The Identity Trap
Part 1: The Dangers of Playing on the Conservatives’ Home Field
Post 22: 27 May 2025
There are two politics of “identity” that stand in exact opposition to one another.1 On the one hand, identity is the central concept in traditional conservatism,2 for which it defines the boundaries of the community, provides the glue of belonging for those within those boundaries, and excludes the “others” caught outside of them. On the other hand, movements to expand full membership to include previously excluded groups necessarily speak in the language of identity—this is the basis of their exclusion.
This reality creates a set of political problems for movement leaders. I’ve written about this before but want to expand my argument in four related directions in this and my next post. Today, I want to focus on some of the rhetorical liabilities of using identity-based language.3 These roughly fall into two categories: the risk of accidentally strengthening the conservatives’ position and, more seriously I think, the danger of succumbing to what I’ll call the pathologies of identity-based argument. In my next post, I’ll circle back to our publication theme to argue that excessive reliance on identity-based rhetoric can slow progress on social rights by making majoritarian coalition-building more difficult. Social citizenship depends on universalism, which means that civil rights movements must ultimately get beyond balkanizing identity politics.
The Conservatives’ Home Field
The fundamental rhetorical problem is that the mere reference to identity provides a permission structure for conservatives to tap into their brand of identity-based hate. The left can’t avoid these references, but how does it find the right way to refer to identity? My goal is to strengthen the message of inclusion while reducing the vulnerability to the predictable conservative counter-message.
The conservative echo is often much louder than the original progressive signal, exaggerating the “discourse”, such as it is, beyond recognition. In the 2024 campaign, for example, we are told that Democrats over-emphasized transgender rights, when in fact the issue was hardly ever actually mentioned by the Harris campaign but repeatedly injected by Republicans. According to Truthout, Republicans spent $215 million on anti-trans ads. That’s an estimated $134 per trans person in America! So it is clear which side is fixated on gender transitions. Some have argued that the ad served not to exaggerate the trans threat but to communicate the idea that “Democrats don’t care about you and your issues”. That seems like the same thing to me. I’ve not seen an academic study of the question, but it seemed to me that the ads worked. We need to figure out why.4
The trans ad example shows how tough the problem is. Falling silent would not have helped; somewhere at some time a Democrat or Democratic ally provided the necessary fuel for this fire. The Harris statement featured in the main ad was five years old (I believe). She was criticized for ever having said anything about this, even though all she did was answer a question honestly and factually.5 The “shh” argument is tantamount to asking the Democratic Party to change its position on the issue. Not only is that unacceptable—we cannot compromise on basic rights—but it also would not work: the old YouTube clip is still out there. A more reasonable response, it seems to me, would have been to frame the issue as a very small part of the larger effort to include everyone in American society. We need to insist that Republicans tell us why everyone should not be treated equally. This can redirect an issue like this to the narrower question of what equal treatment requires—that might or might not include gender-affirming care in federal prisons—while affirming the more important principle of inclusion. Most people probably agree that the medical care offered in prison ought to be adequate, but not better. Perhaps opponents see this care as a luxury, like a facelift. Disabusing them of that notion reframes the whole question to be about real people with actual medical concerns. Even with a shift like that, this is still probably a “loss” for the left, but a much more manageable one. I also think that more effort ought to have been made, then and now, in ridiculing Republican fears of the transgender menace.6 This is elephant and mouse stuff.
Let’s consider another example. A common conservative response to the “Black Lives Matter” movement was to willfully mishear the slogan and counter with “All Lives Matter” or “Blue Lives Matter.” Explaining what the slogan meant—especially emphasizing that the overall goal is universal inclusion, not some sort of Black-first society—can only help at the margins; anyone invested in these bad faith interpretations will not be moved. That said, most people understood the slogan and it was, for a time at least, effective, at least as a tool for mobilizing Democratic voters.7 Pew polls showed overall support for Black Lives Matter peaked at 67% in June 2020 in the aftermath of the George Floyd murder. This faded but still stood at 52% in February 2025.8 This is counter to the common conservative interpretation of the 2024 election, now manifest in the Republican’s anti-DEI fervor, including against post-Floyd police reform and oversight efforts. The jury is still out, but these direct identity-based counter moves do not seem that popular.
A strength of BLM is its connection to the specific issue of police violence. At least some polls in 2020 showed broad support for changes in police policy. A 2020 Gallup poll found a clear majority (58%) in favor.9 This specificity makes it harder for conservatives to argue (or imply) that African Americans are looking to get some sort of special treatment or benefit. The problem of police violence is depicted as a “Black problem”. It definitely is, but this element of the narrative is still overplayed. African Americans are much more likely to be killed by police than White or Latino persons in the US. In 2022, an especially bad year, 4.7 out of every one million Black people were killed. That’s much higher than the rate for White people, 2.0 per million, or Latinos, 2.9 per million. So there is definitely a racial dimension to the problem. But in absolute numbers, more White people are killed each year than Black people—in 2022, 370 versus 198. Moreover, the rise in killings by police is about the same for White, Black, and Latino victims. The problem has gotten worse over the past decade for all groups. This is a major civil rights issue for African Americans—I don’t want to say it’s not—but showing that it is also a problem for everyone would make it that much easier to build support. I’ll return to this theme next week.10
The fact that the belief in equal treatment before the law is a deeply ingrained liberal value creates cognitive dissonance issues for conservatives who would like to use the criminal justice system as a social control mechanism, a way to enforce the identity boundaries of the community.11 Not wanting to specifically refer to race, conservatives typically try to shift the focus to crime, especially “urban” crime. This is, of course, the classic dog-whistle strategy. I note that conservatives have increasingly repurposed crime as an anti-immigrant alarm, now dropping the need for the dog whistle: Republicans pretty much avoid saying “Black crime”, but Trump’s “migrant crime” slogan is as subtle as a baseball bat.
The substitution of crime for race is a successful strategy. Despite a major decline in actual crime rates over time, a persistent majority of Americans always believe that crime is rising. Republicans have invested a great deal in building up this false belief and Democrats have not yet found a good way to re-ground people in reality. I note that the two series are linked at the start of the chart (I will look for a longer timeseries),12 suggesting that it is possible to regain a rational footing. I wish I had the answer.
The Pathologies of Identity
Now I turn to an even more difficult set of issues connected to identity-based language.
I want to take seriously the possibility that the left uses identity in unhelpful ways. My immediate concern is that well-intentioned actors on the left—whose main goal is to expand membership— unwittingly mimic the types of identity references commonly employed by the right—whose main goal is to limit it. I am not claiming that both sides are the same. For the left, I am referring to a pathology associated with reliance on identity, something that we should be aware of and check when we can, mostly by being self-aware. On the right, these are features not bugs. The whole point of the identity mechanisms is to foster exactly those sorts of negative or hierarchical views of others.
The telltales of this identity language are “us” and “them” juxtapositions and, more seriously, references to hierarchical frameworks. It is difficult to avoid us (good)-versus-them (evil) language when this merely mirrors the conservative message, but the left needs to use universalistic terms whenever possible. A simple reason to avoid this is that it is ineffective and likely to backfire. Accusations of wrongdoing must be specific and directed at leaders, who in some cases can be vilified, never at the great mass of followers. I recall Clinton’s “basket of deplorables” gaff. This is a good example: the label strongly reinforced conservative voters’ tendency to see themselves as a maligned identity group; what was an off-hand analytical comment to the left was dynamite for the right.
We might just see the Clinton example as a breakdown of campaign discipline, but it’s more than that: the danger here is that these constructs are inherently hierarchical. In the us-versus-them set-up, the “us” is superior to the “them”. In political debates this typically manifests as moral superiority. It is easy to fall into this trap; some of the MAGA crowd are self-consciously racist and openly espouse morally repugnant views. We have to resist the impulse to generalize this evaluation. The fact that conservative identity-based appeals “work” on all the conservative voters does not mean that we ought to describe all of them as inherently racist or bigoted. The conservative messages are tailored to respond to a basic human need for belonging. In the case of Trump supporters, they also simultaneously remind voters that they perceive threats to their status and provide protection for that status. We are all vulnerable to these sorts of appeals, most of us have anxiety about our status in society, and we all need a narrative of belonging. Progressives need to respond not with judgement but by offering a positive alternative to identity-based appeals.
We should also admit that it feels good to see ourselves as “better than” them. Building hierarchical frameworks in which one’s own group ranks higher is so common it seems natural. For all I know it is. Outside of contexts that we consider political, we do this all the time. But we should try to be more aware of this tendency because even off-hand references to identity make it easier for conservatives to maintain their identity-based battle lines. That needs explication.
We have trained ourselves to call out these mental hierarchies when they reflect or reinforce systems of exclusion or oppression that we now recognize, like colonialism or the white male patriarchy. Becoming “woke” to these assumptions is a major achievement. Conservatives may find this new awareness oppressive, but even the fact that they complain about the need to censor themselves means that at some level they get it too, if only subconsciously. On the left, this awareness can lead to counterproductive hyper-sensitive reactions that are corrosive of ally-ship, but these are pretty rare in practice and can usually be attributed to individual communication failures, and especially to a general lack of trust which makes any utterance by certain speakers suspect to certain listeners.
Where we fail is when we flip the mental hierarchies. This most often happens outside of what we recognize as political contexts (of course, almost everything is a political context). The examples I have in mind are innocuous to me, but they all potentially help keep the conservative identity narrative going. How many times, for example, have we heard an African American sports commentator find white athleticism freakish? Or a female commentator make some blanket women-are-better-at statement? Or a gay commentator dismiss the fashion choices of a straight man? (OK, that last one may be fair.) These comments do not advance or support a system of oppression; indeed, they are part of the effort to unmake those hierarchies. Because they are intellectually lazy, they also don’t really help advance the cause. You cannot combat stereotypes by fostering other stereotypes.
My hypothesis, which I can’t really test at the moment, is that these small violations of neutrality help fuel the conservative obsession with “reverse” discrimination, an essential element of their overall identity messaging. We would do well to minimize them.
Limiting group references would be personally beneficial as well. I mostly attribute these sorts of references to the demand to fill the endless air-time of 24-hour cable TV and podcasts. But they still reveal our general tendency to place ourselves “above” others in some hierarchy of effort, merit, or even natural ability, reflecting beliefs, unconscious, unstated, or explicit, in group superiority. No one is immune. We would all do well to try to be more aware of our own reliance on hierarchical mental models.
I’ll continue this discussion next week.
Next week: The Identity Trap II (3 June 2025)
Notes
These two essays follow up on an issue I have raised a few times before, most notably in “Movements and Membership”.
New readers need to know that I am not referring to neoliberalism here, but true conservatism. See the earlier posts on the Three Political Languages,(I),(II), and on the Three Worlds of Welfare Capitalism.
I use the phrase “identity-based language” instead of “identity politics” advisedly. Identity politics is what conservatives do—they use political means to construct identity. Civil rights leaders and their allies are demanding inclusion for one or more identity groups, hence their need to name them, but (mostly) not defining these identities. Inclusion, the ultimate goal here, involves the recognition that the identity of the whole community must reflect the identities of excluded groups in equal measure to all others.
The substance of the question in the clip featured in the ads was whether the government had a responsibility for paying for the medical care of its inmates. It does. Is that unfair? A little, since people who have not committed a federal crime often cannot get the medical care, including gender care, that they need. But that is not exactly what the GOP was “upset” about.
The Democratic part in the “bathroom” politics in the US Capitol in past year seems to get this redirect and belittle tactic right (relating to where Sarah McBride pees). Sometimes, the Republican position is just silly (they can be scary and stupid at the same time).
See Bouke Klein Teeselink and Georgios Melios, “Weather to Protest: The Effect of Black Lives Matter Protests on the 2020 Presidential Election.” Polit Behav (2025). Not all studies agree. An earlier paper found insignificant mobilization effects: Oliver Engist and Felix Schafmeister, “Do political protests mobilize voters? Evidence from the Black Lives Matter protests”, Public Choice (2022) 193:293-313. Their study focused on pre-2020 events, however. Ultimately, the power of Floyd’s murder is what gave the organizing efforts around BLM their power.
Juliana M. Horowitz, Kiana Cox, and Kiley Hurst, “Views of Race, Policing and Black Lives Matter in the 5 Years Since George Floyd’s Killing”, Pew Research, May 7, 2025. The events around Floyd’s death caused support to peak, but Pew was finding majority support for the sentiment in 2017. An earlier Pew report was also helpful: https://www.pewresearch.org/short-reads/2023/07/12/8-facts-about-black-lives-matter/
Steve Crabtree, “Most Americans Say Policing Needs ‘Major Changes’”, Gallup, July 23, 2020. There was much less support fort the “Defund the Police” slogan. In this same survey, 47% supported reducing police funding. BLM is clearly a better slogan, but that is not saying much. “Defund” is ambiguous and its supporters espouse an enormous range of underlying positions.
I got the police victimization data from the Law Enforcement Epidemiology Project at the University of Illinois at Chicago, the ultimate source is the CDC. Between 2012 and 2022, the rate of victimization rose from 1.8 per million to 2.6 per million. The rise has occurred since 2016 (between 2000 and 2016 one could argue that the trend was stationary).
It is often the case that reforms based on civil rights concerns—that is, group-specific concerns—have broad benefits when implemented. This will usually be because the changes help poorer White people (a theme I’ll return to next week), but sometimes it is just a side-effect. (I am thinking of the benefits that accrue to the male spouses of women who benefit from employment policies that accommodate their giving birth.)
If this statement seems confusing, see my earlier posts on the Three Political Languages (I and II). See note 2, above. We all think about politics using languages rooted in liberal, conservative, and republican worldviews. These are adapted to respond to different core political questions, which can overlap and conflict. Traditional liberalism is the dominant political culture, even for people most motivated by traditional conservatism; this creates an acute psychological problem for them, but we all deal with the same conflicts to some extent.
The Gallup Polls are recorded HERE. The crime data are on the NCVS Dashboard. Reported is the overall rate of violent victimizations (including rape and sexual assault, robbery, aggravated assault, and simple assault. (National Crime Victimization Survey)