The Intent Standard

Ask an American whether something can be racist if nobody meant it that way, and you have not asked a philosophy question. You have sorted the country.

In June 2026 the Fairleigh Dickinson University Poll put the question to 1,211 New Jersey voters in about as neutral a form as it can be put. Fifty-two percent said racist outcomes can happen without racist intent. Thirty-nine percent said something can only be racist if somebody decided to do a racist thing. There was no meaningful difference between white, Black and Hispanic respondents: 51, 56 and 52 percent respectively took the systemic view, all inside each other's margin of error.

By party, the same question opens a canyon: 69 percent of Democrats, 45 percent of independents, and 35 percent of Republicans. Thirty-four points, on the definition of a word.

He is right, and this dossier exists because he is right. Nearly every dispute filed on this site resolves, eventually, into that one question. This page sets out where the intent standard came from, what it does, and why a rule that sounds like ordinary fairness has functioned for fifty years as the most effective protection racial discrimination has ever had in American law.

I. WHAT THE STANDARD ACTUALLY SAYS

Strip away the vocabulary and there are two competing tests for whether a policy is racially discriminatory.

Disparate impact asks what a policy does. If a rule falls much harder on one race than another and cannot be justified on its own terms, it is discriminatory, whatever was in anyone's head.

Discriminatory intent asks what a policy was for. A rule is discriminatory only if somebody chose it because of its racial effect. Not knowing it would fall that way. Not being indifferent to it falling that way. Choosing it because it falls that way.

American law contains both tests. Which one applies depends on whether you are suing under a statute or under the Constitution, and that split is the whole story.

II. 1971: THE COURT ADOPTS IMPACT

Duke Power's Dan River plant had, before 1964, simply barred Black employees from every department except Labor. When the Civil Rights Act made that illegal, the company introduced a high-school diploma requirement and two aptitude tests for transfer out of Labor. Neither had any demonstrated relationship to the jobs. Both screened out Black applicants at dramatically higher rates.

In Griggs v. Duke Power (1971) a unanimous Supreme Court held that Title VII reaches consequences, not just motives: a facially neutral practice with a disparate effect violates the Act unless the employer can show business necessity. Chief Justice Burger wrote for the Court that the Act proscribes "not only overt discrimination but also practices that are fair in form, but discriminatory in operation."

For five years, that looked like the direction of American law.

III. 1976: THE HINGE

Two Black applicants to the District of Columbia police department challenged Test 21, a written examination on which Black candidates failed at roughly four times the rate of white ones and which had never been shown to predict performance as a police officer. The facts were close to Griggs. The claim was constitutional rather than statutory.

In Washington v. Davis (1976), by 7 to 2, the Court held that the Equal Protection Clause does not work like Title VII. A law is not unconstitutional "solely because it has a racially disproportionate impact." A plaintiff must prove discriminatory purpose.

That sentence is the hinge on which the next half-century turns. Congress can still write disparate-impact rules by statute, and did: in Title VII, in the Fair Housing Act, in the Voting Rights Act. But the Constitution itself, the thing that does not need sixty Senate votes, now asks for a state of mind.

Three years later the Court tightened it further. In Personnel Administrator v. Feeney (1979) it explained that discriminatory purpose "implies more than intent as volition or intent as awareness of consequences." The decision-maker must have acted at least in part "because of, not merely in spite of" the harm to the group.

Read that again slowly, because it is the load-bearing sentence in American discrimination law. Under Feeney, a legislature may know with certainty that a bill will devastate Black citizens, pass it anyway, and remain constitutionally innocent, so long as the devastation was a price it was willing to pay rather than a result it was seeking.

IV. 1987: WHAT THE STANDARD COSTS

The clearest measurement of what the intent standard does is a death penalty case.

Warren McCleskey, a Black man, was sentenced to die in Georgia for killing a white police officer. His lawyers brought the Baldus study: an analysis of more than 2,000 Georgia murder cases which, after controlling for 39 non-racial variables, found that defendants charged with killing white victims were 4.3 times as likely to receive a death sentence as those charged with killing Black victims. Prosecutors sought death in 70 percent of cases with a Black defendant and a white victim, and in 19 percent of cases with a white defendant and a Black victim.

The Court did not dispute the study. In McCleskey v. Kemp (1987), 5 to 4, it held the study insufficient anyway. Statistical disparity across a system does not establish that this prosecutor or this jury acted with discriminatory purpose in this case.

Justice Powell, who wrote the majority, later told his biographer that McCleskey was the one vote of his career he would change.

Notice what the ruling requires. Georgia's capital sentencing was racially skewed by a factor of more than four, measured across two thousand cases with thirty-nine controls, and that was not evidence of racial discrimination in the constitutional sense. To win, McCleskey would have needed something no defendant will ever have: a decision-maker willing to say out loud why.

The jar is the whole doctrine.

V. THE TRAP, STATED PLAINLY

Here is the mechanism, and it is worth being precise about it because it is not a conspiracy. It is arithmetic.

The intent standard makes a confession the price of admission. And the one thing American racial politics has reliably produced since 1964 is the disappearance of confessions.

This is not speculation. The strategy's own operator described the process on tape. Lee Atwater, in a 1981 interview, explained that the appeal had begun with open slurs, that by 1968 the slur backfired so the language moved to busing and states' rights, and that by the 1980s it had become more abstract still: tax cuts and spending cuts whose burdens, he noted, fell more heavily on Black citizens than white ones. He was describing, with some pride, a fifty-year project of removing intent from the record. (Full account: THE SOUTHERN STRATEGY.)

Set that beside JIM CROW and the pattern resolves. Jim Crow is the best-documented system of racial subjugation in modern history for one reason: its builders announced their purpose in official proceedings. Knox said the convention was to establish white supremacy. Glass said discrimination was precisely the proposal. Tillman told the Senate they had stuffed ballot boxes and shot people and were not ashamed of it.

Under the intent standard, Jim Crow would be easy to strike down and almost nothing since would be. The regime that said the quiet part into the Congressional Record is legally vulnerable. The one that learned not to is not. A rule that keys liability to candour rewards, precisely and exclusively, the people who stopped being candid.

In July 2026 a federal court wrote that down. Denying an injunction against a Tennessee map that split the state's only majority-Black congressional district three ways, a three-judge panel observed in a footnote that the doctrine "triggers a strange game" in which the majority party, "aware that any mention of race could invalidate the map, coaches its legislators to stay quiet." It then noted that Tennessee's Republicans said nothing about race at any point in the process, and concluded: "Surely, not mentioning race isn't evidence of racial animus." The mechanism described on this page is no longer an inference drawn by critics. It is a paragraph in a published opinion, written by the judges applying it. (Full account: Tennessee cut Memphis into three districts.)

VI. WHERE IT STANDS NOW

The statutory disparate-impact tools that survived Washington v. Davis have been narrowing.

Shelby County v. Holder (2013) disabled the Voting Rights Act's preclearance machinery, over Justice Ginsburg's warning that discarding a working remedy was like "throwing away your umbrella in a rainstorm because you are not getting wet." Texas revived a blocked voter-ID law within hours. North Carolina passed an omnibus voting law within weeks; the Fourth Circuit struck it down in 2016 after finding its provisions targeted Black voters with "almost surgical precision" (a finding of intent, made possible only because the legislature had requested racial data on voting practices before drafting). That is what it now takes.

Disparate impact survives under the Fair Housing Act, narrowly, after Texas Department of Housing v. Inclusive Communities (2015). It survives under Title VII. It has never applied to the Constitution since 1976.

VII. THE REBUTTAL FILE

Standing claims, checked against the record.

VIII. WHY THIS DOSSIER EXISTS

Return to the poll that opened this page, because it contains one more finding, and it is the hopeful one.

FDU ran an experiment. Among white voters living in the whitest quarter of New Jersey towns who already accepted that racism can be systemic, simply asking about racism before asking about their own schools nearly tripled the share who said their schools were segregated (from 6 percent to 17) and moved support for regional magnet schools from 53 percent to 72. Among white voters in the same towns who said racism requires intent, nothing moved at all. (Full account: the New Jersey vision test.)

The lens only works if you are already wearing the prescription. Which is another way of saying that the intent question is not one argument among many on this site. It is the door. Everything documented here (every quotation, every budget line, every registration roll) is visible to a reader who accepts that outcomes can be racist without a racist, and very largely invisible to a reader who does not.

That is why this dossier is filed alongside the others rather than buried in a footnote. It is the standard by which the rest of the archive is read.