against slavery, and their lack of doubt was neither a political nor an intellectual failing. See Kraditor 1989 and Sinha 2016. They did not doubt because there was no doubt. Furthermore, doubts about their cause might have made their campaign less effective. It is also misleading to talk about a refusal or unwillingness to doubt since one can no more doubt at will than believe at will. As Descartes notes, ‘before we can decide to doubt, we need some reason for doubting’. Descartes says this in the Appendix to Fifth Objections and Replies. See Descartes 2017: 75. In the absence of such reasons, it was not open to the abolitionists to doubt their views about slavery. There is also something paradoxical about Russell’s advice to entertain our opinions with a measure of doubt. Suppose that our opinions are our beliefs. How is it possible for one genuinely to believe P, while also doubting that P? Various answers to this question have been proposed. However, the issues are murky, not least because it is unclear what is involved in believing or doubting something. Furthermore, the doubts that are at issue in the present context do not pertain, or only pertain, to a person’s beliefs. As well as beliefs, political actors have convictions, commitments, principles, strategies, tactics, policies, and ideals. Even if belief and doubt turn out to be compatible, it is a further question whether, for example, a person who doubts that P can be said to have the conviction that P. For so-called ‘conviction politicians’, doubt is an anathema because it amounts to not having convictions, or only having weak convictions, and that is as bad as having no principles or only weak principles. Aside from conceptual questions about the impact of doubt on one’s beliefs, there are also practical considerations. President Bush’s point in his conversations with Woodward was that doubts diminish the practical effectiveness of political leaders. Whether this is so or not is an empirical question. On the face of it, Bush’s answer to this question is plausible, despite the unfortunate consequences of his own lack of doubt. One of his predecessors in the White House illustrates the negative impact of doubt on leadership. President Lyndon B. Johnson is someone whose chronic insecurity and self-doubt have been well-documented. McMaster 1997. Far from prompting careful consideration and readjustment of his Vietnam policy, his doubts resulted in a constant need for reassurance and an over-reliance on advisors who were prepared to tell him what he wanted to hear. One might have thought that his doubts would make him more willing to listen to views that were different from his own, but they had the opposite effect and played a significant role in shaping the way that the United States became more deeply involved in the Vietnam war. As McMaster puts it, ‘Johnson’s lack of self-confidence manifested itself in a reluctance to trust those around him’ (1997: 50), and his ‘self-doubt and willingness to forgo the truth would color his relationship with his principal military advisers and shape the way that the United States became more deeply involved in the Vietnam war’ (1997: 51). Whether it follows from this that having calcium in one’s backbone is a political virtue remains to be seen since the notion of a political virtue has yet to be explained. The following discussion of these complex and important issues will be in three parts. The next section will explicate the notion of doubt and the relationship of doubt to belief and conviction. Whether or not it is possible to doubt one’s beliefs while retaining those beliefs, it is possible to doubt one’s own ongoing policies or tactics, or doubt whether one is doing the right thing, even if one is in fact doing the right thing. The question whether such doubts are politically virtuous is a real one, which calls for greater clarity about the idea of a political virtue. In giving an account of this notion, it needs to be acknowledged that there are different types of political actor and that the political virtues of one type of political actor need not be the political virtues of another. This issue will be taken up below. Section 3 will make the case that doubt is a corrective political virtue. One view is that all virtues are corrective, with ‘each one standing at a point at which there is some temptation to be resisted or deficiency of motivation to be made good’ (Foot 1978: 8). The temptations for which doubt is a corrective include the temptations of extremism. To think of doubt in this way is not to suppose that extremism is something to which every human being or every political actor is prone. However, the prevalence of political extremism in one form or another since at least the 1930s suggests that it is a common enough vice to merit its very own corrective virtue. At the same time, President Bush’s remarks about doubt also need to be taken on board. A way to do this is to distinguish between different types of political virtue and acknowledge the sense in which political virtues are both role relative and context relative. For the idea that some virtues are role relative and context relative, see Pigden 2017: 125. Pigden is specifically concerned with epistemic virtues, but the idea generalizes. He also describes epistemic virtues as ‘end relative’. What counts as a virtue depends on the ends you have in view. See Nussbaum 1993 for further discussion of the sense in which Aristotelian virtues are or are not ‘relative’. As far as political leaders are concerned, the virtue they need to exercise in their decision making is not doubt but critical reasoning. Section 4 will address consider the relevance of doubt for deradicalization, the process by which extremists comes to abandon or tone down their extremism. If doubt is an antidote to extremism, then the cultivation of this corrective political virtue in extremists should help to deradicalize them. If extremists like the 7/7 bombers can be brought to question their methods or convictions, this might have a salutary effect. One might think that the very fact that they are extremists makes them immune to doubt. However, even extremists sometimes change their minds and question what previously struck them as indubitable. Deradicalization is possible. The practical challenge, therefore, is to identify ways of utilizing the power of doubt against extremism while also acknowledging the ways in which doubt can be politically problematic. 2 On one view of doubt, ‘to doubt something is to disbelieve it, or alternatively, simply to fail to believe it’ (Salmon 1995: 2). Now consider a subject S who has never considered the question whether P. Suppose that S is King Henry VIII and P is the proposition that Macs are easier to use that PCs. Henry failed to believe that P, not least because he lacked the concepts necessary to understand it. However, it would be peculiar to say that Henry VIII doubted that Macs are easier to use than PCs. To doubt that P one must at least have considered whether P, or thought of P. In Williams’ account of Descartes, it is axiomatic that ‘If A doubts P, A thinks of P’. Another Cartesian axiom is: if A doubts P, A does not believe P. See Williams 1978: 292. In ordinary parlance, a person who doubts that Macs are easier than PCs has considered the matter and is inclined to disbelieve that Macs are easier, or regards it is unlikely that Macs are easier. Such a person need not rule out that possibility that Macs are easier but their degree of confidence that P is low. In contrast, suppose that S is one of our contemporaries who has investigated P and concluded, definitively, that P is false. S disbelieves P but would not ordinarily be described as someone who doubts P or doubts whether P. They have already decided the issue in their own mind. Someone in a state of doubt is undecided. Thus, Salmon’s account is doubly problematic. To be fair to Salmon, he concedes that his account of doubt ‘constitutes a departure from standard usage’ (1995: 1). There are degrees of belief because there are different degrees of confidence in a given proposition. In the same way, there are degrees of doubt. For example, one can be somewhat doubtful that P or extremely doubtful that P. Suppose that S’s degree of confidence that P is 0.7, and that this is sufficient for S to count as believing that P. However, since S is not certain that P, there is some doubt in S’s mind that P. In 2011 President Obama was briefed about the Pacer, a tall man observed walking in circles within the compound of an unusually spacious and secure property in Abbottabad, Pakistan. The CIA suspected that the Pacer was Osama bin Laden. In the words of one CIA briefer, “there’s a good chance he’s our man” but “we can’t be certain”. See Obama 2020: 679. In the end the President ordered a raid on the compound by US Navy SEALS, and it turned out that the Pacer was bin Laden. Obama describes the discussions that led to the raid in chapter 27 of Obama 2020. For further discussion of the President’s reasoning, see Kay and King 2020: 8-10. Prior to the raid, did Obama believe that the Pacer was bin Laden? Since he ordered the raid to go ahead, knowing its many potential hazards, it would be most surprising if he did not believe, on balance, that the Pacer was bin Laden or if he doubted that the Pacer was bin Laden. One can only imagine the public reaction if the raid had gone wrong and it transpired that the President had doubted all along that the Pacer was bin Laden or doubted whether the Pacer was bin Laden. However, it is perfectly understandable that he had doubts about the identity of the Pacer, given that the available intelligence did not settle the question one way or another. This reveals something about the logic of doubt. A person who believes that P cannot be described as doubting that P, but it is possible for a person to believe P while having or harboring doubts about P (as long as they are not filled with doubt). A person who doubts that P or is filled with doubts about P does not believe P. A person who has doubts about P might still believe that P but does not fully believe it, that is, is not completely confident that P. Going back to Russell’s advice, we can entertain our opinions with some measure of doubt if we do not fully believe our opinions or hold them with less than complete confidence. Whether this is good advice or not depends on the subject-matter. There are some matters concerning which a high level of confidence, amounting to certainty, is appropriate, and other matters concerning which it is not. Is it possible to doubt one’s convictions, principles, and ideals while still retaining those very convictions, principles, and ideals? As Michael Lynch points out, a conviction is not just a strongly held belief. It is not a conviction of mine that it is raining right now even though I firmly believe that it is raining. For Lynch, convictions are commitments to action that reflect our self-identity. The sense in which convictions have authority over our actions is that ‘they obligate us to do some things and grant us permissions to do others’ (2021: 14). For example, the religious convictions of the 7/7 bombers gave them permission, at least in their own eyes, to bomb London. Furthermore, ‘once something becomes a real conviction, it is difficult for us, from a psychological standpoint, to doubt. That’s because to doubt it would be to doubt our deepest commitments, to doubt that we are who we say we are’ (Lynch 2021: 140). That is why people feel guilty for not living up to their convictions. The issue here is not whether people can ever abandon or revise their convictions. In the 1950s, revelations about Stalin’s time in power caused many on the left to revise or abandon their Marxist convictions. The issue is whether a person can retain her convictions while also doubting them. This is not just a psychological question. A natural thought is that doubting a conviction amounts to no longer having it and therefore altering one’s own identity. On the other hand, if there are degrees of conviction, just as there are degrees of belief, then doubting a conviction only entails that the doubted conviction is less than full-blooded. It does not entail its abandonment. However, the question whether there are degrees of conviction has not been settled. What is true is that a person who can doubt or question her convictions is very different from one who cannot. Principles are a different matter. Suppose that S is someone for whom a commitment to freedom of speech is a matter of principle. When they confront the question whether freedom of speech should extend to extremists who are intent on fomenting racial hatred, S might start to experience doubts about the principle but still retain it in the following sense: they continue to speak in favour of it and act in accordance with it, even as their doubts are manifest in the reluctance with which they defend the free speech rights of racists. This also bears on the extent to which a person with doubts about an ideal can still be said to have that ideal. Ideals are ‘visions of the ends of life’ (Strawson 2008: 31). Different political ideals are different visions of the ends political life. In view of the plurality of such ideals, one might reasonably entertain mild doubts about one’s own political ideals without thereby abandoning them. The upshot of the discussion so far is that political actors can have doubts about their ideals, principles and possibly even their convictions without necessarily giving up them up. More straightforwardly, they can entertain doubts about their actions, strategies, tactics, and policies. Are such doubts desirable? More specifically, is it politically virtuous to have doubts about such matters? To get a handle on the idea of a political virtue, consider William Galston’s discussion of the question whether toughness is a political virtue. A virtue, for Galston, is a ‘disposition of mind and character’ (1991: 175). Intrinsic virtues are ‘active dispositions that constitute our good, excellence, or perfection qua human beings’ (1991: 176). Instrumental virtues are ‘dispositions that enable us to perform well the specific tasks presented by our situation’ (1991: 176). Every society contains a variety of essential roles without which ‘the political community could not hope to conduct its affairs and accomplish its objectives’ (1991: 178-9). Political leadership is one such essential role. The sense in which toughness is an instrumental political virtue is that is that it is one of the functional excellences of political leadership, that is, an excellence that political leaders need to have if they are to be effective and carry out their role. Toughness as Galston views it, stands between squeamishness and callousness. It is a virtue because ‘it allows the agent to contemplate the performance of intrinsically distasteful and objectionable acts, but only at the right time and in the right manner’ (1991: 182). Other putative political virtues, proposed by Max Weber, are passion, a feeling of responsibility, and a sense of proportion. Weber 1946: 115. Just as, for Galston, squeamishness and callousness are political vices so, for Weber, ‘lack of distance’ is ‘one of the deadly sins of every politician’ (1946: 115). A striking and helpful feature of Galston’s account of instrumental virtues is the way it brings out their role and context relativity. Not all political actors are, or aspire to be, political leaders. Toughness might be indispensable for political leaders, but other political actors can get on perfectly well without it. It is, in this sense, a role relative political virtue. Political virtues are also context relative. The functional excellences of leaders in one kind of political system – say a democracy – are unlikely to be identical with those of leaders in a different kind of political system. This is potentially embarrassing because one might be reluctant to describe the qualities required to be an effective dictator as virtues at all. The discussion so far has tied political virtues to effectiveness. Hence, the political virtues of political leaders are dispositions of mind and character that are required for them to be politically effective. However, it is also possible to think of political virtue in more systemic terms. This is the implication of Mark E. Button’s conception of political vices as ‘the kind of persistent dispositions of character and conduct that imperil both the functioning of democratic-political institutions and the trust that a diverse citizenry has in the ability of those institutions to secure a just political order’ (2016: 1). If these are political vices, then political virtues are dispositions of character and conduct that have the opposite effect. For example, when it comes to political leaders, a disposition to fulfil one’s commitments and be honest with the electorate is a trust-enhancing political virtue, just as a disposition to lie and bullshit one’s way through political difficulties is a trust-sapping political vice. Another way of thinking about political virtues is suggested by Philippa Foot’s account of virtues and vices. For present purposes, Foot’s key idea is that virtues are corrective, with ‘each one standing at a point at which there is some temptation to be resisted or deficiency of motivation to be made good’ (1978: 8). Thus, humility is a virtue ‘only because men tend to think too well of themselves’ (1978: 9). When the notion of a corrective political virtue is added to the other conceptions of political virtue that have been identified here, the question whether doubt is a political virtue can be understood in a range of different ways. There is the question whether doubt is a leadership virtue or whether, as President Bush implied, it is a leadership vice. There is the question whether doubt should be regarded as a systemic virtue in the context of liberal democracy. Finally, there is the issue of whether doubt is, in this context, a corrective political virtue. These are the questions that now need to be addressed. 3 The most compelling argument against the view that doubt is a leadership virtue is a version of Bush’s argument. He focused on the impact of doubt on his colleagues: if he doubted himself or his policies, they would do the same. This ripple effect matters because people cannot be expected effectively to promote and implement policies that emanate from a political leader who is doubtful about those policies. It might be said that doubts about Bush’s Iraq policy would have been justified but that is not the issue here. The fact that a particular policy is misguided is not an argument for the thesis that doubt is a political virtue since doubt is just as likely to hinder the implementation of sound policies as unsound ones. To draw attention, as Condoleezza Rice is reported to have done, to the importance of careful consideration and readjustment is not to make the case for doubt. It is simply to make the case for caution and due diligence. The remedy for a lack of due diligence is not doubt but due diligence. Due diligence depends on critical reasoning, and the Bush administration’s failures in its Iraq policy were failures of critical reasoning. According to Tyler Burge: Critical reasoning is reasoning that involves an ability to recognize and effectively employ reasonable criticism or support for reasons and reasoning. It is reasoning guided by an appreciation, use, and assessment of reasons and reasoning as such. As a critical reasoner, one not only reasons. One recognizes reasons as reasons. One evaluates, checks, weighs, criticizes, supplements one’s reasons and reasoning (1998: 246). It is arguable that Bush and his senior colleagues did not adequately check, weigh, or criticize their reasons for invading Iraq or for assuming that the invasion and its aftermath would be relatively painless for the United States. If they had properly evaluated their reasons, they may well have come to doubt the merits of their policy, but it is not doubt itself that is a leadership virtue. If it were a leadership virtue, then it would presumably be virtuous for political leaders to doubt policies that had been properly evaluated and found to be sound. It is a disposition to engage in critical reasoning about their strategies and policies that political leaders need if they are to be effective and avoid catastrophic errors. Doubt might be the result of critical reasoning but is not itself a leadership virtue. There is also the question of conviction. Democratic citizens have a taste for conviction politicians, and this is sometimes seen as regrettable. A conviction politician is one with strong political and ideological commitments and whose actions are driven by those commitments rather than by a desire for popularity and short-term electoral success. Conviction politicians are not prone to self-doubt, and this can be a leadership asset. It is a good question why voters admire conviction and the absence of doubt in their politicians. Rightly or wrongly, conviction is seen as a sign of authenticity and trustworthiness. Furthermore, voters cannot be expected believe in political leaders who do not believe in themselves. Equally, they cannot be expected to vote for policies whose proponents are doubtful about their merits. Doubts at the top send ripples throughout the electorate as well as the rest of government. Even if doubt is not a leadership virtue, could it still be a systemic political virtue? Suppose that systemic political virtues in the context of a democracy are dispositions of mind and character that abet the functioning of democratic-political institutions and the trust that a diverse citizenry has in the ability of those institutions to secure a just political order. On this conception of a political virtue, it might seem that doubt cannot be one. Democracy, it has been argued, requires trust. As O’Neill puts it, ‘trust is basic for human rights and democracy’ (2002: 27). Democratic citizens need to trust each other, their political leaders, and their political institutions. Doubt is the enemy of trust. To say that one doubts the motives or competence of one’s political leaders or institutions is to say that one does not trust them. If democracies depend on trust, then it follows that doubt is not a systemic political virtue. This argument is too quick in ways that leave the door open for doubt to be a systemic political virtue. First, it is controversial whether and to what extent democracies need trust. This is not the place for a discussion of this issue, but it is arguable that the trust on which democracy depends is compatible with a considerable degree of scepticism. The trust on which democracy depends is ‘trust with good judgement’ (O’Neill 2002: vii) or ‘discriminating trust’ (Warren 1999: 12). Discriminating trust in government or political leaders is grounded in a proper evaluation of their track record. Recent history does not support a high level of trust in government. Johnson lied about Vietnam, Nixon lied about Watergate, and Trump lied about any number of things. In the UK, the Leave side in the Brexit referendum lied about the costs and benefits of EU membership. Descartes was correct that in order doubt we need some reason for doubting, but we have reasons for doubting our political leaders. There is no question that widespread scepticism about our governments makes their jobs more difficult and disagreeable, but it is also our best hope of restraining their poor conduct. Doubt should be our default attitude if we wish to hold our leaders to account. To put it another way, doubt is a systemic political virtue in at least the following sense: it minimizes malfeasance by those in power. Politically virtuous systemic doubt needs to be distinguished from the hyperbolical doubt of conspiracy theorists who foment doubts about mainstream sources of information, including government, for their dubious political ends. On the political ends of conspiracy theories, see Cassam 2019. Doubt is only the start. Consumers of conspiracy theories are ultimately encouraged not merely to doubt but to disbelieve mainstream accounts of climate change, vaccine safety, and specific incidents like 9/11. Such doubts have proved immensely harmful and, in many cases, display poor judgement. The doubt that counts as a systemic political virtue is the mean between excessive trust in government and excessive mistrust. Doubt is the antidote to political naivety but the unwillingness of conspiracy theorists to scrutinize their own doubts – to doubt their doubts – is neither intellectually nor politically virtuous. The remaining question is whether doubt is a corrective virtue. There are many different senses in which this might be so. If corrective virtues stand at a point at which there is some temptation to be resisted, then the first stage in a defence of the idea that doubt is a corrective virtue is to identify the temptation that this putative virtue helps us to resist. Extremism is one such temptation but, for Foot, virtues are about what is difficult for humans generally. They are ‘correctives in relation to human nature in general’ (1978: 11) but there is no reason to think that human beings in general are tempted by extremism or find extremism a difficult temptation to resist. People whose temperament is, as it were, naturally moderate, don’t need doubt to help them to resist the temptations of extremism. They experience no such temptation. However, it is also undeniable that many humans are not constitutionally moderate and succumb too easily to the temptations of extremism. For them, doubt seems like useful corrective. More generally, doubt is a corrective virtue because humans tend to be too sure of themselves. An observation of Foot’s brings out the relevance of the fact that extremism is, if not universal, still widespread: ‘the thought that virtues are corrective does not constrain us to relate virtue to difficulty in each individual man’ (1978: 10). Accordingly, doubt can be a corrective virtue if it pertains to a common difficulty or temptation. If extremism is common enough – as it surely is – then any disposition of character or conduct that acts as a countervailing force has a claim to be classified as a corrective virtue. This brings out the role of context in our thought about political virtues. If we lived in a world in which extremism was not a major issue or one in which few political actors succumb to the temptations of extremism, there would be no need for doubt to act as an anti-extremist corrective virtue. We do not live in such a world, and it is because extremism is a problem that many human beings encounter in their political lives that doubt is in the running for the status of a corrective political virtue. There are three different varieties of extremism, and the 7/7 bombers were extremists in all three senses: the fact that they subscribed to an extremist ideology made them ideological extremists. Their willingness to use disproportionate and indiscriminate violence in pursuit of their political objectives also made them methods extremists. Methods extremists employ or endorse extreme methods in pursuit of their objectives. Lastly, the mindset of the 7/7 bombers, their preoccupations, attitudes, and ways of thinking, made them psychological extremists. The distinction between methods, ideological, and psychological extremism is further elaborated in Cassam 2022. Indeed, they were not just extremists in all three senses but also fanatics. Fanatics trample on other people’s ideals and interests, and are willing to sacrifice their own interests, to realize their ideals. This is exactly what the 7/7 bombers did. It is possible to be an armchair extremist – a person with extreme views but no wish to force anyone else to accept them – but not an armchair fanatic. What has been described as the fanatic’s willingness to ‘force his own ideals down the throats of other people’ (Hare 1965: 178) is in turn a reflection of the nature of those ideals. Fanaticism is incompatible with political ideologies that extol the virtues of pluralism and tolerance. It is not a given that extremism is a vice and therefore also not a given that doubt should be classified as a corrective virtue if it is an antidote to extremism. It has been argued that progress depends on extremism. Olson 2007. According to Martin Luther King Jnr., ‘the question is not whether we will be extremists, but what kind of extremists we will be. Will we be extremists for hate or for love? Will we be extremists for the preservation of injustice or for the extension of justice?’ (Luther King Jnr. 2018: 19-20). The extremism of 19th century abolitionists led ultimately to the ending of slavery, and the extremism of the suffragettes led ultimately to universal adult suffragettes, or so it might be argued. In the words of Merry, a character in a Philip Roth novel, ‘sometimes you have to fucking go to the extreme’ (Roth 1997: 105). Radical abolitionists like John Brown and William Lloyd Garrison were uncompromising in pursuit of the immediate emancipation of the slaves and thought of themselves as extremists or fanatics. As Olson stresses in Olson 2007. Certainly, their views about slavery were ‘extreme’ by the standards of their day, but it was extremism in a just cause. If doubts about their cause had slowed them down or made them less relentless in their pursuit of it that would have been a bad thing. If virtues are characteristics that ‘a human being needs to have, for his own sake and that of his fellows’ (Foot 1978: 3), then how can doubt be a virtue? The assumption that it is a virtue is based on the notion that extremism is generally speaking a bad thing but is that so obvious? One response to this question would be to resist the notion that extremism can be a good thing and has contributed to social progress. The abolitionists did not use disproportionate and indiscriminate violence in pursuit of their cause, and most were non-violent. Their views were radical in context but opposition to slavery is hardly an extreme position when viewed in wider perspective. Talk of psychological extremism is also out of place. Extremists are preoccupied with religious, racial, or ideological purity and with fantasies of victimhood. See the account of the extremist mindset in chapter 4 of Cassam 2022. They are pro-violence, intolerant, and indifferent to the adverse consequences of their actions. Their intolerance includes intolerance of doubt. They have a deep psychological need for certainty and cannot bear the uncertainty and anxiety that doubt brings. They are also prone to apocalyptic and conspiracy thinking. Every one of these preoccupations, attitudes, and ways of thinking can be detected in the mindset of the 7/7 bombers and none in the mindset of the abolitionists. Only the 7/7 bombers were extremists. The abolitionists were radicals rather than extremists, and their political virtues were the virtues of radicalism rather than extremism. This is a summary of the argument given in Cassam 2022, chapter 6. Colin Beck helpfully defines radicalism as ‘contention that is outside the common routines of politics present within a society, oriented towards substantial change in social, cultural, economic, and/or political structures, and undertaken by extra-institutional means’ (2015: 18). However, a problem remains: just as doubt can be an antidote to extremism, it can also function as a brake on admirable forms of radicalism, such as the radicalism of the abolitionists and suffragettes. Consider, first, how doubt might have stymied the 7/7 bombers. One can imagine the bombers filled with doubts about what they were about to do on the morning of 7/7. One can imagine the youngest of the bombers, 18-year-old Hasib Hussain, being paralyzed by doubts about his plan to detonate a bomb on the top deck of a London bus and deciding not to go through with it as a result. In this case, doubt would have been, quite literally, a lifesaver. More generally, doubt can act as an antidote to methods extremism by causing those who are extremists in this sense to question their methods. As far as ideological extremism is concerned, ideological extremists who are prepared to question or doubt their ideals or ideological convictions might end up abandoning or toning down their extremist ideologies. If in practice ideological extremists rarely display such doubts, that is because extremist ideologies are most appealing to individuals with an extremist mindset or, in other words, to psychological extremists. Psychological extremism predisposes a person to ideological extremism, and ideological extremism reinforces an extremist mindset. Fanatics and extremists are both said to have an unwavering commitment to an ideal and to be unwilling to subject the ideal to rational critique. Katsafanas 2019: 7. This suggests that doubt might be an effective antidote to psychological extremism and thereby indirectly also an antidote to ideological extremism. People who doubt their ideals and subject them to rational critique, cannot be said to have an unwavering commitment to those ideals. To be open to doubt is to be open to the anxiety that doubt brings. To introduce doubt into the mindset of the extremist is, in a very fundamental way, to undermine this mindset. Extremists thrives on the certainty that they are right, and view doubts about their cause or methods as a form of apostasy. It remains to be seen whether and how it is possible to sow the seeds of doubt in the mind of extremists but if such a thing is possible then it is a way of depriving them of one of their psychological crutches – the illusion of certainty. These are all ways in which doubt can be an antidote to extremism but how can doubt play this role without also obstructing admirable forms of radicalism? If doubt can paralyze extremists like Hasib Hussain and thereby preventing them from acting, it is equally capable of paralyzing radicals like Garrison and thereby prevent them from acting. Doubt looks like a virtue when it stands in the way of extremism but a vice when it stands in the way of progress. In that case, the obvious conclusion to draw is that doubt per se is neither a virtue nor a vice; it is all a question of context. When writers like O’Farrell represent doubt as a political virtue, this reflects the fact that they are writing in a context in which the threat of extremism is most salient. In a different context, it would be much less obvious that doubt is a virtue. This assumes that doubt is an obstacle to radicalism in just the way that it is an obstacle to extremism. This assumption is questionable. It is not quite true that radical social reformers like Garrison never doubted their principles or their tactics. Since they were under constant attack by defenders of the status quo, it would have been a remarkable psychological feat for them never to have wondered whether they were doing the right thing. However, such doubts were not paralyzing because they were resolvable. Doubts in the mind of Garrisonian radicals serve as a trigger to critical reasoning, and such reasoning enables the resolution of their doubts by showing them to be groundless. Doubt acts as a form of quality control, and commitments that can withstand doubt are more robust for having been subjected to doubt. In this sense, doubt is a political virtue for radicals and not an obstacle to radical political action in pursuit of justice. For the extremist, doubt is much more threatening because it cannot be resolved by competent critical reasoning. Extremists like Hussain cannot afford to doubt because doubts in their case would be unanswerable. The extremists’ low tolerance for doubt is one respect in which their mindset is pre-modern. The sociologist Anthony Giddens describes doubt as a ‘pervasive feature of modern critical reason’ (1991: 3). Modernity ‘institutionalises the principle of radical doubt and insists that all knowledge takes the form of hypotheses: claims which may very well be true, but which are in principle always open to revision’ (1991: 3). The doubt that Giddens describes is a form of Quinean fallibilism. As such, it can be anxiety-provoking because it highlights the possibility that even one’s most fundamental beliefs are mistaken. Radicals can live with this anxiety. Extremists cannot. Radicals can accept that no statement is immune to revision without seeing any reason to revise their core commitments. For extremists, in contrast, there is no question of accepting the principle of radical doubt or of seeing their basic commitments as open to revision. They cannot tolerate the existential anxiety that modernity entails, and this is what marks them out as pre-modern. Suppose, then, that doubt is a corrective political virtue in the following senses: it is an antidote to political extremism that is especially beneficial in contexts in which extremism is on the rise. It is possible for doubt to be antithetical to extremism without being antithetical to beneficial forms of radicalism because the latter can not only withstand doubt in a way that extremism cannot but also be strengthened by having been subjected to doubt and survived. If this is right, then the remaining question is whether it is possible to utilize the power of doubt in efforts to counter extremism. The challenge is to work out how to use the power of doubt to change the minds of individuals who, in virtue of their mindset, are peculiarly resistant to doubt in their thinking. This is the challenge that now needs to be addressed. 4 Radicalization has been defined as ‘the process whereby people become extremists’ (Neumann 2013: 874). Given the distinction between an extremist and a radical, this definition is potentially misleading. One might say that radical abolitionists like Garrison were radicalized but they did not thereby become extremists. However, the characterization of the process of becoming an extremist as the ‘radicalization process’ is now so widespread as to make it futile to insist on an alternative label. In practice, there are many radicalization processes because there are different types of extremism and different individual and idiosyncratic pathways to extremism. Cassam 2018. However, although there is no such thing as the radicalization process, it is relatively uncontroversial that ideological radicalization consists in the adoption of an extremist narrative that purports to provide a justification for the use of extreme methods, including terrorism. For example, the 7/7 bombers accepted a narrative according to which the West was at war with Islam and in which they had no alternative but to fight back. The leader of the 7/7 attackers, a man called Mohammad Sidique Khan, recorded a so-called ‘martyrdom’ video in which he explained and justified his action in the following terms: ‘your democratically elected government perpetrates atrocities against my people all over the world. And your support of them makes you directly responsible, just as I am directly responsible for protecting and avenging my Muslim brothers and sisters. Until we feel security, you will be our targets…. We are at war and I am a soldier’. The narratives that frame the thinking of radicalized individuals and groups are what Lawrence Freedman calls strategic narratives: Narratives are designed or nurtured with the intention of structuring the responses of others to developing events. They are strategic because they do not arise spontaneously but are deliberately constructed or reinforced out of the ideas and thoughts that are already current…. Narratives are about the ways that issues are framed and responses suggested. They are not necessarily analytical and, when not grounded in evidence or experience, may rely on appeals to emotion, or on suspect metaphors and dubious historical analogies (2006: 22-3). This has significant implications for the theory of deradicalization. Deradicalization ‘refers to measures and strategies taken to deradicalize or bring back groups or individuals to a previous point before becoming radical or before embracing violent extremist ideology’ (El-Said 2013: 6). The contrast is with ‘counter-radicalization’, which aims to prevent people from being radicalized in the first place. They are strategies ‘to convince and encourage groups and individuals who have already become violent extremists to permanently repent and abandon violence’ (ibid.). In accounts of deradicalization, there are few, if any, references to the role of doubt. However, in practice doubt plays a significant, albeit unacknowledged role, in deradicalization, and the immediate task is to understand what that role is. To the extent that radicalization consists in adoption of an extremist narrative, one can think of deradicalization as a three-stage process. At stage one, the aim is to persuade the target extremist to doubt his or her own narrative or at least to see the narrative as open to question. At this stage, the deradicalizer operates as a merchant of doubt who tries to induce the target to experience a form of existential anxiety as a result of the experience of doubt. At stage two, the extremist makes the critical transition from doubt to disbelief. Now the target is no longer just doubtful about the extremist narrative but come to view it as mistaken. At stage three, the extremist’s narrative is replaced by a more benign and more realistic strategic narrative. It might seem that deradicalization programmes with this tripartite structure are doomed to failure. If intolerance of doubt is at the heart of the extremist mindset, what hope is there of getting them to doubt their own narrative? This question confuses two distinct issues: whether extremists can tolerate doubt and whether they are immune to doubt. From the fact that doubt is, from an extremist perspective, intolerable, it does not follow that extremists are immune to it. There are many factors that might cause even hardened extremists to begin to question their narratives and ultimately to abandon them. One can imagine an extremist undergoing what L. A. Paul calls a ‘transformative experience’ that brings about a change of mind and a change of heart. Paul 2014. Work on the phenomenon of terrorist dropouts points to more prosaic factors that have been known to initiate a process of deradicalization. Terrorist dropouts are people who decide to leave terrorist groups. Jacobson 2010. Among the factors that impel people to drop out is disillusionment with the group’s leadership and with its ideology and theology. The other factors listed in Jacobson 2010 are petty grievances, unmet expectations, the role of family, a change in the personal circumstances of dropouts, and growing disillusionment with the group’s hypocrisy, especially when its leadership is seen as advancing its own interests in ways that are at odds with its ideology. In effect, they drop out because they come to doubt the group’s narrative, and this suggests that extremists are not immune to doubt. Dropouts have not been through deradicalization programmes. Their doubts are spontaneous rather than actively encouraged. However, the doubts that lead people to drop out of terrorist groups can be actively encouraged by deradicalization programmes. In order to be effective, deradicalization programmes must respect Descartes’ observation that before we can doubt, we need some reason for doubting. The mission is to give violent extremists reasons for doubting their narratives and methods. For this to be an effective strategy, the presented reasons need to be compelling – to ring true – and come from a trusted and respected source. So, for example, the deradicalization of violent Jihadists might focus on their misinterpretation or ignorance of the Quran. By subjecting them to a programme of religious instruction, they might be caused to doubt what they previously took for granted. In case this sounds fanciful, this is precisely how actual deradicalization programmes work. Of particular interest is Saudi Arabia’s deradicalization programme, which encourages detainees to repent and repudiate extremist ideologies: ‘through intensive religious debate and psychological counseling, religious scholars work to demonstrate that they have been following corrupted interpretations of Islam (Boucek 2008: 60). A notable feature of this programme is ‘the participation of distinguished scholars, scientists, and clerics’ (El-Said and Barrett 2013: 211). Saudi authorities encourage respected religious figures to visit extremists in prison to engage them in dialogue. The aim is to undermine the extremists’ narrative and expose defects in their religious understanding. In effect, the authorities use credible authority figures to sow the seeds of doubt in the minds of detainees who, at the end of a course in Islamic jurisprudence, sit an exam which they need to pass to move to the next stage of the programme. The Saudi programme accepts the need for detailed intellectual engagement with the ideology and narratives of Islamist extremists. Ashour 2009. In a battle of ideas, it is essential that one has the intellectual ammunition needed to prevail. In the Saudi case, religious scholars provide the necessary ammunition, and care is taken to ensure that those who sent to engage in a dialogue with extremists are seen as credible and not simply as government agents. The assumption is that many extremists are attracted to extremism by arguments and narratives, and that the remedy is a barrage of counter-arguments and counter-narratives that are designed to satisfy extremists that they have been misled by those responsible for their initial radicalization. The emphasis on the need to convince extremists presupposes a rationalistic conception of deradicalization: convincing extremists to repudiate their extremist ideologies is a matter of giving them reasons to change their views. It is relevant that the targets of the Saudi programme are fundamentalists. Unlike extremism more generally, fundamentalism is the ‘cult of the text’ (Ruthven 2007: 45). The subjects of the Saudi programme are failed fundamentalists. They purport to revere a canonical text whose prescriptions they claim to follow, but many barely know the text and rely instead on the simplifications and distortions of extremist ideologues who know little more than they do. A fundamentalist who misunderstands his creed can be, in certain circumstances, deradicalized by having the fundamentals explained to him by someone with demonstrably superior credentials. As Giddens notes, it is ‘probably rare for even the most fundamentalist of fundamentalist believers to escape radical doubt entirely’ (1991: 181). If this is right, then it provides an opening for deradicalization. The Saudi programme has been described as a Saudi solution to a Saudi problem. It relies on aspects of Saudi culture that are not common to all Muslim-majority countries and is of hardly any direct relevance to Muslim-minority countries where the most influential forms of extremism may have nothing to do with Islam. Far-right extremists do not revere a canonical religious text and cannot be deradicalized by a programme of religious instruction. However, there is a more general lesson to be drawn from the Saudi experience. The lesson is that doubt can be weaponized in the fight against extremism. Admittedly, inducing extremists to doubt their methods or ideologies is only the first step, and there is no guarantee that they will move on to stage two or three of the deradicalization programme. Step three is in many ways the most challenging because extremists may find it difficult to accept anti-extremist narratives even if they are doubtful about extremist narratives. Nevertheless, doubt is the first step on the road to deradicalization. When doubt contributes to the deradicalization of extremists, it operates as a corrective virtue. Clearly, doubt is not a panacea and there are undoubtedly extremists who seem immune to doubt. It is impossible to read the statements of Osama bin Laden without being struck by their absolute moral certainty. Lawrence 2005. One cannot imagine someone like bin Laden ever being pushed to doubt his fundamental assumptions. His retort might be to point out that his anti-extremist, liberal critics are no more willing or able to doubt their fundamental assumptions, including their assumption that bin Laden’s methods were morally repellent. This is exactly the type of pernicious false equivalence that needs to be firmly resisted. A true liberal or, for that matter, a true radical, should be open to doubt about his or her own outlook. However, being open to doubt is not the same as doubting. The challenge for extremists who try to turn the tables on their critics is to provide them with reasons for doubting their standpoint that are as compelling as the reasons for doubting the extremist’s own standpoint. Since no such reasons exist, the most straightforward way of dealing with extremists who try to turn the tables is to call their bluff. O’Farrell makes this point with admirable clarity in his account of the 7/7 bombers. They must have been so very sure they were right, but they were so very wrong. In contrast, Garrison was sure he was right, and he had every right to be sure. That is the crucial difference between the extremist and the radical. REFERENCES Ashour, O. (2009), The De-Radicalization of Jihadists: Transforming armed Islamist Movements (Abingdon: Routledge). Beck, C. J. (2015), Radicals, Revolutionaries, and Terrorists (Cambridge: Polity Press). Boucek, C. (2008), ‘Counter-Terrorism from Within: Assessing Saudi Arabia’s Religious Rehabilitation and Disengagement Programme’, The RUSI Journal, 153: 60-65. Burge, T. (1998), ‘Our Entitlement to Self-Knowledge’, in P. Ludlow & N. Martin (eds.) Externalism and Self-Knowledge (Stanford: CSLI Publications): 149-73. Button, M. E. (2016), Political Vices (Oxford: Oxford University Press). Cassam, Q. (2018), ‘The Epistemology of Terrorism and Radicalisation’, Royal Institute of Philosophy Supplement, 84: 187-209. Cassam, Q. (2019b), Conspiracy Theories (Cambridge: Polity Press). Cassam, Q. (2022), Extremism: A Philosophical Analysis (London and New York: Routledge). Descartes, R. (2017), Meditations on First Philosophy With Selections from the Objections and Replies, ed. J. Cottingham (Cambridge: Cambridge University Press). El-Said, H. (2013), ‘Introduction: Definitions and conceptual framework’, in El-Said & Harrigan 2013: 1-14. El-Said, H. & Barrett, R. (2013), ‘Saudi Arabia: the master of deradicalization’, in El-Said & Harrigan 2013: 194-226. El-Said, H. & Harrigan, J. (2013), Deradicalizing Violent Extremists: Counter-radicalization and deradicalization programmes and their impact in Muslim majority states (London and New York: Routledge). Foot, P. (1978), Virtues and Vices and Other Essays in Moral Philosophy (Oxford: Blackwell). Freedman, L. (2006), ‘Networks, Cultures and Narratives’, in The Transformation of Strategic Affairs (New York: Routledge): 11-26. Galston, W. (1991), ‘Toughness as a Political Virtue’, Social Theory and Practice, 17: 175-97. Giddens, A. (1991), Modernity and Self-Identity: Self and Society in the Late Modern Age (Cambridge: Polity Press). Hare, R. M. (1963), Freedom and Reason (Oxford: Oxford University Press). Jacobson, M. (2010), Terrorist Dropouts: Learning from Those Who Have Left (Washington DC: The Washington Institute for Near East Policy). Katsafanas, P. (2019), ‘Fanaticism and Sacred Values’, Philosophers’ Imprint, 19: 1-20. Kay, J. and King, M. (2020), Radical Uncertainty (London: Bridge Street Press). Kraditor, A. (1989), Means and Ends in American Abolitionism: Garrison and His Critics of Strategy and Tactics, 1834-1850 (Chicago: Elephant Paperback). Lawrence, B. ed. (2005), Messages to the World: The Statements of Osama bin Laden (London and New York: Verso). Luther King Jr., M. (2018), Letter from Birmingham Jail (London: Penguin). Lynch, M. (2021), ‘Conviction and humility’, in M. Alfano, M. Lynch and A. Tanesini (eds.) The Routledge Handbook of Philosophy of Humility (Abingdon: Routledge): 139-47. McMaster, H. R. (1997), Dereliction of Duty: Lyndon Johnson, Robert McNamara, the Joint Chiefs of Staff, and the Lies that Led to Vietnam (New York: HarperCollins). Neumann, P. (2013), ‘The trouble with radicalization’, International Affairs, 89: 873-93. Obama, B. (2020), A Promised Land (London: Viking). Nussbaum, M. (1993), ‘Non-Relative Virtues: An Aristotelian Approach’, in M. Nussbaum and A. Sen (eds.) The Quality of Life (Oxford: Oxford University Press): 242-69. O’ Farrell, J. (2017), Things Can Only Get Worse? Twenty Confusing Years in the Life of a Labour Supporter 1997-2017 (London: Black Swan). Olson, J. (2007), ‘The Freshness of Fanaticism: The Abolitionist Defence of Zealotry’, Perspectives on Politics, 5: 685-701. O’Neill, O. (2002), A Question of Trust (Cambridge: Cambridge University Press). Paul, L. A. (2014), Transformative Experience (Oxford: Oxford University Press). Pigden, C. (2017), ‘Are Conspiracy Theories Epistemically Vicious?’, in K. Lippert-Rasmussen, K. Brownlee, and D. Coady (eds.) A Companion to Applied Philosophy (Chichester: Wiley Blackwell): 120-32. Roth, P. (1997), American Pastoral (London: Vintage Books). Ruthven, M. (2007), Fundamentalism: A Very Short Introduction (Oxford: Oxford University Press). Sageman, M. (2017), Misunderstanding Terrorism (Philadelphia: University of Pennsylvania Press). Salmon, N. (1995), ‘Being of Two Minds: Belief with Doubt’, Noûs, 29: 1-20. Sinha, M. (2016), The Slave’s Cause: A History of Abolition (New Haven: Yale University Press). Strawson, P. F. (2008), ‘Social Morality and Individual Ideal’, in Freedom and Resentment and Other Essays (Abingdon: Routledge): 29-49. Warren, M. (1999), ‘Introduction’ to M. Warren (ed.) Democracy and Trust (Cambridge: Cambridge University Press): 1-21. Weber, M. (1946), ‘Politics as a Vocation’, in H. H. Gerth and C. Wright Mills (eds.) From Max Weber: Essays in Sociology (New York: Oxford University Press): 77-128. Williams, B. (1978), Descartes: The Project of Pure Enquiry (Harmondsworth: Penguin). Woodward, B. (2006), State of Denial (London: Simon & Schuster). 29