Philosophers File Amicus Brief on LGBT Discrimination Cases
This fall, the Supreme Court will be looking at a trio of cases concerning the protection Federal employment discrimination law provides to lesbian, gay, bisexual, and transgender persons. An amicus curiae brief signed by 80 philosophers, in support of the employees in the cases, has just been filed.
The amicus brief was co-authored by philosopher Robin Dembroff (Yale) and law professor Issa Kohler-Hausmann (Yale), and concerns the Supreme Court’s review of Bostock v. Clayton County, Georgia, and Altitude Express, Inc. v. Zarda (both cases concern whether federal laws banning employment discrimination protect gay and lesbian employees), and R.G. & G.R. Harris Funeral Homes v. EEOC (on whether Title VII’s protections apply to transgender employees). Some background on these cases can be found here.
The following is the “summary of argument” section of the 50 page document:
1. The concept of “sex” is inextricably tied to the categories of same-sex attraction and gender nonconformity. Both categories are partially defined by sex and cannot logically be applied to any individual without reference to that individual’s sex. It is simply not possible to identify an individual as being attracted to the same sex without knowing or presuming that person’s sex. Likewise, it is not possible to identify someone as gender nonconforming (including being transgender) without reference to that person’s known or presumed sex and the associated social meanings. It follows that discrimination on the basis of same-sex attraction or gender nonconformity is inherently discrimination “because of sex.”
2. It is conceptually incorrect to state that discrimination against persons who are same-sex attracted or gender nonconforming is “sex-neutral.” If an employer decides to terminate an employee on the basis of same-sex sexual attraction (i.e., a particular sexual orientation) or gender nonconformity (e.g., being transgender), the employer must first presume the employee’s specific sex, and then account for the social meanings, expectations, and stereotypes specific to the employee’s particular presumed sex category. But for the concept of sex, the judgment that an employee violated one of the expectations and stereotypes specific to their sex would be impossible.
3. Title VII prohibits discrimination not simply based on the categories “man” and “woman,” but because of sex. The philosophical underpinnings of antidiscrimination laws represent a societal commitment to alter socially restrictive categories such that they no longer serve as the basis for denying equal treatment or limiting freedoms based on sex. To permit discrimination against individuals who fall into categories that are partially defined by sex would violate the fundamental rationale behind antidiscrimination laws. Moreover, it would require this Court to define “sex” in a way that is illogically constrained and harmful to groups that have historically been the targets of discrimination.
You can read the entire brief here. The oral arguments for the case will be heard on October 8th, 2019.
I’m interested in what people think of this in light of the rest post on Agnes Callard on petitions. An amicus brief isn’t exactly like a petition, of course. But, as a law professor working in an area where there are regular calls for people to “sign on” to Amicus briefs, it’s also clear that the difference is not as great as one might expect – the typical pattern is for the brief it be written by one person or a small group, and then other people to “sign on” so as to “add weight” to it. (It’s clear from the list of signers here that lots of them have no special expertise in the area, so it can’t be the case that they are all signing because they are experts and agree with it as a matter of professional expertise.) I don’t have any special objection to this. It’s a pretty common practice, but also one that courts typically don’t give all that much weight to. (They might, sometimes, care about the arguments, though in truth they are not actually always even read. Courts care very little about having a long list of people sign.) But, I would be curious to see what people think about this instance, in light of the Callard piece – especially having a lot of people, including non-experts, sign on it, as opposed to just having the argument made. (Having the argument made, on its own, is of course fine, but that could have been done just by the authors, w/o the other “signers”.)
(This isn’t a comment on the content – from a quick skim, that seems pretty good, but only the comparison w/ Callard’s recent article.)
Hi Matt,
Callard, I had thought, was objecting to petitions without philosophical arguments for positions. This amicus brief clearly gives arguments, some of them very interesting.
No, she was objecting to them unconditionally.
This is a bit of a shot in the dark, but I wonder if Agnes Callard reads this blog and can answer this, because I’m curious as well…
Hello Matt and K!
I was not objecting to petitions unconditionally–though I grant that the opening phrasing of my point suggested this. I was also not objecting to them on the condition that they fail to present arguments. Rather, I was objecting to petitions by philosophers (qua philosophers) speaking to philosophers (qua philosophers) and addressing philosophical questions (how we should conduct philosophical conversations)–in this context, I believe the attempt to lend force to a claim (or argument) through amassing signatures is problematically unphilosophical. I do want to credit the subsequent discussion on twitter for helping me sharpen the point in this way.
Matt is right that an amicus brief is a funny case, because it is not a petition and doesn’t require more than one signature. I am not sure what to think. However, I am inclined to think that this action by philosophers *is* problematic, here is my argument.
The document was written by philosophers, and it does address a philosophical question (does discrimination against people on the basis of their sexual orientation, or their being transgender, count as gender discrimination?). The question is: is it addressed *to* philosophers? The answer is non obvious, but either way there is a problem.
Suppose yes–that is, suppose that at the level of the Supreme Court the distinction between philosophical and legal activity is no longer a valid one to make. Then I think the argument in my NYT column applies–philosophers should address one another with arguments (alone), and not attempt to lend force to their arguments using lists of names.
Now suppose no. Since the (academic) philosophers are still writing *as philosophers*, they would then be leveraging their professional expertise into persuasion: you non-philosophers should believe that this kind of discrimination counts as that kind of discrimination *because we professional philosophers believe this.* I don’t deny that there is such a thing as philosophical expertise, but I don’t think it works this way–it doesn’t support knowledge by testimony in the way that, e.g., a physicist’s knowledge of physics does.
So this is what I’m inclined to think, but I have some doubts:
(1) Perhaps my clean-cut distinction misses the point–if Supreme Court justices are a borderline case of philosophers, that might matter.
(2) I don’t know much about amicus briefs, and there could be facts I am missing that would change my view, e.g. if appearing on the list served to indicate that you would be willing to testify in the hearing.
(3) There will be a hearing, and arguments will be made there–I don’t really know the relation between the list of names and that forum. If the function of the list were *only* to draw attention to the claims as worthy of attending to in such a hearing, that would matter.
So I don’t feel very sure about this.
But those are my thoughts.
Thanks for asking!
Amicii rarely are given a voice at hearings before the Supreme Court. None come to mind but it may be possible. The number and identities of the philosophers are unlikely to matter to the Supreme Court. None are known for familiarity with contemporary philosophy in the US (or anywhere else). The argument is kind of an outlier given the Supreme Court’s usual approach to statutory interpretation.
I should be a bit more accurate. Amicus is a third party. In very rare circumstances, a third party may be invited by the Court to participate in argument. Not in this case and certainly not this amicus. The hearing is oral argument on briefs. There is no evidence or testimony taken, not in this case. The brief will join a large stack of amici filings on both sides of the case. The Court tends (increasingly) to extend originalism to statutory interpretation, which is a problem for this sort of brief.
Stephen Breyer is familiar with quite a bit of contemporary philosophy. As an undergrad I worked in a business that he frequented and spoke to him about philosophy a number of times. His undergrad degree is in philosophy and he has a PPE degree as well.
J. Bogart is mistaken in claiming that “none [of the Justices] are known for familiarity with contemporary philosophy in the US (or anywhere else).” Gorsuch has had graduate-level training in philosophy at Oxford, and his dissertation was in applied legal philosophy.
Michaela and Grad1: Quite right! My error. Thank you for the corrections. I was thinking too narrowly about opinions.
Thank you for answering!
This is quite interesting. While I believe that discrimination on the basis of sexual orientation is seriously immoral and ought to be against the law, it isn’t entirely clear to me how the main argument on pages 4-5 is supposed to show that discrimination on the basis of sexual orientation *is* discrimination on the basis of sex (construed purely biologically).
The argument claims that for a male to be same sex attracted is to be (1) a male and (2) attracted to males. So, discriminating against same-sex attracted males is discrimination on the basis of sex. That seems correct. However, why would an employer who discriminated against all same-sex attracted persons be discriminating on the basis of sex? They would not be distinguishing between persons on the basis of their sex but upon their having the property of being same-sex-attracted (which property can be had by both males and females).
The authors’ provide an analogy of the “concept of being a wife” which requires that one be “both (i) a woman [female?] and (ii) married.” They claim, plausibly, that if “an employee was terminated for being a wife that termination would be because of sex.” However, a more apt analogy for terminating same-sex attracted persons would seem to be a policy of terminating married persons. That policy (while clearly wrong) would not be distinguishing between people on the basis of sex.
I hope I am not just missing something obvious.
Check out pages 8-12.
Answering ‘Curious’,
Thank-you. I did read them but I found the argument a bit confusing.
The central relevant portion seems to be the argument from the bottom of page 9 to the top half of page 10. The first argument is that “simply because an employer’s policy can be described at a level of abstraction that appears category-neutral . . . does not mean that the policy is in fact category neutral.” Whether this is correct depends in part on whether “describe” is a success term. Of course an employer can lie about their policy. A person who has a policy of terminating wives might claim (falsely) to have a policy of terminating married people. If that counts as “describing” it, then the authors are correct but the issue seems irrelevant as the two policies outlined are distinct even if one is “more abstract.”
The authors then give an example of an employer who terminates a Jewish employee who does not observe Yom Kippur. When “asked to explain his reasons, the employer responds that he would have terminated anyone who is religiously non-observant” and that his policy is “religion neutral.” Setting aside the irrelevance of possible dishonesty regarding his actual policy, let us suppose he has a policy of terminating anyone who is “religiously non-observant.” On a natural reading, this discriminates on the basis of religious belief as those without religious belief will count as non-observant and so isn’t religion neutral. (I am here relying on the authors’ suggestion that prohibited basis of discrimination are “exhaustive forms of categorization.)
Another case: Suppose an employer had a policy of not-hiring those lacking sexual attraction to others (asexuals). Again, that would be clearly wrong (and extremely difficult to apply) but would it be prohibited by laws against discrimination by biological sex? I think we shouldn’t confuse what we wish was illegal from what is illegal.
I agree with Curious. Such discrimination is seriously immoral, but I’ve read this document multiple times and am struggling to see how this inference (section 1) isn’t fallacious.
Conceptualizing relative sexual orientation as the relevant property seems to be doing a lot of the undermining here. Conversely, I think the argument depends on considering as relevant conceptualizing same-sex attraction from a non-relative standpoint. By this I mean if you have a blanket ban on same-sex attraction, that only doesn’t involve sex if you tend that indexed property to be basic, and for no other conceptualizations to be important to the matter. But if you think being, e.g., a lesbian reduces to being a woman and being attracted to other women, then there’s no avoiding the reference to sex.
The counterfactual usually does it for me. If Liz is fired for her same-sex attractions, then it’s the case that she’s being fired for being attracted to women while being a woman, whereas if her attraction to women were held constant but she were a man, then she wouldn’t be fired. Looks there like her firing is contingent on her sex.
The part of the argument that convinced me was the point about application: even if we accept that a policy that is discriminatory towards homosexual or transgender people is sex-neutral, the implementation of that policy involves essential reference to the subject’s sex.
Curious, I think you’re right that, unfortunately, the brief doesn’t do the best job responding to your concern. But I do think the logic is sound.
I believe the critical point is that one can’t determine whether someone is same-sex attracted or gender non-conforming *without looking at their presumed sex/gender*. Being a wife is like this. But being married isn’t analogous: one can just look at marriage records.
The “property of being same-sex-attracted” is really disjunctive: it’s the property of being either a woman attracted to women, or a man attracted to men. There’s no way to tell whether someone is same-sex-attracted in general without referencing their sex. Being religiously non-observing person is a good analogy; it’s similarly disjunctive.
Here’s another analogy: Suppose I don’t hire any person who has the property of being a “No-No”. To be a No-No is to be either a qualified female applicant or to be an unqualified male applicant. This is clearly discriminating on the basis of sex, even though I can describe my policy in a way that appears category-neutral, and it applies to both men and women (*all* No-Nos)!
I agree that the analogy with religious non-observance isn’t quite apt. A better analogy might be an employer who has a blanket policy against miscegenation – they won’t employ anyone who’s married to someone of a different race. It seems to me that if a policy of not employing homosexuals is sex-neutral at some level of description then this anti-miscegenation policy is race-neutral at the exact same level of description.
(Speaking to the conceptual point here, obviously there are historical and political disanalogies)
I think analogies like this one aren’t really helpful for thinking about the underlying logic, since describing an employer who’s against miscegenation stirs up a “that’s bad” intuition that’s easy to mistake for a “that’s illegal according to the relevant statute” intuition. It would be useful to think about cases where our view about the general moral status of the thing is less negative or anyway not as cut and dry. Here are two.
Suppose an employer discriminates against people who are married to members of the same race. I don’t know that I’d want to say this policy isn’t “race-neutral,” and I definitely wouldn’t consider it discrimination on the basis of race.
Or take another case. Suppose an employer wants to fire all racists, but thinks that it’s impossible to be racist against your own race. Then an individual’s own race is clearly relevant to whether or not they get classified as racist — the employer “cannot identify the people he seeks to exclude without reference to” the person’s race. But that doesn’t mean the policy isn’t race-neutral, and it very clearly does not mean the employer is discriminating on the basis of race.
So I think the logic of the inference — i.e. that if a policy requires reference to which category someone fits in X-wise, then the policy is not X-neutral, or even further that the policy therefore constitutes discrimination of the basis of X (or “because of” X as the document sometimes put it) — is just not right.
I agree that those are better cases for carrying less baggage. But it’s not clear to me that those wouldn’t be cases of racial discrimination (particularly the first one).
Devin,
That seems like a better analogy. I remain a bit uncertain whether such a policy (while disgusting) would violate the plain language of Title VII, though I gather that the 2nd Circuit found that it does violate Title VII (Holcomb v. Iona College – “an employer may violate Title VII if it takes action against an employee because of the employee’s association with a person of another race.”). Interestingly, the 2nd Circuit suggests that a white man discharged for such a reason is being discharged because of his race in virtue of the fact that had he been black (and still married to a black woman), he would not have been discharged.
“Both categories are partially defined by sex and cannot logically be applied to any individual without reference to that individual’s sex.”
The term “homosexual” cannot be applied to an individual without *reference* to their sex, but their sex may be a unknown to the employer. That is, one can imagine an employer discriminating against an employee who they know to be homosexual (say, based on personal testimony) but whose sex remains unknown. It is hard to say, then, how this sort of instance of orientation-discrimination would be “based on sex”. Similarly, or even more dubiously, terms like “bisexual” do not seem to me to make any reference to a bisexual individual’s sex.
But if the person you are going to fire would have had a different sex, then, other thing being equal, the person won’t be fired. That line of reasoning might work. However it runs into trouble for bisexuals.
If an employee reason to discriminate against someone is that that person is bisexual as such – that is, the employer is discriminating on the grounds that the person is attracted to multiple genders – then I agree that that it’s not clear how the reasoning in the brief applies. If, as I think is more likely to occur, the employer is instead discriminating on the basis that one of the genders the bisexual person is attracted to is their own, then the reasoning applies exactly as it does in the homosexual case.
*If an employer’s…
I don’t think the reasoning is quite the same. One might argue that discriminating against homosexuals is a case of discrimination because of sex, for if someone’s sex had been different, other things being equal, they would not have been a homosexual. So this is a true “because”. However a bisexual with a different sex would still be bisexual.
Discrimination against homosexuals might be thought of as “because of sex”, for the reason that if someone’s sex had been different, then, other things being equal, this person would no longer be a homosexual. However a bisexual would still be a bisexual, regardless of a change in sex.
Right, what I’m trying to pin down is the relevant notion of reference needed to make true these statements about necessary referring. If Bart is right, that such discrimination against someone who is homosexual/gay necessarily refers to the individual’s sex because of modal facts about their sex, then this seems like it wouldn’t work for bi or pansexuals. I’m also skeptical of this way of fleshing out “refer” as well, but agree with Bart that it may work (though not for bi or pansexuals, etc).
“If, as I think is more likely to occur, the employer is instead discriminating on the basis that one of the genders the bisexual person is attracted to is their own,” I agree this seems more likely, but it also seems like it misses the modal notion of “refer” that seems to be necessary to make the idea work. Which, it seems to me, brings us back purely to orientation-based discrimination (since the same-sex orientation would be present with a bisexual person regardless of their own sex).
If I understand the principle correctly, it really is an unreasonable one.
Here’s an illustration. Suppose Acme Inc. wants to regularly survey its employees to ensure that, among other things, women are happy working at the company. For this reason, the survey (which is otherwise anonymous) asks people to indicate whether or not they’re women. Each employee can only take the survey once, but far more people click the ‘woman’ box than there are women working for Acme. It becomes clear that some men are pretending to be women when they take the survey, which throws off the results. This continues after several warnings, and some male employees on an anonymous forum boast about gaming the system. Eventually, Acme management announces that any men caught filling in the survey as women will be subject to an immediate performance review and may be fired.
Acme management then catches Jack, a male employee who makes no bones about the fact that he’s male, fraudulently filling in a survey as a woman. Jack gets fired. But had Jack not been male, or had management not known that Jack was male, Jack would not have been fired.
Does it follow from this that Acme management is sexist or guilty of sex discrimination? It sure doesn’t seem so. But it seems to follow from the principle being stated here that it is.
I think this is exactly right. Even if there is always an implicit reference to sex, what matters is that the discrimination is made *on the basis* of sex. Some response to a situation may necessarily always make reference to to individual’s sex (or anything else) without the discrimination being something made “on the basis” of the person’s sex (or anything else.
“An employer may not…condition employment opportunities on the satisfaction of facially neutral tests or qualifications that have a disproportionate, adverse impact on members of protected groups when those tests or qualifications are not required for performance of the job….[And] an employer shall not be liable if it can prove that, even if it had not taken gender into account, it would have come to the same decision regarding a particular person.” [Price Waterhouse v. Hopkins 1989]
Is Acme’s survey “gender-neutral”, and is its success a worthy objective that all employees should support? Acme can argue that they would have reached the same decision regardless of bad Jack’s gender – the method by which he actually sabotaged the survey is a distraction. Acme’s reasoning does not rely on a stereotype or expectation about behaviour of an entire class.
Thanks, David. I agree that Acme didn’t act wrongly here, for sure. But I don’t think the reason is what you give, namely that Acme ‘would have reached the same decision regardless of bad Jack’s gender — the method by which he actually sabotaged the survey is a distraction’. If Jack had not been male, he would not have sabotaged the survey at all by his clicking the ‘female’ box on the survey, ,and would not have been disciplined at all.
The reason, it seems to me, why Acme is not in violation is that they _have_ used a gender-neutral method of determining that Jack acted wrongly: they’ve consistently expected all employees to follow the gender-neutral rule, ‘Don’t make false statements when you fill in the survey’. So, they’re fine. But if we follow the principle proposed in this amicus brief (as I understand it), any employer that terminates employees for doing something that can only be known to be a violation if the employer considers the sex of the employee is forbidden. The Acme case, I think, shows that that principle is implausible.
Sorry, garbled the end of that penultimate sentence. But the point should be clear.
Dear Justin,
I know this thread is probably over, but I wanted to chime in anyway.
While I was one of the signatories to the brief, below I’m just speaking on my own behalf; I don’t know if the authors or other signatories would agree with what I say here.
My inclination is to concede that the case you describe falls within the purview of the principle, and also to agree that the employer’s conduct in that case both is and should be legal. But I don’t think that’s fatal for the principle.
Consider 14th amendment jurisprudence as it applies to affirmative action policies at state universities. Judges who’ve ruled in favor of the legality of affirmative action policies have typically not said that they’re not prima facie discriminatory. Rather, all parties agree that they are prima facie discriminatory and thus trigger strict scrutiny, but defenders of the policies argue that they can pass such scrutiny, in virtue of being narrowly tailored to advance a compelling state interest (roughly).
So abstractly, the idea isn’t that the right definition of discrimination will entail that discriminatory policies are all thereby illegal; rather a policy’s being discriminatory triggers heightened scrutiny, but it’s possible to pass such scrutiny.
I don’t know employment discrimination law as it applies to private actors well enough, but I’d hope there’d be some kind of similar structure, so that employment practices that discriminate along certain lines are prima facie suspect, but not thereby automatically illegal. Consider, e.g., a theater company that explicitly rules out hiring light-skinned actors to play Othello, and which likewise rules out hiring dark-skinned actors to play Iago. While I can imagine attempting to come up with a definition of employment discrimination on which that doesn’t count as racial discrimination, much easier, I think, to concede that any general, remotely formal characterization of discrimination will include such a practice, but that such a practice ought to pass whatever heightened scrutiny is appropriate to racial discrimination in employment.
As is probably already clear, I think we should say something similar about your case. I think it will be hard to formulate a principled, general definition of sex discrimination–one that’s not overly restrictive–that won’t cover your case, but I think it’s also plausible that the discrimination in that case is narrowly tailored to advance a legitimate interest, so it should pass whatever heightened scrutiny is warranted for discriminatory employment policies.
People don’t get fired for finding men attractive. They get fired for doing so while being a man.
People don’t get fired for finding women attractive. They get fired for doing so while being a woman.
In both cases, the attraction isn’t sufficient for firing someone. Sex and attraction are both necessary.
Sex not being the sole factor doesn’t mean that this isn’t discrimination based on sex.
The fact that men and women could both be discriminated against doesn’t mean that this isn’t discrimination based on sex.
“Sex not being the sole factor doesn’t mean that this isn’t discrimination based on sex.”
Sure, but the examples discussed above, particularly Justin’s, make me doubt whether sex’s being a “factor” in the sense that it must be *referenced* means that it is a factor in the sense of a basis.
Calling the role of sex mere reference when its a necessary condition for the discrimination strikes me as a bit odd. Saying “Only this population [male-sexed individuals] can engage in this behavior [having romantic or sexual relationships with women]” is just having different sets of rules for people based on sex.
I appreciate the spirit and purpose of this, and agree that notions such as ‘gay’, ‘trans’, and ‘gender non-conforming’ are inextricably tied to sex. I will also support anything I believe works to help achieve human rights, including anything that helps achieve the right verdicts on these cases (i.e. the ruling that all 3 cases involve violations of Title VII). That being said, I disagree that discrimination based on sexual orientation or gender identity is sex-based discrimination.
Sex-based discrimination is discrimination based on the sex of the person who is a potential target of discrimination, not the sex of those they are attracted to or the sex they wish to be/identify as. While those all in some way involve the notion of sex, it is not the case that any discrimination that in any way can be connected to the sex of anyone, or to a relational property whose relata include sex, is discrimination based on sex. In general, the protected characteristics have to be possessed by the person who is a potential target of discrimination. The intention in including sex as a protected characteristic is to prohibit discrimination against women, and this sort of expanded interpretation of sex-based discrimination goes against that intention, while also being unnecessary and philosophically confused.
This sort of situation has already arisen in various places where equality legislation/constitutions list protected characteristics, but gender ID, gender expression, and sexual orientation are not included in the list. Usually there is an ‘or other’ clause at the end, in recognition of the real possibility that the list is not exhaustive. The exclusion of sexual orientation, gender ID, and gender expression is because of homophobia and the lack of salience of trans issues and gender identity discourse at the time of drafting these laws/constitutions. However, the same reason for including in the list of protected characteristics what was originally included, e.g. sex and race, is a reason for including the aforementioned 3 characteristics.
A protected characteristic is a property of a person that puts them in a position of marginality or oppression, and which is irrelevant to the ability of the person to perform properly at their job (this is in the context of employment discrimination). Clearly sexual orientation, gender ID, and gender expression fall under this definition, and so Title VII must be applicable to them. In other words, we can apply the spirit instead of the letter of the law, and it isn’t even a far-fetched or speculative interpretation of the spirit of the law, but a very intuitive and justified application of what is explicitly stated and in every way implied to be the purpose of the law.
I’m not sure why the lawyers pursuing these cases decided to go with the broadening-the-definition-of-sex-based-discrimination approach rather than the approach I described above. Maybe there is some legal expedience that I don’t know about since I’m not a lawyer. In any case, from my standpoint the line of argument they decided to take is highly unsatisfying, but either way I still hope they are successful in getting the 3 cases ruled to be violations of Title VII.