Tuesday, October 02, 2007

No No NO!

The Supreme Court may be taking aim at the anti-retaliation aspect of employment discrimination law:
The discrimination case the justices granted on Tuesday, which has attracted almost no notice, could nonetheless produce an important shift in the court’s approach to interpreting statutes. The question is whether a law that bars racial discrimination in business dealings, including employment, also prohibits retaliation against those who complain about discrimination.

Ordinarily, the court grants cases only to resolve conflicting interpretations in the lower courts. But in this instance, every federal appeals court to consider the issue has agreed that the statute does apply to retaliation. For the court to grant a case in the absence of a lower-court conflict — as it did in the case decided in June that invalidated voluntary integration plans in two public school systems — is often an indication that the case has been added to the docket as a vehicle for advancing a particular agenda.

The federal law at issue in the new case was originally part of the Reconstruction-era Civil Rights Act of 1866. Known now as Section 1981, it does not mention “retaliation.” Neither do most other anti-discrimination laws. In the past, that has been no barrier to the court in finding that protection against retaliation is inherently part of protection against discrimination.

But support on the court for an approach that goes beyond the margins of the constitutional text has been shrinking. Two years ago, the court ruled 5 to 4 that Title IX, a law that bars sex discrimination in schools, also covers retaliation. Justice Sandra Day O’Connor wrote the majority opinion. It is likely that her successor, Justice Samuel A. Alito Jr., would have been among the dissenters.

Given that the new case, CBOCS West Inc. v. Humphries, No. 06-1431, does not meet the court’s most important criterion for review, it is likely that a new majority granted it in order to cut off the retaliation claim and perhaps also to issue a broader ruling against finding rights that are not spelled out in statutes.

Retaliation is not explicitly written into the statute, but lower courts (and the Supreme Court in analogous cases) have all agreed that not reading in an anti-retaliation provision doesn't just create a loophole -- it means you don't have a law anymore. In addition to being wildly unjust on its own terms (firing someone for making a non-frivolous complaint of discrimination is one of the more sadistic employment actions I can think of), if employers can simply terminate anyone who complains of discrimination, then it will be essentially impossible to ever breach the law. In my prior writings on retaliation provisions (this time in Title VII), I complained that they were too narrow--and anyone who thinks that statutory protection shouldn't extend to Robert Jordan in this case is severely lacking in the soul department. But here, the Court isn't looking at limiting the retaliation provision -- it's gearing up to eliminate it entirely.

I agree with Kos: "Congress should pre-empt this case and provide explicit language in the legislation preventing retaliation for discrimination claims." I've been beating the drums for awhile now to get Congress to beef up the retaliation provisions of its civil rights laws for awhile now, and this offers a key opportunity to do so. Discrimination hurts companies too, and by eliminating the incentive for corporations to simply terminate "troublemakers", we don't encourage more litigation -- rather, we encourage employers and employees to develop procedures to resolve these complaints in a just manner without resort to litigation at all. There are no worse incentives in law than those that deliberately and unnaturally pit the interests of employees and employers against each other. In this case, simply restoring the incentive to talk to each other would represent a huge gain for civil rights, as well as the employer's bottom line.


Simon said...

The report is taking quite a liberty to imply that there's something truly irregular about taking a case without a circuit split ("[o]rdinarily, the court grants cases only to resolve conflicting interpretations in the lower courts"). Rule 10 doesn't require a circuit split, and although as Tom Goldstein and other regular practitioners before the court will tell you the best way to get cert granted is to show a circuit split, sometimes it's enough that the case raises an important issue that the Supreme Court hasn't decided (and right now, the threshold is lower than it might otherwise be, given the unusually scarce docket of the last couple of terms). So far as I recall, there was no circuit split in Lawrence or Casey, for example, but we heard not a word from the New York Times to disparage the court's taking those cases. Where was the circuit split in Hamdan?

Noting the absence of a split is fair game, but this story overinflates the significance of the lack of a split in a fairly transparent attempt to fit the grant into their narrative of a court moving to the right.

David Schraub said...

You're right that the NYT kind of overblows the scarcity of taking cases without circuit splits (though I don't really think eliminating retaliation claims is the type of "really important issue" analogous to, say, school desegregation, or detainee rights, or abortion). But I do think that taking the case in absence of a split is a solid sign that a reversal is coming (not always, but a good prediction), so the NYT is likely correct that the Court will use this case to push jurisprudence to the right.

Simon said...

It likely depends on who voted to grant cert and why. As a general matter, it stands to reason that Justices generally don't vote to grant because they want to affirm (a fortiori without a circuit split). But one can imagine obvious exceptions to that rule, such as District of Columbia v. Heller (decided sub nom. Parker v. District of Columbia by the DC Cir): in that case, one can easily imagine voting to grant cert because you want to affirm and "take the ruling national," so to speak." If there are obvious situations where one might grant to affirm, there are surely non-obvious cases where one might grant to affirm. I'm just saying we should use caution (except in obvious situations such as the detainee case whose name escapes me right now, that they agreed to take right at the end of last term - that seems an almost certain bet to be a 5-4 sequel to Hamdan).

I'm going to try and post in more detail about the case as the briefs become available (I'm tied up writing about Santos at the moment), but it bears noting that the Title IX case they refer to is almost certainly Jackson v. Birmingham Board of Education, 544 U.S. 167 (2005), and as Judge Easterbrook noted in a partial dissent in the instant case, Jackson hardly settles this case. Jackson held that Title IX gives the judiciary lattitude to define the word “discriminate” as that term appears in 29 U.S.C. § 1681, but the word “discriminate” does not appear in § 1981 - so how does a decision resolving ambiguity in the word “discriminate” settle questions involving other statutes with different language (statutes enacted at completely different times and in different circumstances, for that matter)? It seems to stretch the in pari materia canon past breaking point. Moreover, says Easterbrook, there is an anti-retaliation rule in Title VII of the Civil Rights Act of 1964, so saying that there isn't one in § 1981 really only has the effect of saying that this lawsuit (and subsequent ones like it) ought to have been brought under a different cause of action, not that there's no merit in it.

PG said...

There is something unusual about the Supreme Court's taking a wholly non-constitutional case without a circuit split. Lawrence, Casey, Hamdan, school desegregation -- all constitutional. Just among recent Title VII cases, this is different from Ledbetter (11th Cir. differed from the other circuits' and EEOC's interpretation); Arbaugh (5th differed from the 7th); Burlington Northern, itself a retaliation case (9-0 decision in favor of the 9th's standard instead of the 6th's)...

Among the the other statutory cases this coming term, even Stoneridge, which strikes me as a slam dunk affirmation of the 8th and 5th Circuits, has a split (thanks to the 9th).

For more purely statutory cases:

Watson v. U.S. (1st, 3rd, 4th, 5th, 8th & 9th on one side; 6th, 7th, 11th & DC on the other);
Bd of Ed v. Tom F. (1st splits with 2nd);
Riegel v. Medtronic (11th splits with everyone else);
LaRue v. DeWolff (4th splits with 7th on the section 502(a)(3) issue);
Hall St. v. Mattel (1st, 3rd, 4th, 5th, 6th split with 9th & 10th).

I could keep methodically working through the docket, but I need to sleep. However, I am guessing that someone who finished canvassing the list probably won't turn up more than a single other case on purely statutory grounds in which there could be a circuit split, but none exists. On a list with a score of purely statutory cases, finding only two that don't have circuit splits where one is possible seems like a reasonable indication that the Court's granting cert is indeed unusual.

What I mean by could, is that pretty much the only purely statutory issues that don't require a circuit split to be heard by the Court, are those handled solely by the Federal Circuit, e.g. intellectual property (Quanta v. LG) and benefits suits against the federal gov't. If no other circuit sees the issue, obviously there can't be a split.

Also, when I say "purely statutory," I mean there isn't a mixed question of constitutional and statutory/ regulatory law. Rowe v. NHMTA and KY v. Davis both are federalism questions.