I often say that the first step in demagoguery is the reduction of politics to identity. And I’m often understood to be making an argument that is very different from what I’m trying to say. It’s important to understand that I’m talking about in-groups and out-groups from within social group theory. So, the “in-group” is not the “group in power.” It’s the group someone is in.
If you meet a new person, and ask them to describe themselves, they’ll typically do it by listing whatever happens to seem to be the most relevant social groups they’re in (their “in-groups”): Christian, Irish-American, Texan, teacher. If I were at a conference of teachers, it would be weird for me to say that I’m a teacher, since everyone there is (it isn’t information anyone needs), and that I am Irish-American would only be irrelevant. I’d list the in-groups most salient for that setting.
We all have a lot of in-groups; our membership in those groups is a source of pride. We also tend to have at least some out-groups. Out-groups are groups against which we define ourselves—we are proud that we aren’t in them. They can get pretty specific. I’ve mentioned elsewhere that my kind of Lutheran (ELCA) often takes pride in not being that kind of Lutheran (e.g., Missouri or Wisconsin synod); college rivalries are in-/out-group; fans of a band often take pride in not being the losers who are fans of that band (or kind of music).
There are two ways I’m often misunderstood when I say that the first step in demagoguery is the reduction of politics to in-group/out-group. The first is that, since I’m saying that social groups are socially and rhetorically constructed, people think I’m saying that social groups have no material reality, and that would be a stupid thing to say. Being a cancer survivor is a very real and material identity. Even categories that are purely socially constructed with no basis in biology (the notion of “Aryans” v. Central or Eastern Europeans) had the very real and material consequences of Hitler’s serial genocides. I’m saying that there aren’t necessary and inevitable connections among social group, material conditions, and how the groups are constructed. What it means to be a “cancer survivor” varies from one culture to another (whether it’s a point of pride or shame, for instance)—that real and material identity doesn’t necessarily or inevitably lead to a specific social group or political agenda.
Second, I’m often understood to be arguing for some Habermasian/Rawlsian identity-free world of policy argumentation in which arguments (and not people), like autonomous mobiles in space, engage with one another. That kind of argumentation is neither possible nor rational.
Of course our identity is relevant to our argument; it’s one of many things we should consider. For instance, that someone is a cyclist means that they can give useful information about what feel like the safest places to ride a bike where they live. That’s relevant information because they’re a cyclist. My opinion about what are the safest places to ride is not relevant because I’m not a cyclist. Unless I’m a traffic engineer who has a stack of studies about accidents in the city. The traffic engineer (who may or may not be a cyclist) and the cyclist have views that should be considered. Neither one is necessarily right.
Thinking about politics in terms of social groups become toxic when we think those groups are discrete (you’re either in one group or another) ontologically grounded categories (meaning that we think we know everything we need to know about an individual when we categorize them into a social group). That notion that, once I’ve put you into a social group I know everything I need to know about your motives, beliefs, politics, and moral worth (you’re a teacher, so you’re a liberal elitist who supports Biden because he’ll increase teacher salaries and you’re greedy). You might really be a cancer survivor, teacher, cyclist, or traffic engineer, but once I know your membership in any of those groups, I don’t immediately know everything about you.
Identity politics is healthy when it is about acknowledging that we have a system that privileges some social groups over others, that some social groups might be possible to ignore (a person could have a long and happy life without ever understanding the distinction between Missouri and Wisconsin Synod Lutherans) but that some are so interwoven into community identity and political rhetoric you can’t not see them (such as “color” in the US), that there are real material conditions of being identified as belonging to some groups versus others, that claims about groups are generalizations that may or may not apply to specific individuals because of overlapping group membership, that overlapping group identities mean that membership in a specific group that guarantee identical experiences (intersectionality).
Those approaches aren’t ways of thinking about identity and its relationship to politics that contribute to demagoguery.
While it’s probably cognitively impossible not to be strongly influenced by notions of in-group, not everyone does so in the same way. In-group identification seems to require some notion of out-groups (or at least non-in-groups). We’re only aware of the boundaries of the in-group (the line that marks “in” so to speak) if there are boundaries, and that means at least the possibility of being outside those boundaries. There must be non-in-group members for there to be an in-group. There also must be groups of people who are outside those boundaries—out-groups. We tend to define ourselves by not being out-group.
What varies is how much hostility we feel toward non-in-group members, whether we group them all as one out-group, and whether we narrate ourselves as in a zero-sum battle. I might take pride in being ELCA and believe that that group has better theology than Missouri Synod, but that pride in my in-group doesn’t require that I feel threatened by members of the Missouri Synod; it doesn’t mean I believe that it is bad for me if something good happens to them, or that it is good for me if something bad happens to them (zero-sum).
When we think in terms of zero-sum, we fail to see ways that we might have shared interests, values, or goals with an out-group or some of its members. We will settle for policies that hurt us, as long as they hurt the out-group; we deny goods to the out-group, even if their getting those goods might benefit us.
So, when I say that we shouldn’t reduce politics to questions of identity, I don’t mean that consideration of identity is always a reduction, but it is a reduction when we assume that there are only two identities, that they are internally homogeneous, and they are inevitably in a zero-sum relationship with each other.