I am defensive about the virtues of big cities. After moving to New York in my early 20s, I came to strongly identify as a city-dweller. When someone points out the downsides of big cities, I retrench. I try to find problems with their arguments against cities or any studies they cite. I reinterpret the points they make, try to show that they only apply in specific circumstances, or that they are not really disadvantages. As a result, I can keep my many rose-tinted beliefs about cities. In turn, this enables me to remain convinced that a city-dweller is the Cool Thing To Be – exciting, attractive, fun, and virtuous.

This is a case of identity-protective reasoning. To protect my identity as a city-dweller, I reason in ways that help me defend the goodness of cities.

We all do this. And often, it is harmless enough. As far as I can tell, the worst outcome of my defensiveness about cities has been bickering with my suburban parents. In fact, counter-intuitively, identity-protective reasoning can be good. If you are constantly exposed to misleading evidence against true beliefs, it can be good to be defensive. For example, we are bombarded with misleading evidence that suggests that Black people are criminals, queer people are mentally unstable, and working-class people are lazy. Having a proud Black, queer, or working-class identity can motivate people to scrutinise this evidence, rightly rejecting its implications.

But in many other cases defending aspects of one’s identity is wrong-headed. It can lead us to maintain false and harmful beliefs in the face of strong evidence against them. For example, identity-protective reasoning sometimes sustains beliefs that vaccines are dangerous. For some people, this belief is tied up with their core identities – as Republicans, anti-establishment hippies, or conservative mothers. Protecting these identities involves defending anti-vax beliefs against the evidence. Similarly, protecting a white racial identity involves resisting evidence of racial privilege. Protecting masculinity may involve resisting arguments against meat-eating. And protecting national identity may involve explaining away evidence of the country’s past atrocities.

So identity-protective reasoning can have positive or negative effects: it can help us track the truth or get us very far from it. How can we avoid falling prey to the bad kind of identity-protective reasoning, and harness its positive powers?

The key here is to notice that the goodness of identity-protective reasoning depends on the identities we are defending. And our sense of self – the identities that matter to us – can change, shifting what beliefs we defend. My move to New York made me more of a city-dweller, turning a moderate preference for cities into deep-seated pride. But I acknowledge (begrudgingly) that this could change. If I were to move to a small town and find community and joy there, I doubt that I would still think of myself as a city person. And I doubt that I would defend the virtues of cities against all arguments. In fact, I might do the opposite. I might undergo a transformative experience, one that changed my fundamental values. As a result, I might become the kind of person who extolls the virtues of fresh air, peace and quiet, and open spaces.

The central point is this: when our circumstances change in ways that change our core identities, we cease to defend beliefs connected with those identities. We become open to a wider range of evidence on those topics – and to changing our minds on them. On the flipside, adopting new social identities can make us close-minded in ways that protect our true beliefs against bad evidence.

If our identities were completely outside of our control, this would be a largely idle observation. But importantly, it is in our power to change our sense of who we are. Here I want to focus on the effects of changes in social networks and communities.

Changes in social networks matter to the evidence we have. By talking with many different people, we get access to a wide range of considerations bearing on our beliefs. Conversely, social networks in which we only interact with like-minded people – epistemic bubbles – only provide evidence that supports our beliefs.

That much is familiar. What is less-often noted is that our social networks and communities also affect our sense of self. And, as we have seen, this in turn affects what we are defensive about. Some communities – call these identitarian communities – might make us very close-minded about beliefs related to the identity the community inculcates.

These effects of social networks – on what evidence we have, on the one hand, and on what topics we are defensive about, on the other – come apart. If we want to pop an epistemic bubble, all we need to do is offer its members access to a wider range of evidence. But merely offering more evidence will not persuade members of identitarian communities. They will simply resist that evidence. To persuade them, we need changes in social networks that restructure their sense of who they are. Which kinds of changes to social networks can do this?

Interacting with people who do not share our central social identities can make us notice new commonalities with them. This can de-stabilise our sense of self and lead us to adopt and value new identities. At least, this can happen when contact involves mutual respect, curiosity, and cooperation.

Let’s take a prosaic example. In successful company mergers, people stop identifying with the smaller company they used to work for. In its place, they adopt a new identity associated with the larger company. This shift in identity makes a difference. Employees in corporate mergers find their interactions less stressful once they adopt a shared identity.

New identities can completely erase existing affiliations, as in the corporate merger case. Or they can function as add-ons to one’s existing sense of self. This often happens when students enter university and add that identity to others.

In either case, a new, over-arching identity makes you less motivated to single-mindedly defend your previously dominant identity. You now need to balance preserving various aspects of yourself – as a student, religious person, and member of a certain ethnic group, for example. In other cases, you no longer need to defend your previous identity at all. Without the defensive motivation, you have less need to resist evidence against beliefs related to that earlier identity.

Indeed, there is evidence that cross-group contact reduces one especially toxic form of defensiveness: partisan motivated reasoning. This happens when partisans reason in ways that help them defend their political identities (e.g. as Republicans or Democrats). By shutting off substantive debate, this causes serious trouble for collective deliberation.

One cause for the increase in partisan motivated reasoning is increased social sorting. As political scientist Lilliana Mason has shown, most partisans now live in partisan bubbles. They barely ever interact with supporters of the other party. As a result, they have ceased to have cross-cutting social identities. As Democrats and Republicans share fewer and fewer (non-political) identities, they grow more invested in defending their partisan identities grows. In an important sense, their entire sense of self stands and falls with their partisan identity.

If social sorting is the problem, cross-group contact might be the solution. In support of this suggestion, there is evidence that Democrats who share important identities with Republicans, and vice-versa, are less likely to engage in partisan motivated reasoning. If we want to reduce partisan motivated reasoning, we need to build networks that make room for shared identities among Democrats and Republicans.

Creating melded, over-arching identities through cross-group contact can reduce bad defensive reasoning. But it will not be enough. We also need to ensure that there are positive social identities that people can adopt to replace the ones they abandon. And we also need social identities that function as a defence against misleading evidence, for instance about the supposed inferiority of marginalised social groups. Here, alternative sub-communities and social movements play a key role.

Alternative sub-communities and social movements often creatively develop new social identities or reimagine existing ones. Sometimes this is an explicit goal of the movement. For example, Martin Luther King Jr. framed the Montgomery bus boycott in 1956 as the source of an identity shift to what he called “the new Negro” – “a person with a new sense of dignity and destiny, with a new self-respect.” Other times, new identities are a side effect of shared projects. Working together toward an important shared goal generates opportunities to notice commonalities. This often generates new identities. In this way, for example, labour activism often gives rise to activist or worker identities.

In addition to creating new identities, social movements and alternative communities inculcate these identities in members. By participating in new communities, you might add new identities to your self-conception. For example, you might come to think of yourself as an activist and not as a passive victim. Or you might come to centre some identities more and others less. For example, as many white women in the 1970s began to think of themselves as feminists, they also came to identify less as housewives.

New social identities change how we reason. Ceasing to identify with whiteness or masculinity as a by-product of union activism can make one less defensive around issues surrounding race and gender. Adopting a proud Black, feminist, or queer identity, can make one more sensitive to injustice and less motivated to defend the status quo.

Identity-change is a powerful motor. It alters which views of the world we leave vulnerable to the evidence, and which ones we cherish. And, as we have seen argued, identity change is often driven by change in social networks. Changing minds requires much more than good arguments. It requires reshaping the social structures in which we interact.

Does this leave any room for philosophical reasoning?

Philosophy is often taken to make two key contributions to changing minds. The first is producing rigorous arguments. But we have seen that, by itself, even the most solid argument is unlikely to change defensive minds. The second is cultivating skills in logical reasoning and fallacy detection. But a defensive reasoner will use whatever tools they have to defend their identity. Critical thinking tools may make them better at viciously protecting their misguided views against good evidence. So perhaps philosophy is at best irrelevant, and at worst actively damaging.

This is too pessimistic. Critical thinking and good arguments on their own do not suffice. But they still matter. If we want people to change their minds for good reasons, we need valid arguments and sound reasoning. But we also need to make sure that people are receptive to those arguments. Changing social connections sets the stage for that reception.

As it turns out, philosophy can also help us set the stage. Philosophy is not merely critical. As Jennifer Morton argues, it is also an imaginative activity. Plato called on us to imagine that all we see are shadows on the wall of a cave; Descartes that everyone around us is an automaton; Rawls that we stand behind the veil of ignorance, waiting to know what our lot in life will be. In asking us to consider these scenarios, these philosophers call on us to realise that there is nothing inevitable about the way things are. In doing so, they invite us to envisage and explore alternative ways that the world could be. Likewise, by using the philosophical imagination, we can come to see that there is nothing inevitable about our social connections – or, more disorientingly, about who we take ourselves to be.

Finally, philosophy can help us build new identities out of the rubble. This is because philosophy involves perspective-taking. In doing philosophy, we strive to understand others’ reasons. We try on their values and way of looking at the world. This helps us achieve what Hannah Arendt calls an “enlarged mentality”. We become better at accessing new standpoints. And this is crucial to forming new shared identities with others.

At a collective level, then, philosophy can make space for new experiments in living. We can envisage new ways of relating to one another, new aspects of ourselves to nourish, and new identities to orient us. This can make us less defensive about beliefs that pose obstacles to social change and sharpen our eyes to injustice. Ultimately, it might get us closer to a just world.

Carolina Flores is a Ph.D. candidate in Philosophy at Rutgers University. Her interests include philosophy of mind and social philosophy. She is currently finishing a dissertation on belief and resistance to the evidence.