Where the argument over equal justice actually runs, what order and reform each protect, and the single standard both sides invoke when no case is attached.
The blindfold is the part of the statue everyone remembers, and it makes the boldest claim there: that justice works by refusing to look at who is standing in front of it — their money, their faith, their name, their neighborhood, the passport they happened to be born holding. Practically nobody in American life argues against that as a principle. The entire fight is over whether the blindfold is on.
The view taken here is that peace and justice are one subject rather than two, that a standard applied to some people and not to others has stopped being a standard, and that this holds for how a government behaves abroad as much as for how it treats the people inside it. What that requires in practice is genuinely contested, and both serious positions deserve to be stated in their own words rather than their opponents'.
Start with any disparity — any measure on which two groups come out differently — and two accounts form.
The first holds that the system is broadly fair and that its failures are individual. The law is written without reference to who it applies to. On this account the overwhelming majority of the people inside it are trying to do a hard job correctly, and the ones who are not can be identified and removed. Outcomes differ for many reasons that have nothing to do with a decision made at the courthouse door, and treating every gap as proof of bias skips over the causes that actually produced it. It also makes a fair procedural objection: if any disparity counts as evidence and no absence of one counts against, the charge cannot be tested, and an accusation that cannot be answered is not an argument.
The second holds that the pattern is the point. You can grant that nearly everyone involved is acting in good faith and still notice that the result comes out the same way, because a system does not need anyone to intend an outcome in order to produce it reliably — only a chain of small discretionary judgments leaning the same direction. When one explanation is offered for a case, then a different explanation for the next, generation after generation, at some point the explanations are the thing needing explaining. To this side, insisting on intent sets a standard nothing can meet and makes the harm the responsibility of whoever suffered it.
Policing is where the argument gets most confused, because two exposures land on the same people. The neighborhood asking for a faster response and the neighborhood asking to be stopped less are frequently the same neighborhood, sometimes the same household. Safety is a service you can receive too little of and too much of simultaneously, and an account that hears only one half of that street is not listening to it.
Money runs alongside all of it. Whether an outcome tracks what you did or what you could afford is a question the law's own defenders raise: the lawyer you can retain, whether you can wait for trial outside a cell, whether appearing on a Tuesday morning costs you the job. None of that is written into any statute, and all of it shapes results.
Foreign policy is the same principle at a different scale, and this page deliberately keeps it there. The question is consistency: whether the standard invoked when a rival government does something is the standard invoked when a friendly one does the same thing, and whether the public can see the reasoning behind the difference when there is one. The serious answer from the other direction is that statecraft is not a courtroom. A government's first obligation is to its own citizens; alliances require working with partners nobody would defend in the abstract; and a demand for perfect consistency can mean walking away from the only leverage that was actually reducing harm. Whether force should be used at all is a separate argument, taken up in War & Peace; this page is about whether one standard is being applied to everyone.
The order side is protecting the presumption that institutions are legitimate until shown otherwise, because a society in which nobody believes the law applies does not become fairer — it becomes a place where the strong do what they can. It is protecting the people who need someone to come when they call, who are rarely the people with the loudest platform, and it is protecting the officer, the prosecutor and the judge from being convicted in public of a pattern they personally did not create.
The reform side is protecting against a rule that makes the injured party prove someone's inner state before the injury counts. It is protecting the presumption of innocence and the process around it — the idea that the worst day of your life should not be resolved by whoever is most convenient to charge. And it is protecting the person for whom the system's rare failure is not rare, because they live where it fails.
Both are protecting someone who cannot afford them to be wrong. Neither list requires contempt for the other.
State the principle with no case attached and agreement is close to total. The same rule for the powerful and the unknown. A process you would accept before knowing which side of it you would be standing on. Reasoning you are allowed to see. Consequences for officials who abuse a position, on the theory that authority and accountability are the same thing viewed from two ends. Anti-discrimination as a floor rather than a favor, because a right that depends on being from a favored group is a privilege wearing a better coat.
The money problem is common ground nobody advertises. Almost no one defends a system in which the outcome tracks what a defendant can pay — not the people who trust the courts, who see it as corrosion of something they value, and not the people who distrust them, who see it as the mechanism itself.
And the domestic and foreign versions of the test are the same test. Would this be acceptable if it were done to us, by someone else, for the reasons they gave? That question does not settle any particular dispute, but it does something more useful than most arguments manage: it applies equally to the person asking it. The habit of running it on yourself first is the whole of the method described in how this works.