The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
A Lawyer Who Litigated Grutter v. Bollinger Comments on the Court's Recent Case on Discrimination Against Majority Groups
I've known Michael Rosman at the Center for Individual Rights for 30 years; he has litigated many important cases, including with regard to racial preferences, and was one of the lawyers in Grutter v. Bollinger. I'm therefore delighted to pass along his short reaction to Ames v. Ohio Dep't of Youth Services; all the remaining text of the post is Michael's:
In Ames v. Ohio Dept. of Youth Services, the Supreme Court unanimously rejected the "background circumstances" rule that some circuits had applied in Title VII cases when the plaintiff claiming employment discrimination was a member of a "majority" group (like whites). The "background circumstances" rule required such "majority" plaintiffs to present some evidence showing that their employer was the "unusual" employer that discriminated against the majority.
Title VII, the Court held, had one text, that text applied to everyone and, accordingly, the same requirements for proving discrimination applied to everyone. Pretty straightforward, and, in my view, undoubtedly correct. So correct that Ohio did not really bother to defend the "background circumstances" rule.
What I have not read in the aftermath of the opinion, though, is the Court's own role in inspiring this bizarre interpretation, and allowing it to fester as a blot on statutory interpretation and Title VII. Not to mention the Court's less-than-candid discussion of its own responsibility.
The appeal in Ames was from the Sixth Circuit. In its first footnote, the Court noted the other circuits that had adopted the rule: the Seventh, Eighth, Tenth, and D.C. Circuits. It cited cases decided between 1992 and 2004 from those circuits. But, as Justice Thomas's concurrence pointed out, the "background circumstances" requirement was first adopted by the D.C. Circuit in 1981 in a case called Parker v. Baltimore and Ohio Railroad Co. So, for over forty years, majority plaintiffs in at least some jurisdictions had to meet an additional requirement that the Supreme Court unanimously rejected in 2025 in a quick eight-plus page opinion. And the "circuit split" has existed since at least 1999, when the Third Circuit explicitly rejected the "background circumstances" requirement.
In its decision, the Court in Ames held that the "background circumstances" rule was an improper addition to the evidentiary framework that the Court had adopted in McDonnell Douglas v. Corp v. Green. It described that framework in broad terms: the Court described the first step of the McDonnell Douglas framework as requiring the plaintiff to produce enough evidence to support "an inference of discriminatory motive." But McDonnell Douglas actually had quite a bit more to say about how to do that (all of which is omitted in Ames).
Specifically, the McDonnell Douglas framework said that plaintiffs could create an inference of discrimination by demonstrating that (1) they were a racial minority, (2) that they applied for and were qualified for a job, (3) they were rejected and (4) the job remained open. In Parker, the 1981 D.C. Circuit case, the court took that first step—being a racial minority—and decided that it had to be modified in "light of common experience" (a phrase that the Supreme Court had used in a later case) if the plaintiff was not a racial minority. Judge Mikva wrote for the court: "Whites are also a protected group under Title VII, but it defies common sense to suggest that the promotion of a black employee justifies an inference of prejudice against white co-workers in our present society."
Thus was born the "background circumstances" test. But its roots in McDonnell Douglas were ignored by both the Court in Ames and even Justice Thomas's concurrence (which was highly critical of the McDonnell Douglas framework).
One last thought. Conspirator Ilya Somin tells us that Ames reaffirms the principle that "discrimination based on sexual orientation is sex discrimination," the principle that was first articulated in Bostock v. Clayton County. But Ames is actually a bit confusing on this point.
Bostock held that discrimination based on sexual orientation is invariably sex discrimination:
[I]t is impossible to discriminate against a person for being homosexual or transgender without discriminating against that individual based on sex. Consider, for example, an employer with two employees, both of whom are attracted to men. The two individuals are, to the employer's mind, materially identical in all respects, except that one is a man and the other a woman. If the employer fires the male employee for no reason other than the fact he is attracted to men, the employer discriminates against him for traits or actions it tolerates in his female colleague.
But the broad proposition is at best questionable. For example, one can have sexual orientation discrimination even where sex discrimination is permissible under Title VII. Consider a male sports team that limits its locker room attendants to heterosexual males. If being male is a BFOQ—bona fide occupational qualification—for the job, then there is nothing illegal about the (admitted) sex discrimination. But there's still sexual orientation discrimination going on, and (in the absence of a rather idiosyncratic understanding of sex discrimination) the sexual orientation discrimination does not appear to constitute sex discrimination, at least as Bostock described it. (After all, the employer is treating men who are attracted to men precisely the same way that it is treating women who are attracted to men.)
In any event, Ames may have said something else. In describing the fact that Ames lost out on a promotion to a lesbian, the Court said that such evidence "would ordinarily satisfy her prima facie burden" in the absence of the "background circumstances" rule. Would it? "Ordinarily," to show sex discrimination, one would need to show that someone of the opposite sex obtained the position in order to satisfy the prima facie burden. (Ames was, in fact, replaced by a male after a second and distinct adverse employment determination.)
The Court's suggestion that a woman showing that she was rejected in favor of a lesbian would satisfy her prima facie burden under Title VII doesn't obviously reinforce that "discrimination based on sexual orientation is sex discrimination," but rather suggests that sexual orientation discrimination independently violates Title VII. If so, this would be a rather startling development given the absence of the phrase "sexual orientation" in Title VII and the text-focused analysis in Bostock.
In any event, the court chose not to reach any of the respondents' alternative arguments, and sent the case back down to the Sixth Circuit for further proceedings. So, I don't think anything in the opinion precludes the employer from arguing that Ames losing out to a lesbian is not sex discrimination in violation of Title VII. Or, for that matter, for Ames to argue that she would have received the promotion had she been a gay male, which certainly would be sex discrimination under Bostock.
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
Grutter - Findings of fact at the trial court - a definite quota
Grutter CA6 - opps lets have de novo finding of facts and take UofM pleadings as truth
the number of blacks admitted every year for 10-15 years straight was 19-21 (based on my recollection of the case). some years, UM bypassed 20-40 white applicants to reach that "critical mass of 20" while some years UM had to bypass 200 -300 white applicants to get that quota - oops "critical mass"
That was always the lie told by the libs when it came to affirmative action. It was sold to the public as something along the lines of if you only had 100 spots, do you wants spots 97-100 to be made up of more rich entitled white kids or do you want a black kid who grew up in the inner city to get a slight bump into those spots.
Now, that would be bad enough if that was only what was going on, but somewhat understandable. What was really happening was that minority groups were getting HUGE bumps in test scores and GPA and excluding massive numbers of whites and Asians to come up with their supposed critical mass.
It was based solely on race and not any personal criteria. A poor white kid who was first in his family to go to college could get bumped by a rich black kid.
It was always about racial numbers and quotas.
“If being male is a BFOQ—bona fide occupational qualification”
Is there any authority that being a particular sex is a BFOQ for being a locker room attendant? A woman can do the job, but the problem seems to be the some men don’t want women in their locker room.
Honoring the culturally-dominant privacy expectations of your customers is a bona fide occupational qualification. I think you could cite any of the cases commenting that single-sex bathrooms, locker rooms and changing areas are still allowed in support of that argument.
Culturally dominant expectations with respect to race and sex are generally not criteria for BFOQ’s.
It would be interesting to see how courts distinguish privacy expectations.
With respect to race, you are correct. With respect to sex, you are wrong. It remains culturally and legally acceptable to insist that the person handing out towels in the women's shower be a female.
Sure, see, e.g., the Title IX regs, 34 CFR § 106.61:
Thanks.
“a locker room or toilet facility used only by members of one sex” is not going to help if the facility is open to folk according to their “gender.”
GEN'DER, noun [Latin genus, from geno, gigno; Gr.to beget, or to be born; Eng. kind. Gr. a woman, a wife; Sans. gena, a wife, and genaga, a father. We have begin from the same root. See Begin and Can.]
1. Properly, kind; sort.
2. A sex, male or female. Hence,
The term gender referred to biological sex - at least until the activists wanted it to mean something else
But since they did want it to mean something else, it does mean something else - at least for the purpose of those jurisdictions that wish to open their "facilities" to :
{people who feel like they really are, or at least ought to be men}
or
{people who feel like they really are, or at least ought to be women}
if that, rather than sex, is the basis for the division of facilities it doesn't fit within the proviso that EV mentioned.
I don't think that is right. Bostock was gay. In order to win his case, did he have to show that a woman replaced him?
I don’t think the Court got into that question. IIRC, Bostock came up on summary judgment against him. The employer said it didn’t fire him because he was gay, but even if it did, they win anyways because the law doesn’t prohibit sexual orientation discrimination.
I believe that the Court simply said that was wrong and left the other issues in his case to be litigated below.
But yours is an interesting question. The straight woman claims she was unlawfully fired and replaced by a lesbian woman. How do Bostock’s contortions work in that case? She wasn’t fired because she was a woman--as evidenced by the fact that she was replaced by a woman.
Would the thought be that she was fired AS a woman , when a similarly situated gay man would not have been fired (as they both would have been attracted to men) and the employer just so happened to hire a lesbian woman as the next in line? To call that sex discrimination proves far too much.
Yes, SCOTUS reversed a summary judgment against Bostock and remanded for further proceedings. But, does it make any sense that Bostock will still lose on summary judgment if he was replaced by a straight man? It doesn't to me.
It doesn't. But it also doesn't make sense if he wins if he was replaced by a gay man.
The only reasonable way to argue that Bostock was discriminated against because of sex if he was replaced by a woman---assuming that was the but for cause of his firing.
It also seems that if Bostock was bisexual you could fire him for that reason because you would also fire a bisexual woman.
But that line of thinking calls all of Bostock into question and I doubt that SCOTUS is ready to overrule that case. However the more applications of it make it seem sillier.
If Gerald Bostock loses on summary judgment when he is replaced by a straight man, then Bostock is a dead letter.
It should be a dead letter. The Court gave a statute an interpretation that was never intended by the authors, wasn't driven by the text, and which likely would have precluded its enactment if anybody had thought that the meaning of it.
I am not sure if Ames would have had a different v oting result had any of the litigants asked the Court to overrule Bostock.
I'm not either. I don't think the justices who would give the statute such an interpretation would have been interested in overruling their decision to do so.
From the majority, Roberts, Gorsuch, Sotomayor, and Kagan are still on the Court, and who'd want to bet that Jackson would vote to overturn it? 5-4 or 6-3 for upholding it, would be my guess.
That's not how it works.
It was, which is why it was authored by the foremost textualist on the court.
And that's really not how it works.
I over thought this one. The explanation is simple.
Per Bostock, sexual orientation discrimination violates Title VII because such discrimination always means there is sex discrimination. Thus, if a straight woman is replaced by a lesbian, that’s prima facie evidence of sexual orientation discrimination, which in turn necessarily means there is prima facie evidence of sex discrimination.
I don’t see why the replacement is relevant (except as useful corroborating evidence.)
1. If you are fired because you are a woman -eg a new manager just dislikes working with women and he just fires you for that reason - then that’s sex discrimination even if you are not replaced
2. If you are fired because you are sexually attracted to people of the same sex as you, under the Bostock theory, even if this policy is applied to all employees regardless of sex, it is nevertheless sex discrimination because in order to determine whether you, Josh, are attracted to people of the same sex as you it is necessary to discover what sex humans you are sexually attracted to and then compare that with your own sex. Since your own sex is a necessary computational item, ie the firing decision will change if we hold the sex of the humans you are atttracted to constant but hypothetically flip yours to the sex that you are not, then that is sex discrimination per Bostock.
3. Thus if you Josh are fired under 2, you have been fired for Bostockian sex discrimination whether you are replaced or not. Even if you are replaced by another gay man (maybe the manager is willing to make an exception for his gay nephew) you’re still being sex discriminated against per Bostock.
Interestingly it seems to me that the poor old bisexuals are not protected by the Bostock logic. If the employers policy is simply to fire you for being bi, your own sex is not a computational item in the firing decision.
Yes, if the job remains open you have a Title VII case. But, if you claim racial discrimination and your replacement is the same race as you, you lose your case. Ditto if you claim sex discrimination and are replaced by a person of the same sex.
Where Rosman got it wrong is his claim that if you are fired because you are straight, that's sex discrimination (so far, so good). And because it is sex discrimination, if you are replaced by a person of the same sex, you lose your case (that's where he went wrong).
Well, he didn't say the second part. He just said that being replaced by someone of the same sex (or same race or ethnicity or religion) doesn't meet the prima facie case of discrimination under McDonnell Douglas. You can still prove your discrimination case with other evidence. (See the last sentence of the post.)
I think he is wrong. Ames being replaced by a lesbian suffices to meet the prima facie standard. It's enough to establish she was fired for being straight, which in turn is sex discrimination per Bostock.
Not as a matter of law. That's a hard fact to overcome, but it doesn't prove that race didn't play a factor in your termination.
"I don’t see why the replacement is relevant (except as useful corroborating evidence.)"
Well, it is conclusive evidence isn't it? If I claim I was fired because I'm over 40, but my replacement is also over 40, isn't that irrefutable proof that I was not fired for being over 40? By replacing me with the second person, I know that it is because of any reason in the world other than my age.
It’s not irrefutable, though it might be difficult to refute.
It might be that the employer’s policy is to fire employees who are both over 40 and not French speaking. If you are fired the fact that you are over 40 is a “but for” in the Bostock formulation.
But the next guy may be fluent in French and so acceptable notwithstanding that he is over 40.
So given the Bostock formulation - the but for - you should win even though your replacement is over 40.
I am not endorsing the “but for” test, as I don’t think it accurately describes the real reason for your dismissal, but that is what SCOTUS is offering us by way of reasoning.
If your replacement was also over 40, then your but-for claim that you would not have been fired had you been under 40 doesn't fly.
Not logical Captain.
If your condition is a conjunct :
If {X and Y} then Z
then each of X and Y are "but for" conditions for Z
This applies to any conjunct proposition - not just for
X = over 40
Y = not French speaking
Z = you're fired
You would not have been fired "but for" your being over 40. Your replacement is hired because he does not meet the other part of the conjunction.
Try this one :
X = male
Y = sexually attracted to males
Z = you're fired
If you are fired and replaced by a straight male ..... your being male is still a "but for."
Yes, he speaks French and you don't. But, that means you can't use him as the evidence you were fired because you were both over 40 and did not speak French. You can only use him as evidence you were fired because you don't speak French.
In contrast, a gay male who is replaced by a straight male can use that as evidence he was fired because he was gay. And then, Bostock says if you are fired because you are gay, you are fired because you are male.
In short, Bostock does not says if you are fired for any old Y, then it must be the case you were fired for any old X. Y must be "attracted to one of the sexes" and X must be "being one of the sexes."
In short, Bostock does not says if you are fired for any old Y, then it must be the case you were fired for any old X. Y must be "attracted to one of the sexes" and X must be "being one of the sexes."
No, you need to be precise here.
Y must be "attracted to one of the sexes" and
X must be "the same sex as in X"
The whole point is that Bostock only manages to conclude that anti-gay discrimination is an example of sex discrimination because - Bostock says - you can't determine whether someone is gay without analysing who they are sexually attracted to and their own sex. It's those last four words that move the needle.
That's why anti-bi discrimination wouldn't be sex discrimination under Bostock - you don't need to know someone's sex to know they're bisexual. And if you have an employee Charles who is bisexual, and you hypothetically swap him into Charlotte who is bisexual, you get the same answer.
Bostock doesn't make anti-gay discrimination a new standalone form of sex discrimination, it argues that it is a form of sex discrimination because the employee's sex is a "but for" in determining why he was fired.
Thus returning to the vital issue of employees above and below the 40 mark, and whether they speak French or not - neither condition is decisive for the employer on its own .... just like being male and being attracted to males. But if you are fired for failing both conditions - over 40 and not speaking French, and the new guy is over 40 and does speak French, you're being over 40 is still a but for, without which you wouldn't have been fired. If you had been 39 you'd still have the job, and the fact that they actually hired a 40 year old who can speak French rather than a 39 year old who can't, doesn't change that fact.
The logic is identical to you being fired for being male and being attracted to males. If you are replaced by a male who is not attracted to males ..... your being a male is still a but for, without which you wouldn't have been fired.
An interesting hypo is: suppose an employer doesn't know whether someone is male or female, but only knows that this person is gay. Would taking an adverse employment action against this person constitute sex discrimination under Bostock?
Perhaps being fired for being bisexual isn't a violation of Title VII under Bostock. But being fired for being gay or straight is.
Yes, being over 40 is a but-for cause of being fired in your hypo. But the fact that they hired an over-40 French speaker in your place cannot be used as the basis for establishing a violation of Title VI (being fired for not speaking French does not violate Title VII). You have to find some other evidence.
In contrast, the fact they hired a straight man can be used as the basis for establishing a violation of Title VII (being fired for being gay violates Title VII).
For this to be true, then EVERYONE who is fired has a claim. I am white but show up to work drunk yet am replaced by a white person who shows up sober----well, white is part of that!
It sounds silly because it is.
This is only true if the employer's policy is :
X = you're white
Y = you're drunk
If X + Y then Z (you're fired)
In such a case you really are being fired for being white (in the Bostock "but for" sense) because if you had been black you wouldn't have been fired however drunk you might have been.
This is precisely equivalent, as a matter of logic, to Mr Bostock.
The employer's policy was to fire gay people. The court reasoned that this policy was equivalent to a policy of firing people for X = being male and Y = being attracted to males. (Also X = being female and Y = being attracted to females.)
There's no law against firing people for being attracted to males. But Mr B won because there is a law against firing people for being male. And under the court's reconstruction of the "we fire gays" policy into the X and Y structure, both X and Y are "but for"s.
There is no "X and Y" structure in Bostock. Instead, it's "if Y then X" where Y is fired for being gay and X is fired for being male or female.
Neither does a claim of sex discrimination if you are replaced by a member of the same sex.
Yes it does when a straight women is replaced by a lesbian. That's discrimination because she was straight and Bostock says that necessarily means she was discriminated on account of being a woman.
Your drunk and white example makes the correct point that Lee's argument is nonsense. There is no connection between being drunk and white. But, you keep missing that being straight and a woman are linked per Bostock.
But to justify the conclusion that sexual orientation discrimination equals sex discriminations you must go through the logical leaps taken by Bostock or else it doesn't work.
Bostock made it clear WHY sexual orientation discrimination was sex discrimination in that case. If in another case it simply cannot be shown that there was sex discrimination, how can Bostock reasoning apply to the new facts? Bostock was very clear it was a textualist opinion.
What do we do with the bisexual employee that fired for being bisexual? Bostock simply cannot logically apply. Sure, the Court (being the Court) can DECLARE it applies, but it doesn't fit within the logical framework.
There is no logical leap because the Court held in Bostock:
Ames extends the ruling to being fired for being heterosexual. Now perhaps you don't have a Title VII case if you are fired for being bisexual. But, that's not relevant to the mistake Rosman made when he claimed a straight woman should lose her sex discrimination case if she was replaced by a lesbian.
The holding of a case is typically tethered to its facts, not Holy Writ to be applied in all circumstances.
In its early posture the Court could have ruled that Bostock made a prima facie case of sex discrimination but didn't rule on the downstream effects or evidentiary issues. It had to make such a ruling because he did not win entirely.
But if it turns out that Bostock was replaced by a man, how can he argue with any words that the English language uses, that he suffered discrimination because he is a man? It is just as impossible as me in my example saying I was fired because I was over 40.
As I said, the Court can proclaim, "Never mind all that logic, we said what we did and we are the Supreme Court!" but it makes no sense in conjunction with their other cases talking about "opposite sex" being an evidentiary requirement.
You may argue that it makes no sense because it would gut Bostock. But I don't see how the bisexual thing doesn't have the same problem.
If Bostock was replaced with a straight man, that's prima facie evidence he was fired because he was gay. The Court held in Bostock that when a man is fired because he is gay, he is also fired because he is a man (the but-for cause, that's the crux of the case). Therefore, being replaced by a straight man suffices as prima facie evidence he was fired for being a man.
Maybe I'm still fighting Bostock.
It is nonsensical for a man who is fired and replaced by another man to be able to claim that he was fired for being a man. The employer has conclusively demonstrated that sex was not the issue for replacement.
Nope. The whole "replaced by someone of a different group" is only a way to establish discrimination when there isn't direct evidence. And an obvious situation where that would not be "conclusively demonstrated" would be one where the hirer and firer are different people. Consider this extreme example: Mr. Smith is fired. There's a series of emails in which his female supervisors discussed how much they hate men, don't want to work with them, and plan to fire Mr. Smith. But after they fired him someone in HR got wind of this, and said, "WTF are you people doing? If you're firing employees for being men we're going to get killed in litigation!" And ordered them to hire Mr. Jones to fill the vacancy created by Mr. Smith's termination. That a man was ultimately hired to replace Mr. Smith in no way legally negates the fact that Mr. Smith was fired for being a man.
Yes, you are fighting Bostock. When a person is fired for being gay, sex is an issue no matter what the employer thinks.
So the analysis Ilya Somin posted here was wrong. Just like all his other posts.
I have practiced employment law for over 30 years in a Circuit that didn't follow the background circumstances "rule." (Why I put "rule" in quotation marks will appear below.) I suspect that we have had a 40-year long Circuit split mainly because it didn't matter very much in practical terms.
I agree that Title VII equally covers Xs discriminating against Ys and Xs discriminating against Xs. I agree that there should be no special, additional burden on Xs suing Xs to state a prima facie case. It seems likely, and I'd love to see numbers on this, that some small number of X v. X cases got short-circuited prematurely in Circuits that required background circumstances, but I doubt it would be more than a handful of the already fairly small number of X v. X cases. Ames was right and it's good to get things right, practical impact or not.
But I don't think it's a big deal. After all, it is just common sense that Xs rarely discriminate against fellow Xs. A lawyer is free to argue, and a jury is free to conclude, that, because Xs don't generally discriminate against fellow Xs, it didn't happen here. (To be sure, they don't have to conclude that, but they can, and often do.) Even at the summary judgment stage, when deciding whether to send a case to the jury, judges in a Circuit like mine can take this "common sense" into account, but they almost never put much weight on it.
Where the other Circuits went wrong is to elevate common sense into a "rule." Once you make it a "rule," it's bound to be applied mechanically and thoughtlessly now and then, and I'm sure some cases got dumped earlier than they should have. Maybe most of them were losers anyway, but maybe some of them would have come out differently if they were allowed to proceed.
So it's good that we have straightened things out, but not much will actually change on the ground.
Rosman's analysis is at odds with Oncale v. Sundowner Offshore Services, Inc. An employee can succeed in a sex discrimination claim even if the discrimination is by or for members of the same sex, as long as the discrimination is "because of sex" as it would be in the case of sexual orientation. This principle is not unique to sexual orientation.