The LGBT Plaintiffs in the SCOTUS Title VII Cases Do Not Rely on Changed Meaning
by Michael C. Dorf
In my latest Verdict column, I discuss the textualist argument for finding that Title VII covers LGBT discrimination, an issue on which the SCOTUS granted cert last week. I more or less endorse the view expressed by Chief Judge Katzmann of the US Court of Appeals for the Second Circuit: (1) The prohibition on discrimination based on sex encompasses a prohibition on discrimination based on sexual orientation or gender identity because of the necessary connection between both of the latter and sex; and (2) the case law already forbids much sex-role stereotyping of precisely the sort that is ingredient in LGBT discrimination. I consider counter-arguments that purport to work within textualism and find them lacking. I conclude therefore that the only plausible basis for ruling against the plaintiffs would have to rely on the fact that in 1964 the Congress that enacted Title VII did not subjectively intend or expect to forbid LGBT discrimination.
One could frame the issue somewhat differently. On DoL last week Prof Segall praised the "honesty" of Judge Posner, who characterized his court's determination that Title VII covers LGBT discrimination this way: "we, who are judges rather than members of Congress, are imposing on a half-century-old statute a meaning of ‘sex discrimination’ that the Congress that enacted it would not have accepted."
Although Prof Segall and I reach the same bottom line and take broadly similar views about these cases, here I'm going to push back a little against that characterization. My main objection concerns Judge Posner's use of the word "meaning." When courts say that "sex discrimination" encompasses sexual orientation and gender identity discrimination even though the Congress that forbade sex discrimination in employment would not have expected or endorsed that conclusion, they are not imposing a new "meaning" on the term "sex discrimination." Rather, they are holding that the meaning of "sex discrimination" has entailments that the 1964 Congress did not realize. That might seem like splitting hairs, but the difference has implications for the persuasiveness of the plaintiffs' argument with self-described textualist justices.
Sometimes the meaning of a legal text could be said to change because of what the literature terms "semantic drift." Let's take a fairly well-known constitutional example. Article IV, Sec. 4 places a duty on the federal government to protect states that seek the former's aid against "domestic violence." In the late 18th century, that term meant something like rebellion but had not yet acquired the additional meaning it has today of intra-familial violence. Suppose, however, that a state legislature found that it lacked sufficient resources to combat a scourge of child and spousal abuse and thus sought additional resources from Congress. Would Congress have an obligation to provide such resources?
Nearly everyone would say no. Originalists would say no because for them semantic drift cannot be a basis for changed interpretation. To an originalist, original public meaning simply is constitutional meaning. For non-originalists like me, semantic drift could in theory be the basis for a conclusion that a provision of the Constitution has acquired a new meaning, but, as I wrote here, there has been very little semantic drift on matters of importance. And even where there has been semantic drift--as in the "domestic violence" example--ordinary tools of interpretation, such as attention to context, will usually suffice to rule out some unexpected interpretation. Accordingly, semantic drift is not a very important phenomenon as a practical matter.
Nor does the argument for the plaintiffs in the SCOTUS Title VII cases rely on semantic drift. They are not saying "the meaning of sex discrimination has changed," but rather that "we now realize that sex discrimination encompasses discrimination based on sexual orientation or gender identity, because we are no longer blinded by our prejudices."
But wait. Hasn't there been some changed meaning too? In 1964 and indeed, until very recently, it was conventional to say that sex was a biological category while gender was a social one, so that one form of sex discrimination consisted of insisting on gender-stereotypical behavior associated with a particular sex. In recent years, however, and mostly due to efforts by and on behalf of transgender activists, many people have come to see sex itself as a social category or perhaps as a complex of mental states, but not as simply a biological state of genes and/or anatomy. When trans folk say they would prefer references to "sex assigned at birth" to "biological sex" or the like, they are making the point that sex is not simply a fact in the world but a result of social norms and practices.
Not everyone accepts that sex is a socially constructed category. For example, here is an argument by MIT philosophy professor Alex Byrne for the proposition that sex is not socially constructed. Who's right? Let me suggest that it doesn't matter, for two reasons.
First, the right of people who are trans, non-binary, genderqueer, or otherwise non-traditional to live their lives as they see fit without suffering discrimination or opprobrium should not in any way depend on whether sex is a socially constructed category. (In another essay on a related subject, Byrne appears to agree. There, he says that "a philosophically sound case for treating everyone with dignity and respect has absolutely no need of" the related claim that sex is not binary.)
Second, the argument that transgender discrimination is a species of sex discrimination in the Title VII cases is perfectly compatible with the notion that sex is a biological rather than a social category. Put differently, whatever one thinks about whether sex should now be understood as a social category, the Court can respect the presumed semantic intention of the Congress that adopted Title VII to use "sex" as a biological category and still rule for the plaintiffs. Indeed, that is more or less how the lower courts that ruled for the plaintiffs proceeded.
To be sure, there is language in some of the lower court opinions that might be thought to rely on notions of sex as social rather than biological. Consider the following statement from the Sixth Circuit in the transgender Title VII case which the SCOTUS has decided to review: "discrimination 'because of sex' inherently includes discrimination against employees because of a change in their sex. Here there here is evidence that [the employer] at least partially based his employment decision on [the employee's] desire to change her sex."
One might think that the possibility of a "change" in sex implies that sex is social rather than biological, but it turns out that's not right. A status can be biological but also mutable. Pregnancy is an example. So are a host of other biological conditions. Moreover, even if a transphobic employer were to say that sex cannot be changed--because this employer defines sex chromosomally--the employer's decision to fire the trans employee would still be based on sex as a biological category, because the employer would be treating an employee who wants to live as a woman but has XY chromosomes and is thus, from the employer's perspective, a man, differently from an XX employee who wants to live as a woman. The conception of sex as used in Title VII as a biological category provides affirmative support for the trans plaintiff's case.
To be clear, I don't mean to be taking a position on whether sex is in some sense really a biological or social category. My very limited point here is simply that the argument for the trans plaintiff in the SCOTUS Title VII case does not rely on the claim that sex is socially constructed. It works perfectly well treating "sex" in Title VII as referring to biological categories of male and female--which is undoubtedly how the 1964 Congress would have understood the term. Accordingly, the contention of some trans activists and their allies that (in other contexts) sex is socially constructed does not turn the current cases into changed-meaning cases.
In the end, the LGBT cases before the SCOTUS seek dynamic statutory interpretation in a way that should be very familiar from related constitutional cases. The framers and ratifiers of the Fourteenth Amendment didn't intend or expect that they were forbidding de jure racially segregated schools or most forms of official sex discrimination. Originalists have nonetheless made their peace with Brown v. Board and Frontiero v. Richardson (which first applied heightened scrutiny to sex-based classifications) by accepting that the fixed meaning of the concept of "equal protection" has entailments that the framers and ratifiers did not anticipate. These were cases of changed understandings of facts and the erosion of prejudices that blinded earlier generations; they are not cases of changed meaning; neither are the LGBT Title VII cases.
In my latest Verdict column, I discuss the textualist argument for finding that Title VII covers LGBT discrimination, an issue on which the SCOTUS granted cert last week. I more or less endorse the view expressed by Chief Judge Katzmann of the US Court of Appeals for the Second Circuit: (1) The prohibition on discrimination based on sex encompasses a prohibition on discrimination based on sexual orientation or gender identity because of the necessary connection between both of the latter and sex; and (2) the case law already forbids much sex-role stereotyping of precisely the sort that is ingredient in LGBT discrimination. I consider counter-arguments that purport to work within textualism and find them lacking. I conclude therefore that the only plausible basis for ruling against the plaintiffs would have to rely on the fact that in 1964 the Congress that enacted Title VII did not subjectively intend or expect to forbid LGBT discrimination.
One could frame the issue somewhat differently. On DoL last week Prof Segall praised the "honesty" of Judge Posner, who characterized his court's determination that Title VII covers LGBT discrimination this way: "we, who are judges rather than members of Congress, are imposing on a half-century-old statute a meaning of ‘sex discrimination’ that the Congress that enacted it would not have accepted."
Although Prof Segall and I reach the same bottom line and take broadly similar views about these cases, here I'm going to push back a little against that characterization. My main objection concerns Judge Posner's use of the word "meaning." When courts say that "sex discrimination" encompasses sexual orientation and gender identity discrimination even though the Congress that forbade sex discrimination in employment would not have expected or endorsed that conclusion, they are not imposing a new "meaning" on the term "sex discrimination." Rather, they are holding that the meaning of "sex discrimination" has entailments that the 1964 Congress did not realize. That might seem like splitting hairs, but the difference has implications for the persuasiveness of the plaintiffs' argument with self-described textualist justices.
Sometimes the meaning of a legal text could be said to change because of what the literature terms "semantic drift." Let's take a fairly well-known constitutional example. Article IV, Sec. 4 places a duty on the federal government to protect states that seek the former's aid against "domestic violence." In the late 18th century, that term meant something like rebellion but had not yet acquired the additional meaning it has today of intra-familial violence. Suppose, however, that a state legislature found that it lacked sufficient resources to combat a scourge of child and spousal abuse and thus sought additional resources from Congress. Would Congress have an obligation to provide such resources?
Nearly everyone would say no. Originalists would say no because for them semantic drift cannot be a basis for changed interpretation. To an originalist, original public meaning simply is constitutional meaning. For non-originalists like me, semantic drift could in theory be the basis for a conclusion that a provision of the Constitution has acquired a new meaning, but, as I wrote here, there has been very little semantic drift on matters of importance. And even where there has been semantic drift--as in the "domestic violence" example--ordinary tools of interpretation, such as attention to context, will usually suffice to rule out some unexpected interpretation. Accordingly, semantic drift is not a very important phenomenon as a practical matter.
Nor does the argument for the plaintiffs in the SCOTUS Title VII cases rely on semantic drift. They are not saying "the meaning of sex discrimination has changed," but rather that "we now realize that sex discrimination encompasses discrimination based on sexual orientation or gender identity, because we are no longer blinded by our prejudices."
But wait. Hasn't there been some changed meaning too? In 1964 and indeed, until very recently, it was conventional to say that sex was a biological category while gender was a social one, so that one form of sex discrimination consisted of insisting on gender-stereotypical behavior associated with a particular sex. In recent years, however, and mostly due to efforts by and on behalf of transgender activists, many people have come to see sex itself as a social category or perhaps as a complex of mental states, but not as simply a biological state of genes and/or anatomy. When trans folk say they would prefer references to "sex assigned at birth" to "biological sex" or the like, they are making the point that sex is not simply a fact in the world but a result of social norms and practices.
Not everyone accepts that sex is a socially constructed category. For example, here is an argument by MIT philosophy professor Alex Byrne for the proposition that sex is not socially constructed. Who's right? Let me suggest that it doesn't matter, for two reasons.
First, the right of people who are trans, non-binary, genderqueer, or otherwise non-traditional to live their lives as they see fit without suffering discrimination or opprobrium should not in any way depend on whether sex is a socially constructed category. (In another essay on a related subject, Byrne appears to agree. There, he says that "a philosophically sound case for treating everyone with dignity and respect has absolutely no need of" the related claim that sex is not binary.)
Second, the argument that transgender discrimination is a species of sex discrimination in the Title VII cases is perfectly compatible with the notion that sex is a biological rather than a social category. Put differently, whatever one thinks about whether sex should now be understood as a social category, the Court can respect the presumed semantic intention of the Congress that adopted Title VII to use "sex" as a biological category and still rule for the plaintiffs. Indeed, that is more or less how the lower courts that ruled for the plaintiffs proceeded.
To be sure, there is language in some of the lower court opinions that might be thought to rely on notions of sex as social rather than biological. Consider the following statement from the Sixth Circuit in the transgender Title VII case which the SCOTUS has decided to review: "discrimination 'because of sex' inherently includes discrimination against employees because of a change in their sex. Here there here is evidence that [the employer] at least partially based his employment decision on [the employee's] desire to change her sex."
One might think that the possibility of a "change" in sex implies that sex is social rather than biological, but it turns out that's not right. A status can be biological but also mutable. Pregnancy is an example. So are a host of other biological conditions. Moreover, even if a transphobic employer were to say that sex cannot be changed--because this employer defines sex chromosomally--the employer's decision to fire the trans employee would still be based on sex as a biological category, because the employer would be treating an employee who wants to live as a woman but has XY chromosomes and is thus, from the employer's perspective, a man, differently from an XX employee who wants to live as a woman. The conception of sex as used in Title VII as a biological category provides affirmative support for the trans plaintiff's case.
To be clear, I don't mean to be taking a position on whether sex is in some sense really a biological or social category. My very limited point here is simply that the argument for the trans plaintiff in the SCOTUS Title VII case does not rely on the claim that sex is socially constructed. It works perfectly well treating "sex" in Title VII as referring to biological categories of male and female--which is undoubtedly how the 1964 Congress would have understood the term. Accordingly, the contention of some trans activists and their allies that (in other contexts) sex is socially constructed does not turn the current cases into changed-meaning cases.
In the end, the LGBT cases before the SCOTUS seek dynamic statutory interpretation in a way that should be very familiar from related constitutional cases. The framers and ratifiers of the Fourteenth Amendment didn't intend or expect that they were forbidding de jure racially segregated schools or most forms of official sex discrimination. Originalists have nonetheless made their peace with Brown v. Board and Frontiero v. Richardson (which first applied heightened scrutiny to sex-based classifications) by accepting that the fixed meaning of the concept of "equal protection" has entailments that the framers and ratifiers did not anticipate. These were cases of changed understandings of facts and the erosion of prejudices that blinded earlier generations; they are not cases of changed meaning; neither are the LGBT Title VII cases.