Introduction

Deliberation’s effectiveness as a method of conflict resolution and democratic decision making is often seen as stemming from the persuasive power of the forceless force of argument. But can an approach to deliberation based on persuasion successfully resolve contemporary conflicts related to partisan polarization, conspiracy theories, or COVID-19, which may be deeply connected to citizens’ social identities? Results from social psychology suggest that, in conflicts such as these, approaches to deliberation based on persuasion and reason-giving may be unproductive, due to the motivated reasoning we unconsciously deploy in response to arguments that threaten our social identity. In these situations, a deliberative process based around co-creation rather than persuasion may be more likely to produce positive deliberative transformations, while still adhering to core deliberative norms.

In this paper I first discuss the importance of social identity to many of the contemporary conflicts that deliberation must help to resolve if it is to serve as an effective basis of democratic legitimacy. I also argue that modern conceptions of deliberation tend to see persuasion by the force of argument as the primary mechanism by which deliberation can induce belief change among participants. I then show that when the conclusions of an argument threaten participants’ social identity they are likely to engage in motivated reasoning, which inhibits the ability of any argument to induce belief change. I argue that Mary Parker Follett’s (1918, 1924) conception of integration, which can be seen as an alternative approach to deliberation based on co-creation, may be more productive than persuasion-based approaches in conflicts closely related to social identity. I illustrate this by contrasting deliberative approaches based on persuasion and co-creation to the contemporary conflict between vaccine advocates and members of the vaccine hesitancy and refusal (VHR) community.

Social Identity and Deliberative Democracy

Social identity theory argues that identifications with social groups are important components of our self-esteem (Tajfel & Turner, 1979). We can identify with different groups simultaneously, but we may identify more strongly with some than others, and the salience of a particular identity can wax and wane depending on the context. The increasing relevance of social identity to deliberative democracy is evidenced by the central role that social identity appears to play in many of the most contentious contemporary political conflicts. Partisanship, which has long been seen as a key barrier to deliberation (Dryzek, 2017, p. 613; Owen & Smith, 2015, p. 229) is itself a powerful form of social identity (Green, Palmquist, & Schickler, 2002; West & Iyengar, 2020), and research argues that the increasing contentiousness of the US political environment is not due to increasing polarization with respect to particular issue positions, but rather the increasing emotional salience and affective polarization of partisan social identity (Abramowitz & McCoy, 2019; Iyengar, Lelkes, Levendusky, Malhotra, & Westwood, 2019; Mason, 2018). The waxing importance of social identity in the political sphere has also been implicated in the rise of ‘post truth politics’ and political conspiracy theories, as well as declining trust in scientific evidence in favor of identity-supporting beliefs (Dryzek et al., 2019; Sternisko, Cichocka, & Bave, 2020). The epistemological implications of social identity may also strongly impact behaviors and attitudes related to public health crises, like the COVID-19 pandemic. Past research has found that skepticism of vaccines and mainstream medical science are often driven not by ignorance, but rather by totalizing belief systems that form a central part of skeptics’ social identity (Attwell & Smith, 2017; Ward, Attwell, Meyer, Rokkas, & Leask, 2017). Similar dynamics are likely to underlie some of the public skepticism surrounding efforts to minimize the spread of COVID-19 or the safety and efficacy of newly developed vaccines (Cruwys, Stevens, & Greenawa, 2020).

Regardless of how or where deliberative democracy is instantiated—be it within legislative bodies (Goodin, 2008), in minipublics or deliberative town halls (Neblo, Esterling, & Lazer, 2018), through ‘participatory constitutionalism’ (Lafont, 2020), or in the public sphere of media discourse (Habermas, 2006)—its utility as a basis for democratic legitimacy in contemporary society depends on its ability to successfully resolve conflicts like these that are closely related to social identity. If, by contrast, deliberation were only effective in conflicts where the topics being debated had little or no implications for the social identities of participants, then it would be ill-suited for solving the ‘crisis of democracy’ that many societies have found themselves facing (Dryzek et al., 2019). It is therefore critical to understand the ways in which social identity impacts the success or failure of the deliberative process. Recent work has already started to investigate these questions, arguing that increasing the salience of social identity in deliberation could prevent groupthink or polarization (Batalha, Niemeyer, Dryzek, & Gastil, 2019) and that opposing social identities can inhibit the acceptance of deliberative norms (Strickler, 2018). However, deliberative theory has yet to consider how the psychological consequences of social identity threat may inhibit what many deliberative theorists see as the central mechanism for deliberation-induced belief change: persuasion through the forceless force of argument.

Deliberation and Persuasion through the Forceless Force

Although deliberation can serve many functions in society, and can produce many beneficial outcomes, its importance for democratic theory stands or falls on its effectiveness as a method of conflict resolution and decision making (Bächtiger, Dryzek, Mansbridge, & Warren, 2018; Goodin, 2017). To fulfill these functions, a deliberation must necessarily lead to some form of belief change among at least some of the participants. Although deliberative theory has moved away from consensus as an explicit goal of deliberation, there is still wide agreement that deliberation ‘would have no point if it did not produce change in the views of at least some participants’ (Mansbridge et al., 2010, p. 78). Alternately, as Goodin (2000, p. 88) puts it, ‘people change their minds over the course of the deliberation (it would hardly be a genuine deliberation at all if they did not, at least sometimes).’ Of course, the transformations produced by deliberation may not involve the particular positions being debated. Rather, participants may be convinced, in the course of deliberation, to revise their ‘meta-beliefs’ about the legitimacy of the values being articulated, the epistemic standards that should be applied, or the range of acceptable alternatives (Dryzek & Niemeyer, 2006). Nevertheless, even in these situations, deliberative success still requires that participants exit the deliberation with different meta-beliefs than they entered with. It is for this reason that deliberation’s ability to actually produce belief change under different circumstances is such a pressing question within deliberative theory (Fishkin & Luskin, 2005; Mackie, 2006; Zhang, 2019) and that documented examples of deliberative transformation are cited as rebuttals to claims that deliberation is unrealistic or ineffective (Dryzek et al., 2019).¹

However, because deliberation is, by definition, non-coercive, it must possess one or more mechanisms for inducing participants to change their beliefs in the absence of coercive force, including the physical inducements deployed in bargaining. Although there are a number of possible mechanisms by which deliberation could induce belief change, current understandings of deliberation have tended to focus almost exclusively on what Neblo et al. (2018, p. 86) refer to as the ‘leading idea’ of deliberative democracy: persuasion via the ‘forceless force (zwanglose Zwang) of the better argument’ (Habermas, 1996, p. 306).²

Persuasion, in this context, refers to a process by which an argument produces a forceless (i.e., non-coercive) force of sufficient strength to induce participants in a deliberation to change their minds. The centrality of persuasion to modern understandings of deliberation can be seen in efforts to clarify how deliberation differs from other forms of communication, such as when Mansbridge et al. (2012, p. 18) distinguish deliberation from coercion by noting that ‘deliberation is about genuine persuasion, not pressure.’ Dryzek (2005, p. 224) likewise sees deliberation as distinguished from agonism because deliberation is ‘oriented to persuasion.’³ Even when the term itself is not explicitly invoked, some form of persuasion is implicitly endorsed as the primary deliberative mechanism for belief change whenever deliberation is defined as primarily about the giving and evaluation of ‘reasons’ for or against particular positions (e.g. Dryzek, 2017, p. 612; Goodin, 2000, p. 81; Owen & Smith, 2015, p. 228).⁴ Critically, these definitions often tacitly assume that deliberation is mainly (or exclusively) concerned with debating and justifying pre-existing ideas that participants had already formulated prior to the deliberation, as opposed to the creation of genuinely new ideas within the deliberation itself.⁵ In formal mathematical models of deliberation, this assumption—that deliberation involves adjudicating between a set of pre-formulated alternatives—is often stated explicitly (e.g. Chung & Duggan, 2020; List, 2018). In this conception, then, deliberation can only produce belief change insofar as participants are persuaded by a sufficiently strong argument to reject positions or beliefs (including, of course, meta-beliefs about epistemology or values) that they previously held or to adopt the pre-existing positions or beliefs of their opponents.

The strength of an argument or reason is not, of course, decided by reference to some a priori standard for rationality, but by criteria established within the deliberation by participants themselves.⁶ Deliberative theory has also begun to move away from a focus on rational justification and now tends to see storytelling, humor, ‘testifying,’ and many other forms of communication as permitted ‘reasons’ in a deliberation (Bächtiger, Niemeyer, Neblo, Steenbergen, & Steiner, 2010; Dryzek, 2000). Successful deliberative persuasion merely requires that, by whatever standard of evaluation a participant is employing—logical coherence, emotional appeal, humor, or something else—an argument (which could include a compelling story, joke, or testimonial) is judged to be sufficiently strong to compel belief change.

For conceptions of deliberation based on persuasion and reason-giving, successfully resolving conflicts is therefore dependent on the power of persuasion to induce belief change via the forceless force of argument. Deliberation is thus less likely to successfully resolve conflicts when—regardless of what types of reasons are permitted or provided, or how these reasons are evaluated—the processes of argumentation and persuasion in general lose their ability to induce belief change. Unfortunately, empirical research in social psychology suggests that is precisely what can occur in response to social identity threat.

Social Identity Threat and Motivated Reasoning

Given that our sense of self-identity is constructed in reference to the groups we identify with (Tajfel & Turner, 1979) it makes sense that information or arguments that threaten the perceived value of these groups could represent an existential threat to our own self-identity (Breakwell, 2015). Research has found that in response to social identity threat we unconsciously engage in various defensive mechanisms to minimize the damage to our self-esteem, especially if we strongly identify with the group in question (Branscombe, Ellemers, Spears, & Dooje, 1999). These defense mechanisms can include not only heightened in-group affect and out-group hostility but also various forms of biased information processing (Cohen, Aronson, & Steele, 2000; Nauroth, Gollwitzer, Bender, & Rothmund, 2014; Nauroth, Gollwitzer, Kozuchowski, Bender, & Rothmund, 2017). Most salient for deliberative democracy, however, are repeated empirical findings that threats to social identity lead individuals to engage in motivated reasoning (Cohen et al., 2007; Dalton & Huang, 2013; Hoog, 2012; Slothuus & Vreese, 2010).⁷

When engaged in motivated reasoning, we evaluate arguments not with an orientation towards accuracy but with a goal of maintaining our existing values, identities, and political preferences (Kahan, 2015; Kunda, 1990; Taber & Lodge, 2006). This poses a challenge for efforts to induce belief change through persuasion when the beliefs we are motivated to maintain are themselves the subject of argument. Motivated reasoning undermines the ability of better arguments to induce belief change by reducing the perceived quality of arguments that would compel us to change our minds with respect to the belief in question. This effect is clearly not limited to any particular type of argument or standard of evaluation. When subject to motivated reasoning, I will evaluate any argument against my preferred position as weaker precisely because accepting it would force me to change my mind.

Even more troubling, perhaps, is the power of cognitive biases like motivated reasoning to prevent participants from engaging in productive deliberation even if they wish to do so. Discussions about the deliberative capacity of individual citizens often concern the question of whether they are willing to change their minds in the face of persuasion,⁸ implying, perhaps unintentionally, that the success of deliberation depends on whether citizens choose to have an open mind and consider opposing arguments fairly. But research has repeatedly demonstrated that motivated reasoning and other forms of biased information processing operate unconsciously (Kahan, 2015; Taber & Lodge, 2016) and are usually invisible to those they affect. Even when we are made aware of the pervasiveness of motivated reasoning, and can identify it in others, experimental studies have shown that humans possess a ‘bias blind spot’ that prevents us from noticing when we are engaging in it ourselves (Kukucka, Kassin, Zapf, & Drorc, 2017; Pronin, Gilovich, & Ross, 2004; Pronin, Lin, & Ross, 2002). I may honestly and deeply believe that I am interpreting opposing arguments charitably, and that I would be willing to revise my beliefs in the face of a sufficiently compelling argument, but the unconscious operation of motivated reasoning may prevent this from ever actually occurring in practice.

Partly for these reasons, past work has already warned that motivated reasoning represents a serious threat to deliberative democracy (Richey, 2012; Rosenberg, 2014). However, although it is clear that motivated reasoning can present deliberation from changing minds, it is equally clear that it does not always do so. It is not infinitely powerful; opposing evidence can sometimes compel belief change even in the face of motivated reasoning (Már & Gastil, 2020; Redlawsk, Civettini, & Emmerson, 2010). Furthermore, research confirms that group deliberation can produce belief change even in regard to contentious topics in politically polarized societies (Barabas, 2004; Neblo et al., 2018), although it is also true that certain beliefs are more resistant to change than others (Zhang, 2019).

However, the fact that motivated reasoning serves as a psychological defense mechanism for social identity threat implies that it is especially likely to inhibit deliberation’s ability to productively resolve many of the pressing contemporary conflicts related to partisan polarization, conspiracy theories, and the reliability of mainstream science, all of which are closely connected to social identity. Even if participants in a debate over one of these topics avoid challenging each other’s social identities directly, they may still experience social identity threat in response to arguments over other beliefs that they view as having implications for the value of a social group they strongly identify with. Because revising these beliefs would threaten their social identity, we can expect participants to unconsciously deploy motivated reasoning to avoid having to change their minds, even in the face of what they might otherwise consider convincing evidence. For example, if I strongly identify as a libertarian, then my self-esteem is partly dependent on the (perceived) truth of certain beliefs that I view as an inherent part of what it means to be a libertarian (e.g., that reducing government intervention in the economy will tend to increase social welfare). A good argument that this belief is actually false would therefore represent a serious threat to my social identity and self-esteem. Through motivated reasoning I can construct a justification for rejecting this argument, thereby preserving my self-esteem, but in doing so, I forestall any possibility of it changing my mind.

Social identity threat may similarly prevent deliberative persuasion from achieving ‘meta-consensus’ (Dryzek & Niemeyer, 2006). For example, in an epistemic meta-consensus, participants may still disagree about the truth of various claims but are operating from within a similar epistemic paradigm (Kuhn, 1970), which allows for common standards of evidence and reasonableness. Achieving an epistemic meta-consensus requires that, for example, participants agree that claims backed only by ‘partisan’ sources should not be admitted into the debate (Dryzek & Niemeyer, 2006, p. 646). However, perceptions of whether a source is partisan or credible are themselves influenced by social identity. When scientific evidence affirms social identity it is more likely to believed, but when evidence from the same source threatens social identity, not only are individuals less likely to believe it, but they also lower their opinion of the credibility of the scientists who produced it (Nauroth et al., 2014; Nauroth et al., 2017). Similarly, epistemic paradigms themselves (e.g., a distrust of mainstream media (Schulz, Wirth, & Müller, 2020) or skepticism of ‘western science’ (Ward et al., 2017)) can serve as key components of social identity. Attempts to persuade individuals to reject these epistemic frameworks in order to achieve epistemic meta-consensus may therefore still provoke social identity threat and motivated reasoning.

We have therefore identified an important weakness in conceptions of deliberation that treat reason giving and persuasion by the force of argument as the sole or primary mechanism for inducing belief change. In a deliberation where salient beliefs and opinions are strongly connected to participants’ social identities, motivated reasoning may prevent any argument from being perceived as strong enough to induce belief change. Given that conflicts related to social identity are among the most pressing and contentious in contemporary society, it is worth considering whether, in these situations, there is some other mechanism by which deliberation could induce belief change.

One possibility involves relaxing the often tacit assumption that deliberation is primarily concerned with adjudicating between pre-existing beliefs, because it is the pre-existing connection of these beliefs to participants’ social identity that is the source of the difficulty. If deliberation instead led participants to jointly develop new beliefs, then this would represent a way in which participants’ minds could be changed without them having to reject any pre-existing beliefs that might have implications for their social identity. Fortunately, recent years have seen a growing interest in a method of dialectical conflict resolution that can be seen as an alternative approach to deliberation based on co-creation rather than persuasion (Bartels, 2015; Mansbridge et al., 2010; Nelson, 2017; Stout, 2019; Stout & Love, 2017; Wright, 2019b). This is the idea of integration, developed by the early 20th century American pragmatist philosopher, organizational theorist, and political scientist Mary Parker Follett.

Integration: Conflict Resolution through Co-creation

Follett defines integration as an alternative to the two usual methods of resolving conflict: domination, viz., victory of one side over the other, and compromise, which, in Follett’s view, merely represents a temporary suspension of the conflict that does not resolve the core desires of either side (Follett, 1942). For Follett (1924, p. 208), neither of these approaches can serve as the basis of democratic government and decision making, because they do not reflect the genuine ‘will of the people.’ In integration, by contrast, participants work together to jointly develop a new win–win solution that addresses the underlying desires of each side (Mansbridge et al., 2010). This new solution could not have been formulated by either side in isolation because it reflects the interweaving of individual positions into a new group thought. Integration is achieved by first shifting discussion from the particular issue being debated (which may indeed represent a zero-sum game) to the underlying desires motivating the disagreement, and then searching for a resolution of these desires that is orthogonal to the original axis of conflict.

Follett provides one of her many examples of this process in reference to a disagreement between a man and his live-in mother over the location of the dining room table: the man wanted the table near the window and his mother wanted it in the middle of the room. Although it is undeniable that the table itself cannot be in two places at once, Follett (1924, pp. 169–170) reframes the question, asking the man,

[w]hat did you and your mother really want? Perhaps not-table-in-window or table-in-middle-of-room at all. Perhaps what she really wanted was to have it where it would be near the butler’s pantry, where it would be easier to walk around, or where it would be near the radiator. Perhaps what you really wanted was more light, or the view of the river. The integration might have been to take down the curtains.

Although the prospect for such win–win solutions may seem remote in many contentious political conflicts, Follett (1942) outlines a series of strategies to help identify and facilitate them. These include bringing underlying differences into the open, breaking up larger concepts and problems into smaller ones, and distinguishing between the symbolic and substantive aspects of the conflict, which may be equally important but can often be dealt with separately. Using these methods, it is often possible to identify avenues for integration even between groups whose positions seem diametrically opposed (Wright, 2019a).

Critically, although it predates the development of deliberative democracy, integration can be productively analyzed as an alternative approach to deliberation that replaces the activity of persuasion with that of co-creation. Like deliberation, integration is a form of dialogue aimed at resolving conflicts and making decisions about matters of common concern through non-coercive means. Like deliberative democrats, Follett (1918, p. 180) sees integration as a mechanism for the production of democratic legitimacy that is superior to voting for both intrinsic and instrumental reasons. The primary difference between integration and contemporary accounts of deliberation is, of course, integration’s rejection of persuasion in favor of co-creation. For Follett, persuading my opponents to reject their own beliefs, or to adopt mine, would represent an undemocratic domination, even if it were achieved through the forceless force of argument.⁹ Because Follett (1942, pp. 30–31) sees conflict as analogous to friction—something that often needs to be managed and controlled but is also a critical source of power for human advancement—she sees the goal of democracy not as eliminating conflict but rather productively channeling conflicts intro creative solutions. Thus, in Follett’s view, a method of conflict resolution that does not make use of the conflict for creative purposes is not genuinely democratic (Follett, 1924, p. 209).

Notwithstanding this key difference, integration has the potential to equal, or in some cases exceed, the instrumental benefits produced by persuasion-based conceptions of deliberation. For example, integration may be better able to harness the ‘error canceling’ power of diversity that serves as the foundation of most epistemic justifications for deliberation (e.g. Bohman, 2006; Estlund, 2008; Hong & Page, 2004; Myers, 2018). This is because, compared to persuasion, co-creation is less vulnerable to groupthink and expert domination (Solomon, 2006) and is explicitly focused on producing a new solution that transcends the perspective and biases of any particular actor (Wright, 2019b).

Because the goal of integration is ‘not to find the best individual thought, but the collective thought’ (Follett, 1918, p. 30), it is not focused on ‘winnowing’ out alternatives and beliefs that cannot be justified by sufficiently strong arguments (Goodin, 2017). Thus, the success of integration is not dependent on participants rejecting beliefs that have implications for their social identity, even when those beliefs cannot be justified. Rather, the goal of integration is for opponents to work together to create a new solution that did not exist before. Because this new idea is aimed at fulfilling the core desires of each side, participants should be able to adopt it without rejecting any of their pre-existing beliefs or unconditionally adopting the position of their opponents. The solution is thus not merely a preexisting belief that participants must be persuaded to accept, regardless of its implications for their social identity. Rather it is, at least partly, their own idea: something that they helped to create and which serves to advance their own interests.

Integration and Persuasion in Deliberations surrounding Vaccination

To illustrate how an approach to deliberation based on integration differs from one based on persuasion in situations of social identity threat, I analyze contemporary debates between vaccine advocates in the mainstream medical community and parents who are part of the vaccine hesitancy or refusal (VHR) movement over whether to vaccinate children against highly contagious and dangerous diseases, such as measles or COVID-19. In reality, the active participation of both groups would be necessary to generate a true deliberative solution to this conflict. However, existing research on the drivers of VHR beliefs can help us predict how these two different approaches to deliberation might produce different results.

In the US at least, VHR beliefs are not generally driven by ignorance about the benefits or risks of vaccines. VHR parents often have high levels of medical or scientific literacy (Gottlieb, 2016) and a high capacity for complex reasoning (Ward et al., 2017). Consequently, efforts to persuade these parents to change their minds by reference to scientific evidence have not been successful (Kata, 2010; Meszaros et al., 1996). Prior to the COVID-19 pandemic VHR beliefs were largely driven by broader distrust of the Western medical establishment and a commitment to a high-intensity mode of parenting that emphasizes continued vigilance and parental control over children’s lives to protect them from harm (Brown et al., 2010; Reich, 2014; Ward et al., 2017). As mentioned above, these beliefs can be strongly connected to parents’ social identity (Attwell & Smith, 2017). Research finds that, for many VHR parents, both their rejection of Western medical knowledge and their commitment to a particular mode of parenting was ‘so totalizing that it came to define their identity and sense of self’ (Ward et al., 2017, p. 7).

This poses a serious challenge for efforts to resolve this dispute via a deliberative approach centered on persuasion. The success of this approach would depend on it being possible to persuade at least some of these parents to revise their opinions about science, medicine, and parenting purely in response to a sufficiently strong opposing argument. But rejecting prior beliefs about vaccine safety, parenting style, and Western medicine would likely produce extreme social identity threat. In effect, it would entail an admission by VHR parents that they have been bad parents and put their own children’s health at risk—one of the very things that motivated their VHR beliefs in the first place.

Furthermore, because many VHR parents also see opposition to the epistemological paradigm of Western medicine in general as an important part of their social identity (Ward et al., 2017), even less ambitious efforts to attain epistemic meta-consensus (Dryzek & Niemeyer, 2006) may produce social identity threat. For VHR parents, even this milder form of consensus—accepting that it is reasonable for proponents of Western medicine to believe in results that have been verified by the methods of natural science—could threaten their social identity.¹⁰ We should therefore expect that defensive mechanisms, like motivated reasoning, will protect VHR parents from having to accept these painful conclusions (Sinatra, Kienhues, & Hofer, 2014) even in the face of compelling counter-arguments, thereby preventing deliberation from inducing any substantial change in their beliefs.

Efforts to resolve this dispute via integration and co-creation, by contrast, would recommend shifting the debate from the initial dispute to the underlying desires motivating the positions of each side. For vaccine proponents in the medical establishment, these would presumably include preventing infection by vaccine-targeted diseases and the establishment of herd immunity. For VHR parents, the research above tells us that their core desire is not to avoid vaccinating their child but rather to promote their child’s health; to engage in an active, vigilant style of parenting; and to maintain a skepticism of the Western medical establishment. Indeed, VHR parents often reject the label of ‘anti-vaccine,’ preferring to identify as ‘pro-safe vaccine’ (Kata, 2012). Thus, although the opposing positions in the original dispute (to give a child an existing vaccine or not) are mutually exclusive, this is not necessarily the case for the underlying desires of both sides.

One possible integrative solution might be for vaccine proponents to ask VHR parents for their assistance in developing an alternative, safer (according to VHR parents) version of the vaccine in question that could be offered to members of the public who are skeptical of the mainstream vaccine. There are various ways in which an existing vaccine might be modified to address the concerns of the VHR community without harming its safety and efficacy (as evaluated by vaccine proponents). One example might be the addition of homeopathic compounds. Because members of the VHR community often embrace various forms of alternative medicine, including homeopathy (Kata, 2010), the addition of such compounds might make the vaccine seem safer and better aligned with their own beliefs about health and medicine. Vaccine proponents are unlikely to share these beliefs, but they may be willing to accept the addition of these compounds, because, from the perspective of Western medicine, homeopathic compounds are so diluted that they have no biological effects at all (Grimes, 2012; Hawke, Driel, Buington, McGuire, & King, 2018; Maddox, Randi, & Stewart, 1988) and are thus unlikely to negatively impact the safety or efficacy of the vaccine.

From the perspective of the VHR community, what may be most important is not the substantive content of these changes but the fact that their community had a hand in implementing them. The resulting alternative vaccine, however closely it resembles the previous version from a scientific perspective, may be seen, from the VHR community’s perspective, as fundamentally different insofar as it is a product of the community itself. Critically, VHR parents would still be free to believe that mainstream medical establishment could not have developed a safe vaccine without their assistance, even though this belief likely could not have been justified in the course of a traditional deliberation. Proponents of Western medicine, by contrast, are still free to believe that the contributions of VHR parents to the vaccine (e.g., the addition of homeopathic compounds) had no actual effect on the safety or efficacy of the vaccine, thus preserving their own social identity as practitioners of an epistemically valid scientific paradigm. Yet, despite this continuing disagreement, both groups would achieve the actual outcome that they desired: a vaccine that is perceived as safe and effective by both the medical establishment and VHR parents.

At the same time, it is possible that this integrative process might still lead participants to eventually revise their existing beliefs. For example, the experience of working together with members of the medical establishment to solve a problem may cause some members of the VHR community to gradually reevaluate their distrust of Western medicine, perhaps even to the point where acceptance of even mainstream vaccines no longer poses a threat. This is, in fact, the mechanism by which Follett sees integration eventually leading to the revision of existing beliefs, in situations where persuasion itself is infeasible:

[t]he theory of consent rests on the assumptions that we think with our “minds” and we don’t…. How often we see cases where we have not been able to persuade people, by our most careful reasoning, to think differently, but later, by giving them an opportunity to enter on a certain course of action, their “minds” are thereby changed (Follett, 1924, p. 198).

Thus, although the rejection of existing beliefs is not a requirement of integration, it may be a byproduct.

In reality, of course, this proposed solution may be unsatisfactory to one or both sides of this conflict. Integrative solutions often seem radical and unsatisfactory, because, by their nature, they incorporate what we see as unjustified ideas from the opposing side. Follett (1942, p. 36), herself, cautions that integrative solutions are not always possible, and they may not even be desirable in all situations. However, expanding our understanding of deliberation to include integration as a possible mechanism for belief change alongside persuasion may increase the likelihood of successful deliberation in situations where social identity threat and motivated reasoning insulates participants from the forceless force of argument.

Conclusion

Some of the most contentious conflicts facing democratic societies around the world are deeply connected to social identity, which may explain why they have proven so intractable. The viability of deliberation as a basis of democratic legitimacy depends on its ability to resolve these sorts of conflicts, but this may not be possible through an approach based solely on reason-giving and persuasion. For these conflicts, an integrative approach that rejects persuasion in favor of co-creation may offer the most viable strategy for deliberative transformation. More empirical research is needed on the effectiveness of integration in practice, but the power of social identity threat recommends that this research may be of critical importance to the project of deliberative democracy.

Competing Interests

The author has no competing interests to declare.