Almost nine years ago, Michelle Alexander published one of the most important books of the twenty-first century. The New Jim Crow thrust the question of mass incarceration into the public sphere in a way no social movement had done in decades, arguably since the 1970s.

Alexander argued that the explosion in mass incarceration in general, and the mass incarceration of African Americans in particular, was not only the byproduct of a bipartisan consensus but a new form of mid-twentieth century racism. Her thesis resonated with the individual experiences of many — myself included.

As a young adult I remember Bill Clinton taking time off from his first presidential campaign to witness the execution of Ricky Ray Rector, an intellectually impaired black man; I remember when California passed its infamous “three strikes” legislation, clogging the jails for years to come; I remember the creation of the “super predator” myth and the equally apocryphal “crack baby.” More than one of my childhood friends spent time in prison. Almost every middle-aged black man I knew had had some sort of negative encounter with the police. Alexander’s book took hold because stories like mine seemed to scale up to account for the rise in mass incarceration and the proliferation of million-dollar blocks (city stretches where over $1 million was spent incarcerating its denizens).

But as powerful as Alexander’s book was, as powerful as the idea that twenty-first century mass incarceration is simply Jim Crow redux, there are two dynamics that it did not explain. It did not explain the fact that middle- to upper-income blacks are far less likely to be directly caught up in the carceral state than their poorer counterparts. And it did not explain the fact that alongside the large number of blacks swept up by the carceral state, there are also large numbers of whites and Latinos. Blacks are significantly over-represented in our nation’s prisons, but they only make up about a third of the US prison population (whites account for approximately 30 percent, Latinos about 23 percent, and a combination of Asian Americans and Native Americans 14 percent).

I was reminded of these figures in the aftermath of Tuesday’s elections, particularly those in Florida. Voters there cast ballots in a contentious race for governor, as well as in a referendum to largely end the state’s policy of felon disenfranchisement (the referendum included an unfortunate exemption for violent offenders). Amendment 4 sought to overturn a 180-year history — when states gave whites without property the right to vote, they often coupled it with legislation that would take that right away, which legislators often used to truncate black voting rights. Ahead of Tuesday’s election, felon disenfranchisement barred well over a million citizens from the polling booth; some estimates suggested that more than 20 percent of black men in the state could not vote due to past felony convictions.

Large-scale disenfranchisement clearly hindered Andrew Gillum’s race for governor. And voter suppression — making it tougher to register voters, throwing up barriers to vote for recent transplants, placing a polling site in a gated community — likely gave DeSantis an added boost. Gillum, the former mayor of Tallahassee, trails Republican Ron DeSantis by just thirty-six thousand votes as of this writing.

Given how close the election is — a recount will likely be held — and the reality of the New Jim Crow carceral state, one might think that the Amendment 4 vote was also close. If Amendment 4 was designed to end the New Jim Crow in a state with a 14 percent black population, then one would presume that this vote was razor-thin as well.

That assumption would be wrong.

Not only did Amendment 4 passed, it passed handily. Amendment 4 received over 1 million more votes than Andrew Gillum did.

While we’ll undoubtedly need more data, what we appear to be looking at is a case in which a number of whites simultaneously voted for a racist candidate and an end disenfranchisement. How can we account for this?

Often, we write about the “identity” in “identity politics” as if it were some hard and fast thing. But the reality is that identities are constructed and malleable — often triggered by politics as much as the other way around.

A white working-class Floridian man may well think black people are genetically inferior, but if he’s a felon (and of the 1.7 million Floridian felons, only about five hundred thousand are black), he can neither vote nor get a job as easily as white non-felons. When that man’s brother or wife or pastor goes to the polls, and Amendment 4 is staring them in the face, it’s not necessarily their racial identity that’s triggered but rather one that connects them with the hundreds of thousands of black men and women affected by felon disenfranchisement. While race may have shaped their vote for DeSantis, it’s likely their class identity that shaped their vote for Amendment 4.

What does this mean going forward? First, we need to be far more attentive to the role that both race and class play in fueling mass incarceration — and the coalitions that can end it. The “New Jim Crow” has created an easy frame to draw in potential activists, but it doesn’t map well enough onto reality. Forgetting about class is at once erroneous and constricting, narrowing the possibilities for multiracial coalitions. Second, to the extent that we engage in electoral politics, we should focus more on ballot initiatives like Amendment 4, which sidestep two-party politics and can potentially sidestep enduring racial divides. These initiatives are sorely underused as vehicles for progressive change.

The passage of Amendment 4 was perhaps the brightest spot on Tuesday night. With the right analysis and the right strategy, we could have more progressive victories like it.