Two or three weeks ago, I had an intuition, a glimpse of a thought that has kept coming back to me since: the discourse of norm erosion isn’t really about Trump. Nor is it about authoritarianism. What it’s really about is “extremism,” that old stalking horse of Cold War liberalism. And while that discourse of norm erosion won’t do much to limit Trump and the GOP, its real contribution will be to mark the outer limits of left politics, just at a moment when we’re seeing the rise of a left that seems willing to push those limits. That was my thought.

And now we have this New York Times op-ed by Steven Levitsky and Daniel Ziblatt, two of the premier scholars of norm erosion, about the dangers of norm erosion. Nowhere in it will you find the word authoritarianism, though there is a glancing reference to “Trump’s autocratic impulses.” What you find instead is concern about “dysfunction” and “crisis.”

What you find is this:

Democrats are beginning to respond in kind. Their recent filibuster triggering a government shutdown took a page out of the Gingrich playbook. And if they retake the Senate in 2018, there is talk of denying President Trump the opportunity to fill any Supreme Court vacancy. This is a dangerous spiral.

Now imagine — bear with me — that it’s 2020, and Sanders is elected with a somewhat radicalized Democratic Party in Congress. Or if that’s too much to swallow, imagine some version of that (not necessarily Sanders or the Democrats but an empowered electoral left) in 2024. Or a realignment of the sort the US saw in 1932. Realignments always involve a contestation over norms; realignments change norms; realignments erode norms. And all of these counsels against norm erosion and polarization — which many people in the media and academia are invoking against Trump and the GOP — will now come rushing back at the Left.

And how could they not? When you set up “norms” as your standard, without evaluating their specific democratic valence in each instance, the projects to which they are attached, how could you know whether a norm contributes to democracy, in the substantive or procedural sense, or detracts from it? How could you know whether the erosion is good or bad, democratic or antidemocratic?

Levitsky and Ziblatt mention two norms: mutual toleration and forbearance in the exercise of power. Sometimes forbearance serves the cause of democracy; sometimes it does not. But by their lights, a lack of forbearance, by definition, becomes a problem for democracy.

Consider this revealing moment in the piece:

Could it happen here? It already has. During the 1850s, polarization over slavery undermined America’s democratic norms. Southern Democrats viewed the antislavery position of the emerging Republican Party as an existential threat. They assailed Republicans as “traitors to the Constitution” and vowed to “never permit this federal government to pass into the traitorous hands of the Black Republican Party.”

The authors want to posit the 1850s as a moment that “undermined America’s democratic norms,” strongly suggesting that prior to the 1850s, there was a robust enjoyment of democratic norms in America. Most of us would argue that when one portion of the people enslaves another, denying them their humanity (and the vote), there’s no real democratic norm in play. (Not to mention that one-half of the population, white and black, didn’t have suffrage at all.) And while it would have been awfully nice if the Southern slaveholders had simply agreed to vacate the stage of history peacefully, most of us realize that was never in the offing. Outside the South, wrote C. Vann Woodward, the end of slavery was “the liquidation of an investment.” Inside, it was “the death of a society.” These weren’t the sort of people to go gentle into that good night.

If American slavery were to be eliminated, someone had to call the question. That’s what the abolitionists (and the Republican Party) did. They polarized society. (For a representative example of how polarizing their discourse could be, read this.) And the result — however awful the Civil War was (and make no mistake, it was more awful than you can imagine) — was not the destruction of democracy and its norms but the creation of democracy — a “new birth of freedom,” Lincoln called it — which then got undone after Reconstruction, which was also a politics of norm-shattering.

As Jim Oakes has shown, the Southern Democrats were right to be terrified of the Republican Party, to see that party as an existential threat. The Republicans did want to destroy slavery, they did want to break the back of the slaveocracy, to gut a long-standing way of life. They wanted to do it peacefully, but they also understood that if war came, it would offer an opportunity to do it violently, an opportunity that they would not fail to seize. The Republicans were norm-breakers: they didn’t just want to limit the expansion of slavery into the territories (and whether limiting expansion of slavery was a norm in antebellum America is very unclear; in fact, we might say that the argument over that question was more the norm than any settlement of it; see Mark Graber’s book on Dred Scott); they wanted to limit that expansion as prelude to destroying the institution everywhere. Freedom national.

Levitsky and Ziblatt know that norm erosion and polarization were afoot during the 1850s. Only they want to put the onus entirely on the slaveholders. That way, they can take a stand against norm erosion without endorsing slavery; that way, they can pin the polarization of the era entirely on the Southern Democrats; that way, they can have their discourse of norm erosion and eat it, too. That’s politically understandable, in some sense, but wildly off the mark, historically.

And perhaps, in the end, not so politically understandable. For it suggests — no, says — that had the southerners merely shown some forbearance toward the Republicans, democratic norms would have persisted. On the question of slavery’s persistence, Levitsky and Ziblatt have nothing to say.

(I should note here that Ziblatt and I had an interesting exchange about these issues on Twitter.)

A similar, though perhaps less fraught, moment arises in their treatment of the Constitution:

We should not take democracy for granted. There is nothing intrinsic in American culture that immunizes us against its breakdown. Even our brilliantly designed Constitution cannot, by itself, guarantee democracy’s survival. If it could, then the republic would not have collapsed into civil war 74 years after its birth.

One of the last books Robert Dahl wrote was How Democratic is the American Constitution? His answer: not very. Yet in the same way that the discourse of norm erosion re-describes antebellum America, half of which was a slaveholder society, as a democracy, with democratic norms needing protection from polarizing forces, so does it re-describe the Constitution as a “brilliantly designed” text that is necessarily, though not sufficiently, connected to “democracy’s survival.”

What the op-ed does is show what the real object of concern is in the discourse of norm erosion: not authoritarianism, as I said, but extremism — whether that extremism comes from slaveholders or abolitionists, the Republicans shutting down the government to deny people health care or the Democrats shutting it down to allow immigrants to live here. Both sides do it.

To press the point a little further: Let’s take one of the cases that Levitsky and Ziblatt mention — the recent shutdown. Now, one of the strongest arguments in favor of the notion that Trump is an authoritarian is the treatment of immigrants. The Democrats shut down the government in order to secure some sort of deal that would allow hundreds of thousands of Dreamers to live in the country where they have lived most of their lives. The question is: assuming a shutdown would secure that outcome, is a shutdown democracy-enhancing or norm eroding? Arguably, it’s both, and not in an unrelated way: arguably the latter is necessary to the former. Arguably, norm erosion is not antithetical to democracy but an ally of it.

If your highest value is the preservation of American institutions, the avoidance of “dysfunction,” the discourse of norm erosion makes sense. If it’s democracy, not so much. Sometimes democracy requires the shattering of norms and institutions.

I fully recognize that the devil is in that “sometimes.” Some norms should be shattered, some should not. Some norm erosion undermines democracy, some enhances it. But that’s the real discussion we need to have: not a general toxin against norm erosion as somehow the bane of democracy — which may set us up for a centrist politics but not for a democratic one — but a more normatively informed discussion of what democracy requires.

For now, I’ll merely leave us with this thought: democracy is a permanent project of norm erosion, forever shattering the norms of hierarchy and domination and the political forms that aid and abet them.