ASHFALL INSTITUTE | SUBDUCTION ZONE
Building Bridges — Episode Thirteen
THE MODEL IS NOT CLEAN
P. A. Moore
Ashfall Institute | Subduction Zone
Written under the collaboration disclosed in Time Is a Wheel and We Are the Ground. Concept, argument and judgment: P. A. Moore. Research and composition: Claude Opus 5.
I have leaned on the 1930s in this series more heavily than on anything else.
Twenty-seven days from a signature to the first enrollee. Three million young men. A government office that sent people out with notebooks because it understood a window was closing. Buildings, bridges, murals, trails, and a demonstration that a country can decide to do a large thing quickly.
I meant every word of it. And I have been presenting a model without its ledger, which is precisely the thing this series exists to object to.
So here is the ledger. Not as a concession, and not as an apology for having used the example — as the condition of using it honestly. An idealized model is worthless. You cannot rebuild from a photograph that has been retouched, because the retouching always hides the load-bearing part.
The camps were segregated, and it took a direct order to make them so
The founding legislation of the Civilian Conservation Corps prohibited discrimination on account of race, color or creed. That clause was in the law from the beginning.
Black enrollment was capped at ten percent — pegged to the share of the national population, in a Depression that had fallen on Black families disproportionately harder. A quota calculated from the census is not neutral when the need is not distributed like the census.
Then it got worse deliberately. In 1934 the Corps' director, Robert Fechner, asked the Army to review the question. The Army concluded that the CCC should not enforce segregation. In 1935 Fechner overruled that finding and issued an order making the complete segregation of colored and white enrollees the rule.
Asked about it by the NAACP, he wrote that he was satisfied the Black enrollees preferred it, and that segregation of this kind was not discrimination and could not be construed as such.
The camps that resulted were often placed on national park land far from any town, partly to keep them away from hostile white communities — which is to say the men were protected by being made remote, and the hostility was accommodated rather than confronted. Where integrated camps were permitted at all, it was sometimes justified by an asserted natural adaptability of Black men to kitchen work, and Black enrollees in those camps were assigned to cook while the technical outdoor work went to others.
Every one of those decisions was made by an administrator, against a written non-discrimination clause and against his own Army’s advice. That matters for what follows. It was not the temper of the age acting on its own. It was a person, with a name, making a ruling.
The women got camps too, in the sense that a rounding error is a number
Three million men passed through the CCC.
For women there was Camp TERA at Bear Mountain in 1933, and then a program built by Hilda Worthington Smith under Harry Hopkins, with Eleanor Roosevelt pushing from behind. By the summer of 1934 there were twenty-eight camps across twenty-six states and the District of Columbia. Across the whole life of the effort, roughly ninety residential schools and camps.
Set that against three million.
The press called them the She-She-She camps, and the name is the argument. It was not a program with a nickname. It was a nickname doing the work of a policy — a way of making the whole idea sound like a joke about women talking, so that no one had to say out loud that the largest employment program in the country’s history had been written for men.
I want to put one more thing beside it, from Episode Eight. When the government sent writers out with notebooks in 1936 to record the people who had been born into slavery, they got 2,300 interviews and we would have nothing without them. Nobody sent anyone out with a notebook to ask the women in the valleys how they did what they did. The same decade, the same capability, the same budget line. One window was recognized. The other was not, and it closed.
The exclusions were written into the permanent law, not just the temporary programs
The camps ended. The Social Security Act did not.
The 1935 Act excluded agricultural laborers and domestic servants from old-age insurance. That was roughly half of all workers in the American economy, and it fell very unevenly: at least sixty percent of the country’s Black workers were in those two categories. Black workers were about eleven percent of the workforce and about twenty-three percent of the people the Act did not cover.
I am going to be careful here, because the honest version is more useful than the satisfying one. Historians disagree about why. One account holds that the exclusions were the price demanded by Southern committee chairmen who were not going to permit a federal check to reach a Black tenant farmer. Another, argued from inside the program’s own history office, holds that agricultural and domestic work was excluded for administrative reasons — that in 1935 there was no workable way to collect payroll tax from a farm or a household — and points out that other countries excluded the same categories without the same racial politics.
Both may be true at once, and that is the part worth sitting with. A policy can be built on a real administrative constraint and be kept in place by people who like where it lands. You do not need a conspiracy for the outcome. You need a difficulty, and then an absence of anyone with the standing to insist that the difficulty be solved.
The exclusions were narrowed in 1950. Fifteen years is a long time to be told your work is the kind that cannot be counted.
The dam took the valley
The Tennessee Valley Authority is the crown jewel of the period, and it is the clearest case of the cost.
Building Norris Dam required emptying the reservoir floor. Nearly three thousand families — close to fourteen thousand people — were required to leave land they held, by eminent domain, for a public good they had not asked for and largely did not receive first.
And Norris, the model town built for the project — planned, electrified, admired, photographed — did not admit Black families. The stated reason was respect for local custom. Black leaders pointed out at the time that poor Black and white families had been living side by side in those valleys long before the TVA arrived, and that the custom being honored was in part being invented. The NAACP filed complaints about TVA hiring, housing and training in 1934, in 1935, and again in 1938.
Three complaints in five years is not an oversight. It is a finding that was received and set aside.
And the thing that was not a program at all
Between 1929 and 1936, somewhere between four hundred thousand and two million people of Mexican descent were removed from the United States.
The estimates vary that widely because nobody was keeping a proper record, which is itself the point. What is not in dispute is that a very large share of those expelled — commonly put between forty and sixty percent — were United States citizens by birth, overwhelmingly children.
It was not primarily a federal deportation program. It was carried out by cities and counties and states, with private cooperation, in a climate the federal government had helped create. It ran through the same years as the alphabet agencies, in the same country, under the same flag, and it has almost no place in the story we tell about the decade.
A country that could organize three million men into camps in a summer was, at the same moment, organizing the removal of its own citizens and not writing down how many.
Why I am still using this example
Here is where I have to be exact, because the easy move is to say the whole thing was rotten, and that is not what I think and it is not what the record shows.
The failures above are failures of who was let in. They are not failures of capacity.
Those are different, and only one of them is a claim about what is possible. The 1930s did not demonstrate that a good government did a good thing. It demonstrated that a mediocre, compromised, viciously bigoted government retained the raw capability to plan and execute at enormous scale, in weeks, with the technology of 1933. The camps got built. The dams got built. The checks went out. The interviews got taken.
If the argument were we should do it the way they did it, the ledger above would destroy it. The argument is the capability exists and we have been told for forty years that it does not. The ledger does not touch that. If anything it strengthens it, because the capability survived being run by people that flawed.
Everything on that list is a decision about eligibility. Not one of them is a discovery about the limits of what a country can do.
What would have to be different, and it is not a pledge
Good intentions were not in short supply in 1935 either. Something structural has to change or the outcome repeats, so here is what I would actually require, and every item comes out of an earlier episode in this series.
Eligibility written into the statute in the specific, and enforceable by the excluded person. The CCC’s non-discrimination clause existed and was overridden by an administrator’s memorandum. A right that only the administrator can enforce is not a right. It is a preference.
The rolls published. Who applied, who was admitted, who was turned away, by place. Fechner’s ten percent cap was arithmetic done in public and objected to at the time. It survived because objecting and prevailing are different things — but the objection was possible only because the number was visible. Every exclusion above was known while it was happening. None of them required later scholarship to discover. They required someone with standing to make the objection stick.
The record kept in more than one place. This is Episode Twelve and I am not going to repeat it, except to note that the reason we can write this episode at all is that the NAACP kept its own files, the Army kept its finding, and Fechner’s letter survived in someone else’s archive. The unflattering half of the New Deal record exists because it was not held solely by the New Deal.
And a named accounting of who is not in the room. In 1933 the answer was Black men, women, tenant farmers, domestic workers, and anyone of Mexican descent in Los Angeles County. In 2026 the list is different and I am not going to pretend I can see all of it — that is exactly the failure mode. But I know the first entry, because this whole series is about it: the harm from the thing not built has no name attached, and the people it falls on are never in the room where the decision gets made.
The retouched photograph
I have spent twelve episodes arguing that the party holding the information should not be the party deciding what the information means. That the record has to be kept where the interested party cannot reach it. That a model with the ugly parts sanded off is not a model, it is an advertisement.
It would be indefensible to make that argument and then hand you a clean 1930s.
So: the decade I have been holding up did great things and did them while excluding, displacing and expelling people, and both halves are load-bearing. The great things prove the capability. The exclusions prove that capability is not the safeguard, and never was.
We are being told right now that we cannot do large things anymore. That is false, and the 1930s prove it.
We are also being told, more quietly, that if we ever did do a large thing again, it would of course be fair this time.
Nothing proves that. Nothing has ever proved that. It is the only part we would have to build from scratch.
Next time: the last one.
Sources: African American enrollment cap, Robert Fechner’s 1935 segregation order over the Army’s contrary conclusion, his correspondence with the NAACP, and camp siting and work assignment practices, as documented by the Corps Network’s Moving Forward Initiative, the Living New Deal project, and PBS. Women’s camps, Camp TERA, Hilda Worthington Smith and the scale of the women’s program, from the Living New Deal project and the University of Pennsylvania Weitzman School’s research on women and the CCC. Social Security Act exclusions and their racial distribution, and the competing administrative-feasibility account, from the Social Security Administration’s own policy history and from subsequent scholarship. Norris Dam displacement figures, the exclusion of Black families from the town of Norris, and the NAACP complaints of 1934, 1935 and 1938, from the Loyola University Chicago eminent domain project and contemporary accounts. Mexican Repatriation figures and the citizenship share, from Immigration History and reporting collected at Bowdoin College and NPR. All figures were checked against multiple independent accounts before publication; where the sources disagree, the range is given rather than a single number.
Building Bridges is a fourteen-part series. Previous: Episode Twelve — Two Ways of Remembering. Next: Episode Fourteen — A Full-Time Job at Six O’Clock.
P. A. Moore is the pen name of Pamela King, philosopher and artist. Available through the Ashfall Institute.