The Case That Does Not FitPart I — The Cases Before Us
Chapter 4 — The Discipline Chart
The Discipline Chart
The spreadsheet had no face on it.
It had columns: school, grade, incident type, suspension, expulsion, race, disability status, sex, date, length of removal. Some cells contained numbers. Some contained percentages. Some were blank because the category did not apply or the data had not been entered. At the top of the sheet, a filter could narrow the view to one campus, one semester, one offense, one group of students.
Nothing in the spreadsheet said why a particular child had been removed from class. Nothing in it showed the teacher's tone, the hallway noise, the prior warnings, the fear in the room, the parent meeting that did or did not happen, the administrator trying to keep order, or the student who had been sent out three times that month and now appeared as one more mark in a cell.
The chart was a distance-making object. That was its danger. It could flatten children into categories and decisions into counts.
It was also its value.
No single teacher sees a district. No principal sees every classroom. No parent can know whether what happened to one child is isolated or ordinary. Local discretion often hides inside local explanation: this student was disruptive, that incident was serious, this administrator used professional judgment, that removal followed policy. Each decision may have a reason. Each file may look defensible by itself. The spreadsheet asks a different question: what happens when the decisions are counted together?
This chapter begins by reversing the moral direction of the previous three. Sometimes the general rule is what protects the person from the local judgment that has mistaken them for someone else. Sometimes a chart is not bureaucracy imposing abstraction on lived reality. Sometimes it is the first way an institution sees a pattern that local reality has normalized.
An administrator and a civil-rights reviewer sit beside the same screen. The first filter narrows the spreadsheet by school. The second narrows it by offense. The pattern sharpens around words that look ordinary until they are counted: defiance, disruption, disrespect, threat. One campus has far more referrals under those categories than another campus serving similar grades. A third filter separates suspensions longer than three days. The numbers do not accuse anyone yet. They make the room quieter.
The reviewer opens individual files. The first incident has a plausible explanation: a teacher wrote that the student refused repeated directions and the class could not continue. The second file also has a plausible explanation: a hallway confrontation escalated after a warning. The third includes a parent conference, a prior behavior plan, and a note from an assistant principal who appears to have followed district policy. None of the files says, in plain language, that race or disability caused the decision. None proves the chart's story by itself.
Then the reviewer returns to the aggregate view. The same vague categories are doing more work at one campus than another. Students with disabilities are appearing in removal counts that seem out of proportion to enrollment. The files are not identical, but the pattern is too persistent to dismiss without investigation. The administrator can feel both truths at once: each decision deserves to be judged on its facts, and the accumulation of those facts may reveal something no single file can confess.
The spreadsheet does not relieve the district of proof. It does not tell the principal to reverse a suspension because a percentage looks bad. It does not make disparity into discrimination. It does require a next step: check the coding, compare campuses, ask whether categories are too elastic, examine training and alternatives, test the explanations against the pattern rather than allowing each explanation to bury it.
On July 23, 2026, the U.S. Department of Education announced that it was rescinding Title VI disparate-impact regulations. The department framed the move as a return to intentional-discrimination enforcement and argued that prior disparate-impact provisions could pressure schools toward racial balancing or decision-making based on outcomes rather than unequal treatment of individuals. The final rule was published in the Federal Register at 91 FR 46733 on July 24, 2026, and took effect that same day.1
The department's concern should be stated before the objection.
Civil-rights law is not supposed to make individuals instruments of demographic management. If a school disciplines a student for a serious violation, the moral question in that case cannot be answered solely by asking whether the discipline chart will look worse afterward. A student is not a unit used to calibrate a spreadsheet. A teacher threatened in a classroom is not required to absorb risk because a district fears an aggregate disparity. A principal should not be pressured to treat one child more leniently or more harshly because of a racial category rather than because of the facts before her.
Intentional-discrimination enforcement protects an important principle: the state should not treat individuals unequally because of race. It focuses inquiry on decisions, motives, classifications, and unequal treatment. It asks whether officials did something to a person because of a protected characteristic. It resists the temptation to infer wrongdoing from statistical disparity alone. That resistance matters because unequal outcomes can arise from many sources: prior disadvantage, neighborhood conditions, poverty, disability identification, reporting differences, small sample sizes, policy differences, or chance. A rule that treats disparity as proof may punish institutions for conditions they did not create or push them toward quotas they cannot openly defend.
There is a real danger here. Effects-based review can become crude. It can encourage administrators to manage numbers instead of justice. It can make staff suspicious that any decision affecting the chart will be judged first by demographic consequence and only later by fact. It can produce a different kind of arbitrariness: individuals handled not as themselves but as representatives of a category the institution is trying to balance.
That is one protection rules must preserve.
The case for effects-based review begins where the intentional-treatment rule becomes too narrow to see institutional harm.
Discrimination does not always announce itself as motive. A district can adopt a facially neutral policy that repeatedly removes one group of students at much higher rates. A school can rely on discretionary categories like defiance, disruption, disrespect, or threat in ways that absorb cultural bias without anyone writing bias into the rule. A pattern can persist across classrooms because each adult believes he is responding to the case before him. No single decision-maker needs to confess hostile intent for the aggregate result to reveal a problem.
If civil-rights enforcement looks only for intentional unequal treatment, it may miss the kind of institutional distortion most likely to survive in modern administration. Few policies say, in plain language, that one race should be punished more harshly. Few officials describe their choices that way. Yet students can still experience a system in which the general rules are applied through local judgments that repeatedly fall harder on them.
The spreadsheet does not prove why. It does not prove unlawful discrimination by itself. But it can reveal that the institution has a question it cannot responsibly ignore.
This is the protective function of effects-based review. It gives institutions a reason to ask whether neutral policies are producing unequal burdens that cannot be justified. It prevents local decision-makers from hiding behind the innocence of each isolated file. It says that legitimacy may require attention not only to the motive inside a decision, but to the pattern produced when many decisions accumulate.
For the student repeatedly removed under vague discipline categories, the general rule may be the only thing strong enough to interrupt local discretion. Without the chart, each removal remains a local fact. With the chart, the institution must ask why the removals form the shape they do.
This is why the discipline chart is morally unstable.
It can be a warning light. It can also be a false accusation. It can reveal a pattern that no one meant and everyone reproduced. It can also tempt people to treat disparity as self-explanatory when the underlying causes require more careful investigation. It can protect students from biased discretion. It can also pressure administrators to make individual decisions with one eye on demographic totals.
Neither danger is imaginary.
The intentional-treatment focus protects people from being reduced to group outcomes. It insists that if the government accuses an institution of discrimination, the accusation should be tied to unequal treatment, not simply unequal results. It protects a student who violated a rule from being used as a statistical adjustment. It protects teachers and principals from being told that discipline must follow a demographic target rather than the facts of a classroom.
The effects-based focus protects people from a different reduction: the claim that no one was wronged because no one can prove hostile intent. It insists that institutional legitimacy includes attention to predictable consequences. It protects students whose repeated exclusion is rationalized one incident at a time. It asks whether policies written in neutral language are functioning, in practice, as channels for unequal burden.
The book's invariant appears here from the opposite direction. A discipline standard, a civil-rights regulation, a reporting requirement, and a statistical review process can all restrain local actors who might otherwise treat similar conduct differently for reasons no one sees. The counterweight is equally necessary: a purely effects-based rule can mistake people for numbers, treating an individual decision as suspect because of the category it affects rather than because the decision itself was wrong.
Legitimacy depends on preserving both protections.
A school discipline system is an especially difficult place to preserve them because classrooms require fast judgment.
A teacher cannot run a regression analysis while a fight is happening. A principal cannot wait for district-level review before deciding whether a student may return after a threat. Schools must maintain order for the students who were harmed, distracted, or frightened, not only for the student facing discipline. Rules that make every removal legally hazardous can leave staff feeling abandoned and other students unprotected.
That is the strongest case against blunt effects pressure. If administrators believe the chart matters more than classroom reality, they may under-enforce necessary rules. The cost will not fall evenly. It may fall on students in the same schools already least protected by stability. A civil-rights system that ignores safety and order in the name of aggregate parity can fail the very children it means to protect.
At the same time, classroom urgency is exactly why pattern review matters. Fast decisions are vulnerable to habit. Categories like "defiance" and "disruption" can absorb adult interpretation. Two students can perform similar behavior and receive different consequences because one is read as immature and another as threatening. A teacher may not notice the pattern because each decision felt justified in the moment. A principal may not notice because each referral came with a narrative. The district may not notice until the spreadsheet is compiled.
Local judgment needs room to act. It also needs institutions capable of reviewing what local judgment repeatedly does.
That review should be disciplined. A disparity should trigger inquiry, not automatic condemnation. The institution should ask: Are the underlying offenses comparable? Are categories too vague? Are some classrooms referring students at much higher rates? Are disability-related behaviors being punished as misconduct? Are restorative alternatives available evenly? Are administrators overriding teachers consistently? Are policies producing removal for conduct that could be handled differently? Are sample sizes large enough? Are records accurate? What explanations are supported by evidence?
These questions treat the chart as a beginning, not a verdict.
The dispute over Title VI disparate impact is often narrated as a choice between colorblind law and racial justice. That may be the political vocabulary surrounding it. Institutionally, the question is more precise: when should patterned consequences require a rule to answer for what it repeatedly does?
If the answer is never, institutions can preserve neutrality while exporting harm through discretion. They can say the policy is facially equal, the officials meant well, and each case had reasons, while the same groups bear the burden year after year. The law would protect against explicit unequal treatment but become weak against structured indifference.
If the answer is whenever outcomes differ, institutions can become hostage to numbers. They may infer injustice where the evidence is not sufficient, adopt hidden quotas, or treat individuals through demographic categories in ways that compromise the very equal treatment civil-rights law is meant to secure.
The hard answer lies between those failures. Patterned consequences should have enough institutional force to reopen inquiry, demand explanation, and require correction when neutral policies cannot be justified. They should not, by themselves, replace the work of proof. A spreadsheet can tell the institution where to look. It cannot tell the institution what it will find.
That distinction is fragile. It requires regulators, courts, districts, parents, teachers, and advocates to resist the simplest version of their own argument. The intentional-treatment advocate must admit that motive is not the only way institutions harm. The effects-based advocate must admit that outcomes are not self-interpreting and that individual treatment cannot be sacrificed to aggregate design.
The chart is useful because it refuses to let isolated decisions remain isolated. It is dangerous because it can make isolation disappear too completely.
Return to the spreadsheet.
No cell contains a motive. No formula can say whether a principal acted from bias, fear, care, pressure, prudence, exhaustion, or a good-faith reading of the rules. The chart cannot tell whether one suspension was necessary, another excessive, a third discriminatory, and a fourth unavoidable. It cannot see the classroom.
But the chart can see what the classroom cannot see.
It can show that one category of students is removed at a rate no one expected. It can show that vague offenses carry most of the disparity. It can show that one school is unlike the others. It can show that a policy defended as neutral produces consequences that ask for explanation. It can show enough to make denial irresponsible.
The spreadsheet is not justice. It is not proof. It is not a substitute for individual judgment. It is a demand that individual judgments be accountable to the pattern they collectively make.
No cell in the spreadsheet contains a motive. The pattern exists anyway. Whether the pattern proves injustice is the question the spreadsheet cannot answer by itself.
Footnotes
-
U.S. Department of Education, "U.S. Department of Education Removes Disparate-Impact Title VI Regulations," July 23, 2026, https://www.ed.gov/about/news/press-release/us-department-of-education-removes-disparate-impact-title-vi-regulations; Federal Register, "Rescinding Portions of the Department of Education Title VI Regulations To Align With the Statutory Prohibition Against Discrimination Based on Race, Color, or National Origin," 91 FR 46733 (July 24, 2026), Doc. 2026-15019, effective July 24, 2026, https://www.federalregister.gov/documents/2026/07/24/2026-15019/rescinding-portions-of-the-department-of-education-title-vi-regulations-to-align-with-the-statutory. ↩