Tuesday, February 07, 2017

The Answer to Judge Clifton's Math Question

by Michael Dorf

During the just-concluded Ninth Circuit oral argument in the immigration case, Judge Clifton seemed to temporarily stump Washington SG Noah Purcell when he asked whether Purcell could come up with any case in which a facially neutral law was deemed discriminatory based on the fact that it harmed only fifteen percent of the disfavored class. Purcell more or less regained his footing by answering--correctly--that in order to prevail on a discrimination claim a plaintiff has never had to show that the challenged policy harms all members of the disfavored class or that it harms only members of the disfavored class. Nevertheless, he did leave the door slightly ajar by not addressing the numerical threshold.

I'd like to help SG Purcell out with a hypothetical. Suppose that a racist police chief wants to act out his racism but disguise the character of his actions. Thus, he instructs his beat officers to dramatically step up unjustified stop-and-frisk activity in the city's designated high-crime neighborhoods. Suppose further that only 15% of the city's African Americans live in the designated high-crime neighborhoods but that the population of the high-crime neighborhoods is 95% African American. (By comparison, Somalia is 98.6% Muslim, Iraq is 96% Muslim, and Yemen is nearly 100% Muslim. Syria is "only" 87% Muslim.) If there were solid evidence that the police chief had in fact decided to use the high-crime neighborhood designation simply to get at African Americans, this would be a slam-dunk case of unlawful racial discrimination. And that would continue to be true even if the designation of a neighborhood as "high crime" had been made by the previous (non-racist) mayor as a basis for a legitimate crime-control initiative, such as designating where to send extra patrol cars.

Thus, there is nothing to Judge Clifton's "math" objection. He might think that the evidence of subjective motivation is insufficient. (At one point he disturbingly referred to the evidence that is not only in the record but public record as mere "allegations.") But if the subjective motivation evidence is given its fair weight, there is nothing in the 15% figure or the fact that Congress and the Obama administration previously listed the seven countries for a different counterterrorism purpose that undercuts the state's case against the Trump executive order.


Diane Klein said...

You are on it! It was all I could do to live-tweet smart remarks (like "Drinking game: every time Flentje said NATIONAL SECURITY."

Donald R. Frederico said...

Professor Dorf: I've been retweeting some of your recent blog posts and thought you might be interested in some of mine: http://reflectionsofabostonlawyer.blogspot.com/2017/02/washington-v-trump-stays-thing.html - Don Frederico

Joe said...

Reminds me of Planned Parenthood v. Casey where the spousal notification law was argued to be okay since only a small portion of women were affected by it. Such "math" didn't work there.

Michael C. Dorf said...

Thanks for the comments and Don--nice to hear from you--and good analysis! The only place I might disagree is that the federal government has some case law on its side for a no-review position, at least with respect to aliens outside the U.S. who have not previously been here. I think the federal government should lose because I am persuaded by Adam Cox that those precedents are outdated, (see https://www.justsecurity.org/36988/muslim-ban-held-unconstitutional-myth-unconstrained-immigration-power/ ), but the argument is less terrible than it might appear.

Joseph said...

You use all the appropriate qualifying words to make this work in principle and I agree with the formulation. I think the judge's math question was mistaken if intended to masquerade as doctrine but he was at least pointing to how extraordinary the situation would need to be with that kind of math in play. Setting aside whether the immigration order is justified and how the court can determine that, showing that motivation for the policy was 'simply to get at Muslims' is challenging.I don't know what percentage of non-racist motivation saves a policy or if there is an easy mathematical answer.

Miguel said...

The question I would ask Judge Clifton is that, if 15% is okay, at what point does the percentage become not okay? There is, I believe, no way to craft a nonarbitrary standard based upon Judge Clifton's approach.

matt30 said...

I thought there was a slightly more intuitive answer than even the one you gave here. Iran, Iraq, Libya, Somalia, Sudan, Syria and Yemen are all countries with significant numbers of actual refugees seeking asylum/relief or a country from which a citizen getting educated abroad has a high likelihood of not returning home.

A Saudi or Indonesian isn't a target of the ban specifically because they aren't very likely to make the US their permanent home. And the purpose of the ban, in full view of the context in which it was signed, is to prevent resettlement of these people in the US where they would otherwise have a strong claim against a deportation action.

el roam said...

Thanks for the post , Theoretical moral analysis , won't do much here :

This is because , the issue is not mere discrimination , but , balancing between competing values . discrimination is wrong , yet , wrong as a general rule . If, one discriminates the other, not on a relevant basis, this is wrong (solely for being: black, Jew, woman, whatsoever) yet:

If discrimination , is relevant and legitimate , due to the issue discussed , then , it is : relevant , and legitimate , and legal .

Take for example , synagogues :

Generally speaking , you can't discriminate women , solely for being as such ( woman ) , yet : if , for religious purposes , one needs to separate women from men during prayers in synagogues , then :

This is relevant for the situation discussed , legitimate , and legal .
A detective , can't target a black , solely for being so , yet :

He may question the witness to an offense : Whether the suspect was black , and if indeed , chase him as such , since , it is relevant to the ongoing investigation .

I can't analyze further here now, too many complications , but, in the third circuit, there was a very relevant case, where, mass surveillance against Muslims has been made by New York city police department, and , I quote :

" Plaintiffs allege that the Program was based on the false and stigmatizing premise that Muslim religious identity“is a permissible proxy for criminality, and that Muslim individuals, businesses, and institutions can therefore be subject to pervasive surveillance not visited upon individuals, businesses, and institutions of any other religious faith or the
public at large. "

End of quotation :

Yet , one may find there opinions , that mere discrimination , whatsoever , is wrong and unconstitutional , Very recommended for the issue:



el roam said...

Just clarification to the example in my comment above , concerning synagogues :

The separation between women and men , creates alleged or relevant discrimination , not by the separation itself , but rather :

By barring so women , to pray , and be present , at the holy center of the synagogue , where all sacred books and staffs and alike , are situated . So , they are distanced to marginal location , in holy terms may I say ….


Michael C. Dorf said...

The answer to Joseph Simmons' question is really quite easy if you look at the mass of disparate impact litigation in the federal courts. Here's the general standard from the Arlington Heights case: "

"[Washington v.] Davis does not require a plaintiff to prove that the challenged action rested solely on racially discriminatory purposes. Rarely can it be said that a legislature or administrative body operating under a broad mandate made a decision motivated solely by a single concern, or even that a particular purpose was the 'dominant' or 'primary' one. In fact, it is because legislators and administrators are properly concerned with balancing numerous competing considerations that courts refrain from reviewing the merits of their decisions, absent a showing of arbitrariness or irrationality. But racial discrimination is not just another competing consideration. When there is a proof that a discriminatory purpose has been a motivating factor in the decision, this judicial deference is no longer justified."

The same standard applies in religious discrimination cases. There are some close legal questions in this litigation. Whether the plaintiffs have done enough to show a likelihood of success on the merits if the courts reach the equal protection claim is not one of them.

As to matt30's point, that would make a lot more sense if the executive order used the seven-country only to bar entry of people seeking to be treated as refugees. It doesn't.

Joe said...

'simply to get at Muslims'

Sounds like that is simply too high of a standard as compared to current case law.

"in full view of the context in which it was signed, is to prevent resettlement of these people in the US where they would otherwise have a strong claim against a deportation action"


Put aside the background that suggests motivation, which some want us to handwave away as too hard to prove. Both sides have arguments, but when arguments rely on such broad, easily refuted arguments, it is not doing the government much good.

ScriptKeeper said...

Every key point of this was going through my head last night while live, Professor: 'Literally every sex or race discrimination claim ever isn't literally against all women or all black people...' Literally every single one. Judge Clifton acted like he was not a lawyer, let alone a judge.

Joseph said...

In response to Joe, I wrote 'simply to get at Muslims' because "simply" was the word used by Prof Dorf and it is my impression that he chooses his words with care. As Prof Dorf wrote, that scenario would be a "slam dunk case." That is what motivated my question about the more difficult realities of this case.

Professor, I appreciate the response and it is helpful but I'm not sure it answers my question. According to the precedent you quote, where there is evidence of racial motivation, courts may review the merits of a decision. My question concerns what happens next. What I posited is whether there is some percentage of non-racist motivation which saves a policy.

Deciding the national security merits or improper motivation of Trump's order will be difficult to some and easy, even obvious, to others. I think there are great obstacles to a court reaching the equal protection claim, but as an intellectual matter I am interested in how that could play out.

Asher Steinberg said...

On this point, I was intrigued by something Eric Posner said at his blog:

"Imagine that a judge had asked: “if we knew that Trump acted on anti-Muslim animus AND that the seven countries posed special security threats, what is the right outcome under the Establishment clause?”

The answer must be, I think, that the government wins. Otherwise, the government would be unable to protect people if the president is a bigot. No court would come to that conclusion."

Now, I don't think that's right, anymore than the police chief in your hypothetical acts constitutionally because the neighborhoods he's policing really are high-crime neighborhoods. But I think he has a sort of point. Suppose that, if Trump's intentions were transparent to us, they went something like this. In the campaign, Trump proposes a Muslim ban because he believes Muslims in the main are more likely to be terrorists (and much more likely to be terrorists than we actually are) and we can't screen out the terrorist Muslims with total accuracy. Trump is quickly advised that this would be illegal and is advised that he could legally ban entry from a list of majority-Muslim nations, though not all of them, and quickly pivots to talking about banning immigration from certain countries. Once elected President, Trump then asks his advisers to give him a list of the majority-Muslim countries whose citizens pose us the greatest national security risk, his intentions here being that he thinks "radical Islamic terrorism" is a unique security risk and he wants to do what he legally can to keep out the radical Islamic terrorists until "extreme vetting" measures can be devised to his satisfaction, as he believes the current vetting isn't good enough, or isn't certain if it's good enough until he reviews it. Is that religious animus? Even at the stage of the Muslim ban, I don't know if you have animus per se or merely illicit stereotyping, but at the point where we are now I don't know if you even have that.

Now, there is an alternative story on which Trump hates Muslims, therefore initially wanted a ban, was told he couldn't have a Muslim ban, and instructed his advisers to devise an alternative measure that legally excluded as many Muslims as he could. That would of course be a case of animus. But I don't see much evidence that that's going on. Rather, all the public statements plaintiffs rely on suggest to me that Trump is concerned about "radical Islamic terrorism" and is, at worst, mistaken about the incidence of radical Islamic terrorists in the Muslim population, though it's unclear that that mistake has anything to do with the EO as it actually exists. And even if it does, unless Trump's mistaken to the extent of believing all or most Muslims are terrorists, I don't see how it's animus to be mistaken about the size of a hard-to-measure subgroup of a population.

Finally, of course if you define animus as the intent to disparately affect a religious/racial/ethnic minority, regardless of whether the reason for that desire is some kind of bigotry, a belief that all or most members of the group share some unfavorable trait, or merely a reasonable mistake or accurate perception about how many of them do, then you could very possibly find animus here. But I don't see how that could be the law, or it wouldn't even be necessary to scrutinize racial or religious classifications; they would all be definitionally motivated by animus and invalid per se.

matt30 said...

Asher Steinberg, I think Justice Kennedy's attempt to shoehorn animus into the Equal Protection and Due Process law in Windsor and Romer isn't analytically helpful.

The best response to your quandary is to go back to the traditional analysis. Are muslims abroad in war-torn countries a protected class (or perhaps more narrowly, are muslims that would otherwise be eligible for a visa a protected class)? If so, then the government must show that there is a compelling interest to regulate and that the classification is narrowly tailored and/or has a high degree of fit to the problem being addressed. That second part is a factual matter. It doesn't matter if Trump has a blind hatred for muslims or whether he holds the mistaken belief that muslims granted a visa for entry into the US pose a significant threat to national security. All that matters is whether there is factual support for the government's position (which is a judicially cognizable empirical question).

IMO, the disparate impact test is really just a proxy for determining whether a classification is overbroad for the purpose of determining whether a policy is narrowly tailored. In other words, it's irrelevant that you know exactly what percentage of "terrorists" are in a protected population so long as you can show that whatever previous vetting process was better suited to addressing the problem.

The animus language is just used a device (maybe only a rhetorical one?) to identify bad arguments. Those who opposed gay marriage or promoted sodomy laws didn't do so for NO reason, just reasons that could not be supported by provable facts.

matt30 said...
This comment has been removed by the author.
matt30 said...

"As to matt30's point, that would make a lot more sense if the executive order used the seven-country only to bar entry of people seeking to be treated as refugees. It doesn't."

I don't think you need that at all. If the white governor of a racially divided state decides to eliminate polling places only in majority black districts where people are highly motivated to vote, then its doesn't matter whether the order specifically targets "elderly black voters" or "designated high crime areas."

Obviously, the intent is only to target voters the governor believes will vote against him. Nevertheless, the order doesn't need to affect all blacks living in the state (or even a majority of them) nor does it need to single out a specific subgroup of highly motivated opposition to be constitutionally impermissible. All you need is a good degree of fit between the discriminatory intent and the way in which the policy furthers that purpose