The Scare Comes After

The obvious thing to say is that this is inaccurate — that democratic socialism, as practised in Copenhagen and as advocated in Queens, is a set of arguments about how capitalism distributes its output, pursued through elections, and that this is not what the word communism has ever meant. Academics who study the movement have been saying so patiently for a year. They are right, and it does not matter, because accuracy was never the function.

The function is to convert a policy argument into a security question. Policy arguments have to be answered. Security questions only have to be raised. If the proposition on the table is "should the state pay for dialysis," someone eventually has to produce a number. If the proposition is "are these people loyal," nobody has to produce anything at all, and the burden lands permanently on the accused.

This is the oldest move in the American political repertoire, and the record on it is unusually complete.

i · the word does the work

Strip out the ideology and look at the mechanics. A label like this is an enclosure: it fences off a region of the possible and marks it hazardous, so that nobody has to walk in and check. The efficiency is remarkable. One syllable — red — retires an entire century of economic argument from serious consideration, at a cost to the person deploying it of roughly zero.

What makes it work is not persuasion. Most Americans, asked directly, favour most of the individual policies. What makes it work is asymmetry of effort. Defending yourself against the label requires paragraphs; applying it requires a word. Any exchange that runs on that ratio has a predictable equilibrium, and it is not a debate.

Hold onto that ratio. It is the only moving part in this essay, and everything that follows — the crackdowns, the purges, and the thing that is happening right now that has never happened before — is that one number changing value.

So the question is not whether the red scare is imaginary. Of course it is imaginary — the seed of this whole cycle is that a third of a major party's coalition now claims a label that, by every available measure, describes people who vote, canvass, and run for county comptroller. The question is what an imaginary scare is for, and the answer is that an imaginary enemy is superior to a real one in every operational respect. It cannot be examined. It cannot be negotiated with. It cannot be defeated, which means the emergency never has to end.

ii · 1912, and what came after

Here is the part that should worry the people currently enjoying the polling.

In 1912 Eugene Debs took 901,551 votes for president — roughly six percent, the high-water mark for American socialism. Beneath that number sat something more substantial: something like twelve hundred Socialists holding elected office across the country, seventy-nine of them mayors, in twenty-four states. Milwaukee, Schenectady, Berkeley. Not a fringe. An infrastructure.

The crackdown did not precede that success. It followed it.

The Espionage Act arrived in 1917 and the Sedition Act in 1918. Debs gave a speech in Canton, Ohio, and received ten years. In 1919 and 1920 Attorney General A. Mitchell Palmer's raids swept up thousands; in December 1919 the Buford sailed for Russia with 249 deportees aboard, Emma Goldman among them. In January 1920 the New York State Assembly suspended five duly elected Socialist assemblymen and expelled them in April — not for any crime, but for belonging to a party. Charles Evans Hughes, a Republican and a former Supreme Court justice, called it what it was and was ignored. Milwaukee sent Victor Berger to Congress; the House refused to seat him, twice.

Debs ran again in 1920 from a cell in the Atlanta federal penitentiary and got 913,693 votes — more raw votes than in 1912. It made no difference. The infrastructure was gone by 1924.

Now the objection, which is a good one and has to be answered rather than stepped around.

None of that was aimed at the ballot box on its face. The Espionage Act was a war measure, passed two months after a declaration of war. Debs went to Atlanta for an anti-war speech, prosecuted for obstructing recruitment — not for polling six percent. The Palmer Raids followed an anarchist bombing campaign, one of whose devices detonated at Palmer's own house, and fell hardest on the foreign-born rather than on Socialist officeholders. Read the record narrowly and you can explain 1917 through 1920 without mentioning socialism's vote share once.

Which means the causal claim has to be stated more carefully than "success invites repression" — because stated that way, it is an inference from adjacency, which is precisely the move this essay spent its first section objecting to.

Here is the careful version. Strength is the necessary condition. The shock is the permission.

Nobody builds an apparatus to dismantle a party polling two percent; there is nothing to take apart and the effort is not worth its cost. The machinery becomes worth constructing only once there is an infrastructure to aim it at. But constructing it also requires a public willing to tolerate the construction, and that willingness is never produced by the movement's own success — success produces opponents, not emergencies. The tolerance has to be supplied from outside: a war, a bomb, a revolution abroad, a strike that shuts the steel industry. Both conditions have to arrive together, and in 1917 they did.

That is a weaker claim than the essay would like, and a considerably more useful one, because it is falsifiable and it tells you what to watch. Strength alone predicts nothing. Strength plus an exogenous shock predicts the instruments.

iii · the second pass

The 1919 version was crude. The 1947 version was elegant, and it is the better template for what comes next, because almost none of it looked like repression while it was happening.

The permission arrived on schedule. In 1946 there were 4,985 work stoppages involving roughly 4.6 million workers — the most the Bureau of Labor Statistics has ever recorded in a single year, opening in January with 750,000 steelworkers, the largest single strike in American history. That November the Democrats lost both chambers for the first time since 1930. Labour had demonstrated exactly how much power it held, and an incoming Republican Congress had been handed a mandate to answer it.

Note that this cuts against the lazy version of the thesis, and the correction matters. 1946 was not a year in which the label stopped working at the ballot box; it was a year in which the party nearest the labour movement was routed. The strength that counted in 1946 was industrial, not electoral — and the shock that supplied permission was the demonstration of that strength itself. Sometimes the movement supplies its own trigger.

Taft-Hartley followed in 1947, over Truman's veto. It did not ban anything. It simply required union officers to sign affidavits disclaiming Communist affiliation in order to access the National Labor Relations Board — a procedural condition, entirely voluntary in the sense that a union was free to decline and lose its legal standing. In 1949 and 1950 the CIO expelled eleven affiliated unions representing about a million workers. In 1954 Congress passed the Communist Control Act, which stripped the Communist Party of legal standing; it cleared the Senate 81 to 1 and was signed by Eisenhower. Its lead sponsor was Hubert Humphrey, a liberal, who wanted to prove that liberals could be trusted.

That last detail is the one worth keeping. The second Red Scare was not primarily executed by its loudest proponents. It was executed by the centre of the coalition being scared, purging its own left flank to establish that the accusation did not apply. That is what the label is designed to produce, and it is far more effective than anything the accuser can do directly.

The results are still visible from orbit. The United States is the only wealthy democracy without a mass labour or social-democratic party, and the only one without universal health coverage. Those two facts are the same fact. The vocabulary was made radioactive in 1919 and re-irradiated in 1950, and a country that cannot say a word in public cannot organise around the thing the word names. The scare was imaginary then too. It worked anyway. It is, measured in outcomes, one of the most successful political projects in American history.

iv · the fight that isn't about what it's about

Which brings us to the argument currently running inside the Democratic coalition, framed as economics versus identity, and generally reported as a values dispute between the party's material-interests wing and its cultural-recognition wing.

The polling makes that framing hard to sustain. Democratic-aligned Americans who identify as democratic socialists are, in the CNN numbers, younger, less likely to hold a college degree, more likely to earn under $100,000, and — notably — more likely to be white than the Democrats who do not. Fifty-nine percent are under forty-five. This is not the seminar wing. It is closer to the opposite: the part of the coalition with the least credentialed relationship to politics, arriving at a class analysis without having been assigned one.

A coalition that has spent seventy years unable to use the vocabulary of class in public will inevitably reorganise its internal conflicts around vocabulary it is permitted to use. The economics-versus-identity fight is not the cause of anything. It is a downstream artefact of an enclosure imposed in 1920 and maintained since — the shape a disagreement takes when the direct language for it has been fenced off. Remove the fence and most of the participants discover they were arguing about the same distributional question in two dialects.

v · the deviation

Something genuinely has shifted, and it deserves marking, because a deviation from a pattern this old is rare.

The label is not holding. A third of the coalition has adopted it voluntarily. Sixty-four percent of Democratic-aligned adults who do not identify as democratic socialists say they would be at least fine with a democratic socialist nominee. Candidates carrying the term have taken House nominations from sitting incumbents — Darializa Avila Chevalier over Adriano Espaillat in upper Manhattan, Melat Kiros over Diana DeGette, a thirty-year incumbent, in Denver — after a democratic socialist won the mayoralty of New York City. A word that ended careers for a century is now, in one of the two American parties, a modest asset.

That is a real change in the power map, not merely in the rhetoric.

And the essay has been sitting on the explanation since its second section.

The label never worked by persuasion. It worked by asymmetry of effort — a word to apply, paragraphs to rebut. But that ratio was never a property of the word. It was a property of the distribution system. Rebuttal was expensive because rebuttal had to be routed: through a newspaper that would print it, a broadcaster that would book you, a union hall, a party committee, a publisher — every one of them a chokepoint operated by somebody with a strong professional reason not to be seen carrying your paragraph in 1953. The accusation travelled free. The answer paid a toll at every gate, and most of the gates were closed.

Remove the gates and the ratio inverts. When a million people can each publish the paragraph, and the accused can answer without asking anyone's permission, the cost of rebuttal collapses toward the cost of accusation. The word did not become less frightening. It became cheaper to argue with. That is not a change in rhetoric or in generational taste — it is the same mechanic running backwards, driven by the identical variable that made it work in the first place.

It also closes the 1947 pathway, which is the more consequential half.

That template required the centre of the scared coalition to purge its own left, and purging requires control of two specific things: nominations and affiliations. Humphrey could put his name on the Communist Control Act. The CIO leadership could expel eleven affiliates, and did. In 1912 the Socialists were an external party and could simply be excluded from the ballot line. In 1949 the left was inside an institution whose centre still held the levers.

Neither condition survives. Today's left is inside the Democratic Party and winning its primaries — against a thirty-year incumbent in Colorado, against a sitting member in Manhattan. A coalition centre that cannot control its own nominations cannot purge anybody. There is no affiliation to revoke, no charter to withdraw, no card to demand. The pathway that made 1947 elegant is not merely difficult now. It is absent.

So it is worth stating plainly what would mean this essay is wrong, since the essay runs on historical resemblance and has just spent four sections warning about exactly that inference.

If the cost of rebuttal has genuinely inverted, and no exogenous shock arrives, then the pattern does not run. The label stays broken, the distributional argument gets had on its merits for the first time since 1919, and a century of enclosure ends without a third pass. That is a live possibility and not a remote one. Precedent is what a pattern looks like from inside it; precedent is not a mechanism, and this essay does not get to assume otherwise merely because it has a good hundred years behind it.

vi · the prediction

But notice what the mechanism analysis predicts if it does hold — because it is not what a century of precedent would suggest.

If the label broke because rebuttal got cheap, then anyone who still needs the label's function has lost the option of winning the argument. Better rhetoric will not restore the ratio; nothing restores the ratio, short of restoring the gates. What remains available is the class of instruments that never required an argument to be won in the first place.

So the forecast is not that the rhetoric will intensify — rhetoric is cheap and already saturated, and its saturation is evidence that it has stopped working. The forecast is a transition in kind, and it has a signature you can watch for: the moment the noun stops appearing in speeches and starts appearing in instruments. Designations. Funding conditions attached to federal money. Nonprofit status reviews of organisations that train candidates. Conspiracy and racketeering theories applied to the people who organise protests rather than the people who attend them. Loyalty questions embedded in eligibility for something mundane. None of it will be called a red scare. In 1947 none of it was called one either; it was called a procedural condition on access to the labor board.

And it will need its permission. Strength is now demonstrable — that is what the primaries were. The missing variable is the shock, and shocks are not scheduled. Watch for the two arriving together: an incident, and within the year, an instrument that would have been unthinkable the month before.

The tell will be small and administrative, and it will arrive shortly after an election the label failed to decide. Watch for the paperwork, not the speeches. The speeches are how you know the paperwork is being drafted.