Keywords

From the outside, it is easy to see tyrannies in terms of repression and of loathing. When we think of Nazism, we think of the Gestapo, the camps, the terror which rendered opposition perilous at the very least. Equally, when we think of Italy’s fascist period, our overwhelming image is of grim-faced blackshirts. Certainly, we would not wish to diminish in any way the violence and brutality of either regime. Yet, if we want to understand how such systems worked, why they were able to thrive, and hence how they can best be opposed, such a focus may be misleading. From the inside, the most striking aspect of tyrannies may be the sense of participation and of devotion. To put it slightly differently, when we analyse the outpourings of Nazis and Fascists, we tend to focus on “hate speech”. What should concern us more is “love speech”.

Much of this love is centred on the figure of the leader. This is clear in the words of Athe Gracci, cited above, taken from Duggan’s (2013) account of the voices of ordinary Italians in the period of fascist rule. He regards such devotion as much more than a curiosity. Indeed Duggan’s central argument concerns “the crucial importance of the figure of Mussolini to emotional and political engagement with the regime” (p. xvii).

A similar picture emerges from Eberle’s (2012) study of the thousands of letters written to Hitler by ordinary Germans between 1925 and 1945. Rather than force securing consent, Eberle argues, Hitler was explicit that his ability to use force depended upon his popularity. This in turn depended upon creating an intimate relationship between the population and their leader. As Harris puts it, in her introduction to Eberle’s analysis: “the secret to the Third Reich’s success during its height in the 1930s and early 1940s was Germans’ sense that they could engage in a conversation with their leader, and that he was in some way listening” (2012, p. 1).

If we want to understand the psychology of tyranny, then we must address why and when people embrace authoritarian leaders. How can people be devoted to figures who take away their freedom and threaten extreme violence to any who would question, let alone oppose, them? But before we can do this it is first necessary to address the nature of leadership itself and to specify what authoritarian leadership is.

On the Nature of Tyrannical Leadership

Leaders, Followers and Social Groups

There is a vast literature on leadership, which we have no space to review in any detail here. However, a general survey of the field (e.g. as provided by Haslam, Reicher, & Platow, 2011) reveals that in recent decades “great man” theories—which suggest that leaders have a set of special personal qualities that set them apart from the mass—have given way to theories which see leadership as resting upon a social relationship between leaders and followers. On the one hand, transactional models see leadership as a form of exchange: if the leader can give followers what they want, they in turn will do what the leader wants. Transformational models see this as underestimating the extent to which leaders reshape the desires and goals as followers. But these come perilously close to reinstating notions of leaders as being endowed with special qualities such as charisma (Burns, 1978; Bass & Riggio, 2006).

Perhaps what is most striking about these various approaches is less what they include than what they exclude. To be more specific, they generally ignore the fact that leadership is something that takes place in groups. A leader is always a leader of a specific group—a faction, a party, a nation, whatever. For those of us who look at Hitler and Mussolini now, they appear as strange, almost absurd figures and we certainly have no sense of intimacy or devotion to them. They were leaders of Germany and of Italy at a distinct moment in time. Thus, to understand leadership we cannot limit our focus to the leader alone, we cannot limit it to just leaders and followers; we have to examine the relationship between leaders and followers within a social group defined by both place and time.

One obvious exception to this is Freud’s Massenpsychologie und Ich-Analyse, first published in 1921 and translated into English in 1922. Freud starts with an appreciative summary of Gustave Le Bon’s classic crowd psychology—the notion that people lose their identity in the crowd, revert to a common primitive “racial unconscious”, and therefore become both extremely suggestible and extremely atavistic in their actions (Le Bon, 1895/1947, see also La Macchia & Louis, 2016). However, he criticises Le Bon for neglecting the importance of leadership in his analysis. The leader, for Freud, is akin to the father of the primal horde or the hypnotist in therapy. He (for Freud’s vision is highly gendered) controls both individual group members and binds them together in the group. To use Freud’s own words:

“A primary group… is a number of individuals who have substituted one and the same object for their ego ideal and have consequently identified themselves with one another in their ego” (1922, p. 80).

Put slightly differently, it is the shared vertical relationship of veneration to the leader which generates the horizontal bonds of identification between group members. This stress on a love relationship (a libidinal tie) between leaders and followers captures the intense personal relationship that we have commented upon. But the problem of characterising this relationship as prior to group identification is that it places no boundaries on the power and influence of the leader. As acknowledged by even those who are sympathetic to Freud (e.g. Ernesto Laclau in his, 2005, analysis of populism), this all too easily leads to the conclusion that leadership is necessarily a top-down process in which anything the leader says will be accepted by followers. In these terms, leadership inherently involves subjugation to the leader’s dictatorial will, with the only choice being between benevolent or else toxic dictatorship. Democratic leadership, by contrast, is an oxymoron and a logical impossibility.

It follows that, if we want to avoid an equation of leadership with tyranny, we need to provide an alternative conceptualisation of the relationship between group members, leaders and social groups. The social identity approach (Tajfel & Turner, 1979; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) provides just such a reconceptualisation.

The Social Identity Analysis of Leadership

As outlined by Hogg (2016), the social identity approach places the act of identification—seeing oneself in terms of one’s membership of the relevant social category (e.g. “I am German”, “I am a Catholic”)—firmly at the start of group process . Where people identify with a given social category they seek to conform to the norms, values and beliefs which characterise this category. They will therefore be more likely to be influenced by those who are in a position to understand these norms, values and beliefs and to interpret what they mean in concrete terms for action in context (Turner, 1991, see also McGarty, Haslam, Hutchinson, & Turner, 1994). In other words, group members will be swayed by those who are themselves emblematic of the group—in more technical language, those who are most in-group prototypical.

However, even if someone is indeed highly prototypical, it doesn’t mean that anything they propose will be supported. For even if they themselves represent what the group stands for, it is still necessary that what they propose and what they do is seen as consonant with group understandings and group interests. Thus, from a social identity perspective, the primacy of identification places limits upon the process of influence and hence upon the ability of anyone to sway group members.

While not labelled as such, this model of influence clearly lays down the basis for a model of leadership. Over time, social identity theorists have made the connection more explicit. At first, the emphasis was very much on the construct of in-group prototypicality (e.g. Hogg, 2001; Hogg, Hains, & Mason, 1998). More recent work endorses the importance of prototypicality but suggests it is just one of several dimensions along which leadership is framed by social identity process (for a summary of evidence, see Haslam et al., 2011; see also Steffens et al., 2014, for a validation of this multidimensional approach).

First, then, leaders need to be seen as “one of us”, as part of the group and as representing group values. But it is important to stress that this does not mean that leaders should be typical of other group members. It means that they should represent the core characteristics which we value in our group and which make us distinctive from others—characteristics we may believe in but which few of us actually live up to. Prototypical, then, is very different to typical. It means that the leader has to be an extraordinary rather than an ordinary member of the group (Steffens, Haslam, Kessler, & Ryan, 2013).

Second, leaders need to be seen as “acting for us”. Certainly to be seen as acting for one’s own interests is extremely corrosive for effective leadership, which explains why would-be leaders are often reluctant to be seen as seeking power. In this regard, the emperor Cincinnatus is often presented as a model of the good leader—a man who had to be persuaded out of retirement to defend Rome against the Aequians and who, once he had succeeded, quietly went back to his farm (Livy, 1922). It is even worse to be seen as acting for an out-group. Indeed, far from favouring leaders who are just and fair, unless there are explicit group norms to the contrary, we generally favour leaders who will favour in-group over out-group members (e.g. Platow, Hoar, Reid, Harley, & Morrison, 1997).

Third, it is not enough to simply act in the in-group interest . Successful leaders need to deliver for the in-group. They need to transform group norms and values into lived realities—a process we refer to as “collective self-realisation ” (Reicher & Haslam, 2006a). This does not necessarily mean economic success or victory in competition; what counts as success will depend upon the specific norms and goals of the group in question and will therefore be very different in different circumstances. But whatever form it takes, collective self-realisation lies at the root of the powerful positive emotions that can occur in groups (Hopkins et al., 2016) and the powerful positivity towards leaders who are seen to have made it happen.

Collective self-realisation can occur at two levels. On the one hand, it is a matter of building norms and values into the practices of the group itself. In this way, group practices become a performance of identity and the leaders who create these performances act as impresarios. Hitler’s Nuremburg rallies are a case in point. These were choreographed so that the leader would emerged from the order ranks of Nazis and mount a raised dais above them: of the mass yet dominating the mass, an embodied demonstration of the fuhrerprinzip on which the Nazi vision of an ideal German society was to be built (Spotts, 2002).

On the other hand, collective self-realisation is a matter of actions taken to transform the practices of the wider society. Shifting from Hitler to Mussolini, it is relevant to cite one of the Italian dictator’s acolytes, Giovanni Giurati. Mussolini, he believed, was the man “to chase moral and civil disorder, heresy and war, not just from Italy, but from the face of the earth” (Duggan, 2013, p. 79). Or, to cite another, Alberto De’ Stefani, Mussolini did not just exalt “national sentiment and the power of the State, in opposition to the democratic, pseudo-liberal, pacifist and humanitarian ideologies”—something seen to exemplify an ancient Italian tradition going back to the Romans—he was the man who turned “the word” into “action” and, we may add, who turned action into the (mercifully brief) reality of the Italian fascist state (Duggan, 2013, p. 79).

The fourth and final dimension of “ identity leadership” is on a different level, and encapsulates the other three. As we have argued, those who share identification as members of a common social group seek to act together on the basis of the meanings associated with the group identity. This means that those who are able to define the nature of this identity have the potential to mobilise collectivities in favour of their proposals. That is, it is through their capacity to define identities that leaders acquire social power. As a consequence, rival leaders, seeking to mobilise people to different ends, will contest the meanings associated with group identity. Leaders, then, must craft a sense of us. To be successful, they must be skilled entrepreneurs of identity (Reicher & Hopkins, 2001).

Such entrepreneurship must encompass all aspects of what a leader does, leaders don’t just wait and hope that they will be seen as exemplifying the group prototype; they actively construe the group identity and their own selves so as to establish a consonance between them. This is not limited to what they say about themselves and the group, but extends to all dimensions of what they do and even how they look. Let us use two examples, one serious and one more frivolous, to illustrate this point.

By many estimates, Franklin D. Roosevelt was one of the greatest of all US Presidents. But when, at the age of 39, in 1921, he was struck down with what was thought to be polio (then, often dubbed “infantile paralysis”) it was thought to be the end of his political career. After all, surely a politician had to be virile, energetic, autonomous—everything that was negated by his disability. But Roosevelt persisted. Still, when he first stood for President in 1932, during the depths of the depression, and decided to go on a whistle-stop train tour around the country, his advisors strongly counselled against displaying his ravaged body to the public. Yet the sight of Roosevelt painfully, but successfully, dragging himself from the train to his podium resonated both with his audience’s sense of hardship and with his message of an America able to triumph over economic paralysis (Leuchtenburg, 1995; Rosenman, 1952). This is best expressed in Roosevelt’s most famous words, taken from his inaugural speech of March 4th 1933: “this great nation will endure as it has endured, will revive and will prosper. So first of all, let me assert my firm belief that the only thing we have to fear is fear itself—nameless, unreasoning, unjustified terror which paralyses needed efforts to convert retreat into advance”.Footnote 1 In short, Roosevelt had the entrepreneurial skill to convert his body from a liability into a symbol of the nation’s path.

Now, let us turn from the (almost) sublime to the (almost) ridiculous. In 1993, the Conservative MP, Bill Walker, stood up in the British Parliament to propose legislation that would make Scottish devolution more difficult. In order to claim that he was talking for, and not against, the Scottish people and the Scottish interest, Walker, resplendent in tartan, began: “I stand before you, madam speaker, wearing the dress of Highland Scotland”. But at this point he was interrupted by a fellow MP who objected: “On a point of order Madam Speaker, my honourable friend… suggested that he was in Highland dress. He is in nothing of the kind. He misled the House and I have reason to believe that he is wearing little red pants under his kilt” (cited in Haslam et al., 2011, p. 154). Such is the gravitas of the “mother of parliaments”. Our point, though, is that when it comes to establishing that you are of the group, that you are for the group and that you deliver to the group, then everything in your performance—even your underwear—can matter.

Leadership and Democracy

One of the key contributions of the social identity approach is that it resolves the issue of agency, which has bedevilled traditional approaches to leadership and which—as we have seen—also constitutes a key problem for the Freudian account. The question is whether the agency of leaders and that of followers are inherently at odds with each other; in sociopolitical terms, is strong democratic leadership possible? Whether wittingly or not, most approaches suggest it is not possible. They flip-flop between rendering the leader completely autonomous and able to impose anything upon followers (great man theories) to making leaders completely dependent upon pre-existing preferences of followers (transactional theories) and then back again (transformational theories). But, by placing the leader–follower relationship within a frame that equally encompasses both of them—the social group—the social identity approach opens up the possibility that both can be involved in defining the nature of social identity and hence determining forms of social action. Strong leadership can facilitate rather than exclude the participation of group members in this process. It can promote respect and harmony within and between groups as much as derision and conflict.

It is important to stress that, in opening up the possibility of strong democratic leadership, social identity models do not exclude undemocratic forms. On the contrary, it is precisely through its stress on leadership as a process of social identity management, and the centring of analysis on the balance between leaders and group members in arriving at the definition of social identity, that these models are able to specify how and when leadership will be more democratic or more authoritarian.

In our analysis of The New Psychology of Leadership , we have identified three broad types of leadership based on three different orientations to the definition of identity (Reicher, Haslam, & Platow, 2014). The first, we call “ideal democratic” leadership. Here, the leaders seek to facilitate an inclusive discussion about “what we are” and hence what our policies and priorities should be. Here, the leader assumes no necessary priority and the mass of group members are given an equal voice in defining the collective identity. This idea is often promulgated through the metaphor of a “conversation”—a national conversation, a fireside conversation or whatever. But often when leaders claim to be having a conversation, it obscures a second and far more common form of leadership which we term “asymmetrical” or “hierarchical” leadership.

Hierarchical leadership denotes a situation where leaders claim priority in terms of the definition of identity and seek to essentialise their particular versions of “who we are” as the only possible versions. In other words, even as they are involved in constructing identity, they seek to hide the activity of construction. They portray themselves more as archaeologists revealing what was already there and will always be there.

There are a number of ways through which such essentialised versions of identity can be achieved. Perhaps the most common is the use of history. Or rather, to be more precise, a number of events will be selected to represent the “true” identity of the group and hence to define an essence which continues unchanging across time and outside of history. Different choices of events will then lead to very different versions of identity (see Reicher & Hopkins, 2001, for an extended analysis).

But even if hierarchical leadership seeks to establish a leader’s version of identity as the one “true” version, it still positions that leader as no more than an interpreter of who we are. That is, a distinction still exists between the leader and the identity and this has a number of important implications. To start with, as an interpreter, there is always the possibility that the leader is fallible in their interpretation of identity. As a consequence, criticisms of the leader’s version as well as alternative versions of identity are possible and legitimate. Equally, criticisms of the leader are possible and legitimate. All in all, hierarchical leadership still allows for democratic debate even if it tries to tip the terms of the debate in the leader’s favour.

All that changes with the third type: authoritarian leadership . Here, the leader is elided with the group itself. He or she becomes the embodied manifestation of the identity. For example, the Indonesian dictator Sukarno symbolically represented himself as the Javanese hero and demigod Bima (see Wilner, 1984). He altered his speech away from the quiet cadences of the Indonesian elites to the crude booming tones associated with Bima. Physically, he emphasised his large muscularity, again so dissonant with the graceful fine features of the Javanese aristocracy but so reminiscent of Bima. He even drew upon the colour black, a symbol of strength always associated with Bima, by carrying a black baton wherever he went. Sukarno did not just draw on Bima, he became Bima and hence Indonesia. For a more succinct way of encapsulating this relationship between leader and group, we can refer to the climactic words with which Rudolf Hess concluded the 1934 Nüremburg rally: “The Party is Hitler. But Hitler is Germany just as Germany is Hitler” (cited by Kershaw, 2001, p. 69). Moreover, Hitler returned the compliment: “I know that everything you are, you are through me, and everything I am, I am through you alone!” (cited in Fest, 1974, p. 159).

As we shall now examine in some detail, once the distinction between the leader and the group is obliterated, the space for debate and dissent also disappears. Under authoritarian leadership, the leader quite literally defines the group and gives group members no say over what the group is and what the group can or should do. They can only follow. It is important to stress that not all leaders who are elided with the group will use this to deny debate (Nelson Mandela is a case in point). It is equally important to stress that those who do so will not limit themselves to rhetorical denial of dissent, our point rather is that we should be wary of turning representatives into icons for this creates the conditions for leaders to become tyrants.

The Tyrannical Consequences of Authoritarian Leadership

As analysts, looking at tyranny from the outside, we understandably concentrate on all its many unpalatable dimensions. But from the inside, and for tyranny to succeed, it must have its attractions. Let us start with one of these, one which we have already had cause to mention when discussing Freud’s Massenpsychologie. We refer to the intense, libidinal, almost erotic tie between leaders and the led. While we may doubt Freud’s explanation, that does not mean to say that we dismiss the phenomenon. Indeed, it is striking, when one reads accounts of those who have attended rallies addressed by authoritarian leaders, that instead of feeling an anonymous part of the mass, they have a sense of making personal contact, of being personally addressed by the leader and of the intensity of that relationship. To quote from one early Nazi who saw Hitler speak: “I looked at him as he passed by and felt that he met my glance. All who have ever seen him must have felt the same way. The shouts of Heil Hitler continued” (cited in Abel, 1986, p. 271). And another: “I felt as though he were addressing me personally. My heart grew light, something in my breast arose. I felt bit by bit something within me were being rebuilt” (cited in Lindholm, 1990, p. 102).

The ability to engender such feelings is often held as the mark of a leader’s charisma. Yet, unless one can explain the basis of charisma this only serves to redescribe—or, at worst, to mystify—the phenomenon. From a social identity perspective, we are more likely to see a leader as charismatic if we see them as prototypical of our group (“one of us”; see Platow, van Knippenberg, Haslam, van Knippenberg, & Spears, 2006). What is more, when leaders are seen as prototypical of a group with which we identify (e.g. committed Democrats who see Obama as prototypical of the Democratic Party) we have a sense of having a personal bond with them (Steffens, Haslam, & Reicher, 2014). Moreover, experimental research that reproduces these patterns shows that these are not simply associations, but also causal pathways. In-group prototypicality is thus not just a correlate of, but also a basis for charisma, and also for a sense of personal connection to the leader.

To understand this, it is necessary to appreciate that the social identities we derive from group memberships are every bit as meaningful to us personally as the identities attached to our unique individuality. They define who we are, what we care about, what we strive for (Reicher, Spears, & Haslam, 2010). Accordingly, insofar as we represent ourselves personally in terms of our group membership then we also feel a sense of personal connection to those who represent the group.

In sum, whereas Freud argues that group members are linked horizontally to each other through their relationship to the leader, a social identity approach argues that leaders and group members are linked to each other through their mutual relationship of identification with the group. In this way, we retain the ability to explain how followers feel personally engaged with leaders, but without implying that this is always an unequal engagement.

Extending this same logic a step further, it follows that the more we identify with the group and the more the leader is identified with the group, the stronger our bond to the leader (a point confirmed empirically by Steffens, Haslam, Reicher, Platow, et al., 2014). At the point where the self is fused into the group and the group is elided with the leader, then the follower becomes inseparable from the leader. This point is chillingly articulated by one early Nazi, describing his experience of hearing Hitler speak in Bonn in 1926: “The German soul spoke to German manhood in his words. From that day on I could never violate my allegiance to Hitler” (cited in Abel, 1986, pp. 152–153).

Moreover, in this quotation, we see not only the intensity and the intimacy of the bond to an authoritarian leader, but also begin to glimpse the impossibility of dissent. If the leader has become the embodiment of the group (e.g. Hitler as “the German soul”), then any distancing from the leader becomes a distancing from the group, any attack on the leader becomes an attack on the group; difference becomes apostasy and debate becomes betrayal. Moreover, and here’s the rub, it becomes distancing, attack and betrayal of ones own self as defined through the group. Accordingly, divergence from the path the leader sets out becomes unconscionable – and the impossibility of dissent is not just a matter of self-policing by group members, it is also vigorously policed by authoritarian leaders and their agencies.

To illustrate these points with a more contemporary example, Penic, Elcheroth and Reicher (in press) recently examined the relationship between what have been called “modes of attachment” to the nation (Roccas, Klar, & Liviatan, 2006), and criticism of the nation’s excesses. More specifically, the research was interested in the claim by Roccas and her colleagues that such criticism is not a function of how strongly one identifies but rather a matter of how one orients to group authorities and whether one regards criticism as a way of strengthening or weakening the group. Critical patriots, argue Roccas et al., using evidence from Israel, can both love their group and criticise their group (see also Packer, 2008). Indeed they can love their group precisely by criticising it. We found exactly this pattern in Serbia, in the aftermath of the wars of the 1990s. But we didn’t find such a relationship in Croatia, principally because, in that country, there was no space for critical patriots. Why?

The answer lies in the fact that, for Croatians, these wars, locally named “the Homeland War” were seen as founding the modern nation, and the Croatian leader of the time, Franco Tudjman became regarded as the “Father of the Nation”. In this context, those who openly criticised either the war or Tudjman’s role within it were deemed unpatriotic, alien and illegitimate. This, indeed, was the reaction to an episode of a documentary programme entitled “Tudjman’s legacy” aired on Croatian television, which became a cause célèbre and the subject of an extended debate in the national parliament. In concrete terms, the outcome of this was that the show was suspended for a month, and several of those involved in its production were given formal warnings. Such was the outcry against the editor that he had to be placed under police protection.

Particularly revealing is the following intervention in the parliamentary debate. Rather than criticising Tudjman, averred the speaker, his speeches should have been broadcast: “so our children can see who the great man was, instead of my daughter asking me in the morning (…) “how come they are speaking like that about our Franjo?” (cited in Penic, Elcheroth, G, & Reicher, in press). Here, criticism of the “father” (as expressed through the personalised diminutive “Franjo”) is represented as an assault that distresses the children of the nation, as manifested through the child of the speaker. And what could be more nasty, vicious and unwarranted than upsetting a harmless little child?

There is one final twist to our argument. If followers identify with a group which is embodied through the leader, then an attack on the leader becomes not only an attack on the group but also an attack on oneself. Understanding this helps to make sense of one of the more perverse features of twentieth century dictatorships. Getty and Naumov (1999) provide a comprehensive and relentless document-based account of the Soviet terror of the 1930s. They show that the repression was not simply the work of Stalin, but a matter of consensus. From the inside it was seen less as a terror and more as a “war on terror” aimed against agents or dupes of fascism who posed a pressing threat to the Soviet Union (see Overy, 2004). But, what is perhaps even more remarkable is that many of those indicted and facing execution colluded in their own repression—even though they knew themselves to be entirely innocent of what were often entirely trumped-up charges.

The power of the analysis lies in how, through presenting the original transcripts of the trials, Getty and Naumov reveal what the revolutionary leader Karl Radek once called “the logic of confession”—a logic which, once fully developed, makes these extraordinary acts seem almost banal. The starting point is the familiar elision between leader, party and people. This is exemplified by the speech of Ivan Akulov at the plenum of the Communist Party Central Committee (the CC) in January 1933: “Stalin’s policy is our policy, the policy of our entire party. It is the policy of the proletarian revolution… and these gentlemen [i.e. those accused of dissent] will never succeed in separating us from our leader” (p. 80).

The next step is to suggest that, in challenging an accusation by the Party, rather than accepting one’s guilt (as a good Communist should) one is claiming that the Party (and hence both Stalin and the people) are fallible, thereby weakening them precisely at the moment when need to be strong in order to counter the mortal threat of German fascism. This is a repeated theme in the trials as defendants who dare defend themselves are hectored for that very act. The defendants are thereby positioned as the aggressors. This is exemplified in the following attack on Bukharin, possibly the most prominent of all those who were purged: “When I read Bukharin’s note concerning the charges against him I felt such disgust, as if you saw before you a snake, a viper. I’m sure everyone of you felt the same. From first word to last this note is steeped in vile insinuations and assaults from the CC. It is steeped in a spirit of confrontation, in which he perceives himself as offended or oppressed by somebody” (p. 385).

The dilemma, then, is that the defendants themselves were committed Communists, whose lives were committed to the revolution and who accepted the leading role of the Party (and the General Secretary within it) as the voice of the proletariat—and who were also aware of the gathering threat of fascism. They accepted that to defend themselves as individuals would weaken the leader, the party and the people through which they defined themselves personally. Did they then seek to protect their personal selves at the cost of damaging their social selves or vice versa? Many put their social self first and confessed even if that meant death. Bukharin himself tried a middle path: “while pleading guilty and admitting to the overall validity of the fantastic charges made against him, he nevertheless refused to confirm specific details of the supposed conspiracy.... he may have been trying to fulfil his party duty… by simultaneously confessing and defending his personal honor” (p. 526).

Bukharin was duly executed the day after his trial. He was well prepared for this fate, though. For in the course of his trial, he wrote a letter to Stalin—“perhaps the last letter I shall write to you before my death”—in which he assured the Soviet leader that he understood how the fate of the proletariat (as promoted by Stalin) was more important than his own fate. His sole expressed regret was not for his own imminent death but for the implication that he was disloyal to the group and its leader. As he put it: “my heart boils over when I think that you might believe that I am guilty of these crimes” (p. 558, emphasis in the original). More generally, to adapt Radek’s turn of phrase, the elision of leader with the group entails a logic of tyranny whereby not only is criticism, dissent and debate rendered illegitimate and subject to repression, but dedicated group members collude in this repression even when they are its victims.

On the Acceptance of Tyrannical Leadership

This far we have been analysing the nature of tyrannical leadership: we have argued against the notion that strong leadership is necessarily undemocratic but rather claimed that tyranny stems from a particular relationship between leaders and followers in defining the group identity whereby, whatever the leader is, does or says ipso facto characterises who we are. We then examined exactly why and how such a form of leadership leads to the destruction of any space for debate or dissent. What we have not yet done, though, is to explain why and when followers would accept or even embrace such forms of leadership. That is what we will do briefly in the space that remains to us—briefly, because these are matters that (to some extent) we have addressed before (e.g. Haslam & Reicher, 2012a,b; Reicher & Haslam, 2012, 2014).

One of the strange paradoxes of our discipline is that, although we have been much concerned with issues of authority and tyranny, we fail almost completely to address these “why and when” questions. That is because we tend to take submission to authority as given even as a characteristic of particular individuals (as in authoritarian personality research) or of human beings in general. Indeed the two most famous bodies of research in the history of psychology—Milgram’s Yale Obedience studies and Zimbardo’s Stanford Prison Experiment —are characteristically represented as showing, respectively, that people are predisposed to follow the orders of authority, no matter how toxic, and that people are predisposed to accepting the roles in which they are cast, no matter how oppressive (see Reicher & Haslam, 2006b; Reicher, Haslam, & Miller, 2014). So it doesn’t make sense to ask when we succumb to tyranny because it is assumed that whenever there is tyranny, we can’t help but succumb to it.

But this leads us to a second and perhaps even stranger paradox. Even as these classic studies are represented as demonstrating that human conformity is natural, so what they actually show is the ubiquity of resistance. Across all the obedience studies conducted by Milgram—where experimenters urge participants to inflict an escalating series of electric shocks on “learners” (actually Milgram’s confederates) each time they make errors on a learning task—the majority of participants (52 %) eventually disobey rather than obey the experimenter.

Equally, in Zimbardo’s Stanford experiment—where students were randomly divided into Prisoners and Guards and immersed in a simulated Prison—nearly all the Prisoners challenged the Guards at the start, some continued to the end, and only a minority of the Guards (of whom one, dubbed “John Wayne”, stands out) acted repressively or tyrannically. What is more, in one final twist to the tale, it is precisely this message of resistance alongside conformity (rather than of the inevitability of conformity) which makes these studies relevant to the wider world. For even in the most brutal carceral regimes that human beings have managed to devise, there is always some evidence of resistance to the extent that sometimes the prisoners are successful in taking over the prison (Haslam & Reicher, 2012c).

It was these considerations that led us to run our own prison study. We wanted to revisit Zimbardo’s contention that people “naturally” adopt the group roles into which they are cast and, in particular, that they will abuse positions of power. We also wanted to reconsider the lesson that Zimbardo drew from this: that we always should avoid groups and power. In our study, as in Stanford, ordinary men were divided into Prisoners and Guards. If anything, this showed that our previous scepticism about the inevitability of conformity was too mild. The Guards were highly reluctant to accept their power and use it to maintain the system. The Prisoners were highly reluctant to accept their powerlessness and to accept the system. The result was that, in a short period of time the Prison hierarchy collapsed. But, for present purposes, it was what happened next that was particularly interesting (see Reicher & Haslam, 2006b for details).

After the fall of the Guards’ regime, the participants met together and decided to set up a non-hierarchical system—a “Commune”—in which all chores and all resources would be discussed collectively and allocated equitably. Unlike the original Prisoner-Guard arrangements, this was a system that most of the participants themselves embraced and believed in rather than one imposed from the outside. But despite this, the new system quickly collapsed and in the wake of this those who believed in it were ready to accept the imposition of a quasi-fascist regime. The reasons why this happened can be encapsulated by three key moments in the short history of the Commune (these can be seen in full detail by watching the BBC documentary series “The Experiment”; Koppel & Mirsky, 2002).

First, two of the “Communards”, who had taken responsibility for the allocation of chores, were discussing what to do if people declined to do the task they had been given (they were aware that there were dissenters amongst them who were not keen to do their share). “Give them another task”, suggested one. “What happens if they don’t want to do that too”, replied the other. And then there was a long silence. Certainly, there was no suggestion of going beyond polite requests in the attempt to enforce the procedures of the Commune and deal with those who sought to undermine it.

The second moment centred on a bowl of salty porridge. Breakfast, on the second day of the Commune was all but inedible. While this was simply a blunder on behalf of the caterers, it was not seen as such. The dissenters used the opportunity to argue that it was a signal from us, the experimenters, that we would not tolerate an egalitarian system and that, if the Communards tried to persist, we would in effect starve them into submission. For the Communards themselves, the situation began to look hopeless: they faced internal dissent that they didn’t know how to deal with without asserting coercive power (something they were unwilling to do); they also faced external dissent which they felt was unchallengeable. As one supporter put it, they now had all the responsibilities and efforts of making a system work and yet still it seemed unworkable.

The third moment was when the dissenters took centre stage. After the breakfast incident, and seeing the demoralisation of the Communards, they called a meeting of all participants. The supporters of the Commune sat round a long table, the dissenters stood at the front and the lead figure (PB) berated the assembly in foul and violent language. He started by invoking the breakfast incident: “We’re eating fucking shit because the regime of yesterday [i.e. the Commune] doesn’t work. Irrespective of your fucking beliefs, a little bit of moderate force does work”. PB went on to threaten the Communards with further disruption, but also to offer himself as a bulwark against disruption, to berate them for their failure to stand up for themselves, but also to offer them praise:

“You shithouses I said it, didn’t I? You didn’t have the balls. These guys [i.e., the other dissenters who were centrally involved in disrupting the Guards’ regime] have got fucking balls because they put their money where their mouth is and did something about it. And I know you’re sitting there quietly because to a lot of you the words I’m saying do make sense. And I find standing here that I have more in common with these two than you lot. I hate to say, you’re fucking great guys, but you’re arseholes. You’re alright while you’re fucking eating the food and drinking the beer, but when it comes to payday, get rid of the fucking surplus men. That’s what it’s down to, isn’t it?”

In effect, he both threatened them with patriarchal violence and offered them patriarchal protection. This offer was then made explicit: revert to the previous system but with the Communards as Prisoners and the dissenters as Guards. The Communards certainly didn’t embrace this suggestion, but it is striking that, with one sole exception, they did not protest, instead sitting quietly, looking despondent, some with their heads in their hands or on the table. PB finished by reminding everyone of the impasse they faced: “Right, I’m ajourning then. Give me a shout when you have a bad fucking dinner”.

Soon after this, the study came to an end. The Communards admitted that they had lost faith in their ability to run an effective system and that their faith in democracy had diminished. This was matched by psychometric evidence suggesting that their authoritarianism had increased to a point where it was statistically indistinguishable from the authoritarianism of the dissenters. It is worth underlining the importance of this finding for it suggests that those whose original ambition was to create an egalitarian social system had by the end of the study become as authoritarian as those who wanted to disrupt that system and implement a harsh hierarchy. Lest anyone miss the wider resonance of what they proposed, these would-be “new guards” had requested a uniform of black shirts, black berets and black sunglasses. Even if they would not voluntarily allow themselves to be locked (literally) into the new tyranny, they had lost the will to defend the old democracy.

In analytic terms, we suggest that the slide towards authoritarianism derived from the failure of a democratic group which in turn stemmed from the unwillingness of group members to assert group power in defence of their group values. These were the conditions under which group members began to experience their democratic rights and responsibilities as a burden. These were the conditions under which an authoritarian discourse centred upon submission to patriarchal power became credible. These were the conditions under which the offer of patriarchal protection became attractive.

In effect, what we are arguing is that authoritarian leadership becomes attractive under conditions where alternative systems fail to create a viable order. Some of those conditions of disorder are produced by the authoritarian leadership itself: they simultaneously organise social conflict and promise to keep it under control (an approach that was central to the rise of the Nazi Party; see Bendersky, 2000). Some of those conditions arise out of the inadequacies of alternative leaderships. Indeed the great tragedy of our Communards was that, in avoiding the exercise of power for fear that it would turn them into tyrants; they helped create the conditions under which tyranny could flourish.

Perhaps this is a good place to recall the closing words of Berthold Brecht’s parable of Nazism: The Resistable Rise of Artuo Ui:

If only we could act instead of talking,

We wouldn’t always end up on our arse.

That was the thing that nearly had us mastered;

Don’t yet rejoice in his defeat you men!

Although the world stood up and stopped the bastard,

The bitch that bore him is in heat again Footnote 2

Authoritarian leadership becomes attractive, then, not because of deficiencies of the psyche but because of deficiencies of action. And far from avoiding groups and power (as Zimbardo, 2006, counsels) it is essential to exercise group power in order to enact democracy. As long as we fail to understand this, the conditions for tyranny to reproduce itself will be assured.

Conclusion

Our argument in this chapter can be simply summarised: strong leaders are not necessarily tyrants just as strong groups are not necessarily tyrannies. To believe that they are is not just a wrong-headed understanding of the problem, it is part of the problem. For tyranny is fundamentally a matter of the specific content (the values and norms) of specific groups, and of a specific relationship between leaders and ordinary group members in defining that content. Democracy requires strong effective leaders involving group members in open debate as to how to sustain strong, inclusive and equitable groups. Nelson Mandela is a good case in point. His willingness to don a Springbok shirt (previously a symbol of apartheid privilege) at the final of the 1995 Rugby World Cup was critical in bringing Whites into the “Rainbow Nation” and marginalising those supporting conflict against majority rule (Carlin, 2008). It follows that those who undermine the possibility of strong democratic leadership—either in theory or in practice—play the tyrants’ game for them.