« Jeremy Telman: John Marshall's Constitution
Michael Ramsey
| Main | Eric Segall on Discretionary Originalism
Michael Ramsey »


A Response to John Vlahoplus’ “Textualism and Title VII” [Updated with a Further Exchange]
David E. Weisberg

A recent post from John Vlahoplus criticizes my earlier post (which commented on an article by Prof. Richard Primus) for arguing “from the logic of contemporary English usage rather than early 1960s usage,” and for failing to cite “dictionaries, other reference works, or other sources from the corpus of English usage around 1964.” 

I consider myself a textualist but not an originalist (for reasons set forth here), so I don’t approach Title VII with a presumption that the words and phrases in the statute had a different meaning in 1964, when it was enacted, than they have today.  Moreover, Mr. Vlahoplus’ post makes no reference whatsoever to any change in logic, usage or meaning of any word or phrase in Title VII from 1964 until today.  In fact, he himself relies on two definitions—of ‘discriminate’ and ‘because of’—from the current, up-to-the-minute online edition of the Oxford English Dictionary, and those definitions have not changed in the interval between 1964 and today.  So, there seems to be agreement that, with respect to Title VII, current meanings and 1960s-era meanings are identical.

Mr. Vlahoplus notes that Title VII includes the word “discriminate,” and he asserts that “firing someone for being LGBTQ discriminates against them,” because it is “unjust and prejudicial.”  I agree.  But one doesn’t pluck a single word out of a statute to determine the whole statute’s meaning.  Title VII does not prohibit an employer from acting in a way that “discriminates against,” or is “unjust and prejudicial” towards, actual or prospective employees.  Either formulation would, I think, be void for vagueness.  But, in any event, the statute is much more specific.  A relevant question, e.g., is whether an employer who fires a male employee because the employee has relations with other men has “discriminate[d] against” that employee “because of [his] sex.”

Mr. Vlahoplus says the answer is yes, because that answer follows from a concept with which lawyers have long been familiar: ‘but-for causation’.  But for the male employee’s sex, he would not have been fired.  That establishes, we are told, that he was fired “because of” his sex, in violation of Title VII.  (Mr. Vlahoplus makes a point of citing a statement written in 1963: “Plaintiff suffered injuries because of the negligent operation of a cab.”  But that statement has exactly the same meaning today that it had when it was written; Mr. Vlahoplus does not even hint otherwise.  I therefore cannot imagine what special significance supposedly attaches to its origin in 1963.)

Another thing lawyers have long known is that but-for causation is typically relevant in tort cases to determine whether defendant’s actions or omissions caused some physical injury to plaintiff’s person or damage to plaintiff’s property.  That is, the concept of but-for causation was developed by common-law courts to deal with complicated causal chains that produce effects in a world governed by the laws of physics and the other natural sciences. 

The phrase ‘because of’ is not limited to the description of taxi accidents.  Sometimes it is used in statements—“He was fired because of: (a) his sexual relations with other men, or (b) his sex”—that are about motives and decisions, rather than cause and effect in the physical world.  The question then arises: can but-for causation properly be extended to questions of motivation, such as those relevant to Title VII?  I offered, in my earlier note, two reasons for concluding that but-for causation cannot be so extended.  Mr. Vlahoplus does not address either reason. 

The first reason: the Primus/Vlahoplus position, if strictly adhered to, generates plain falsehoods.  If the male employee, under oath, were to insist that he was fired because of his sex, he would have perjured himself.  If he has sworn to tell the truth, the whole truth, and nothing but the truth, he may not properly answer the question, “Why were you fired?”, by repeatedly asserting, “I’m male, and I was fired because of my sex”.  The falsity of that answer would be underlined if his employer, similarly under oath, testified along these lines: “We have employed and continue to employ many men—we don’t discriminate against males—but we fired this particular man because he had sexual relations with other men.”

The second reason is that the Primus/Vlahoplus view leads to an absurd result (and I am confident the reductio ad  absurdum argument form was invented prior to 1964): an employer simultaneously violates and does not violate Title VII.  If their view of but-for causation were correct, one could say that the male employee was fired because of his partner’s sex, rather than because of his own sex.  So, according to Primus/Vlahoplus, the employee was fired because of his sex (a violation of Title VII); but it could equally well be said he was fired because of his partner’s sex (not a violation of Title VII).  Due process is not satisfied when a tribunal arbitrarily picks one of two equally apt descriptions of some event, with the result that a defendant is found to have or not have violated the law.

Although Mr. Vlahoplus does not address either of the objections to but-for causation in the Title VII context, I continue to believe they are valid.  A man fired because of relations with another man is fired because of his sexual behavior or activity, not because of his sex; that statement has the same meaning and is just as true today as in 1964.  The proper way to outlaw discrimination based on LGBTQ lifestyles is to outlaw discrimination based on LGBTQ lifestyles.  Title VII does not currently do that.        

UDPATE: John Vlahoplus replies:

In his latest post on textualism and Title VII, David E. Weisberg fails to address both Richard Primus’ discussion of causation and my hypothetical criminal law example (a law mandating wearing only clothing historically associated with one’s sex is discrimination “because of” sex, not punishment for the act of wearing the “wrong” clothing).  He also fails to notice my statements that the terms “to discriminate” and “because of” have long had the meanings I cite.  The current Oxford English Dictionary bases the definitions on usage from well before 1964.

But let us focus specifically on causation and 1964-era usage.  Professor Primus addresses causation here, implicitly referring to authorities that date to 1964 and earlier:

If you’re tempted by the thought that firing a person for having a same-sex partner doesn’t discriminate on the basis of sex because the employer would fire people of any sex who have same-sex partners, ask yourself whether a law prohibiting people of any race to marry outside their racial groups, or to ride in a railroad car designated for people of a different race, discriminates on the basis of race. (It does.)

In 1964 such discrimination was “because of” race.  See Justice Stewart’s concurring opinion in McLaughlin v. Florida (1964), which invalidated a statute punishing nonmarital interracial cohabitation:

These appellants were convicted, fined, and imprisoned under a statute which made their conduct criminal only because they were of different races. So far as this statute goes, their conduct would not have been illegal had they both been white, or both Negroes . . .  [I]t is simply not possible for a state law to be valid under our Constitution which makes the criminality of an act depend upon the race of the actor.

McLaughlin v. Florida, 379 U.S. 184, 198 (1964) (Stewart, J., joined by Douglas, J., concurring) (emphasis added), here.

Substitute “because they were of the same sex” for “because they were of different races,” and it is clear that the employers’ decision to fire the Title VII plaintiffs “depend[ed] upon the sex of the actor” just as the Florida statute “depend[ed] upon the race of the actor.”   

Discrimination that depends on the sex of the actor being the same as that of his partner is “because of” his sex, just as discrimination that depends on the race of the actor being different from that of her partner is “because of” her race.   A textualist could easily conclude that firing the Title VII plaintiffs for being LGBTQ discriminated against them “because of” their sex, as many understood legal causation in 1964 in discrimination law as well as more generally in tort and contract law.

FURTHER UPDATE:  Final thoughts by David Weisberg:

Mr. Vlahopus is now 0 for 2; he has had 2 chances to address my objections to the ‘but-for causation’ analysis that was central to his original post, and he has never done so.  I think this is an admission that he is unable to do so, because his latest post does not even mention but-for causation.  Instead, he focuses on Justice Stewart’s concurrence in McLaughlin vFlorida, a 1964 case in which the Court unanimously held that a Florida statute criminalizing cohabitation by heterosexual couples of different races violated the 14th Amendment’s Equal Protection Clause.  Justice Stewart’s very brief concurrence states, in relevant part:

I cannot conceive of a valid legislative purpose under our Constitution for a state law which makes the color of a person's skin the test of whether his conduct is a criminal offense. These appellants were convicted, fined, and imprisoned under a statute which made their conduct criminal only because they were of different races…[W]e deal here with a criminal law which imposes criminal punishment. And I think it is simply not possible for a state law to be valid under our Constitution which makes the criminality of an act depend upon the race of the actor. Discrimination of that kind is invidious per se.  (379 US at 198, footnote omitted.)  

Anyone can pick words and phrases out of any opinion, but that approach doesn’t shed any substantial light on whether Title VII (not the Equal Protection Clause), which imposes civil (not criminal) penalties, prohibits a private employer (not a State) from firing an employee who engages in homosexual activity (not activity that necessarily involve persons of different races).