One key ruling — that most class actions seeking monetary compensation cannot be brought under lenient procedures that were originally designed for suits seeking only injunctive relief — was unanimous, sending a sharp rebuke to the U.S. Court of Appeals for the Ninth Circuit, which had held otherwise by a 6-5 margin.

At the same time, the more far-reaching rulings in the case, which relate to the more fundamental question of just how much in common a million and a half women must have before than can sue as a class for gender discrimination, were decided along narrow, familiar ideological lines, 5-4.

Justice Antonin Scalia wrote the opinion of the Court, and Justice Ruth Bader Ginsburg wrote the partial dissent.

Here are the headlines:

Unconscious discrimination. The majority drove a stake — multiple stakes, really — through the heart of a very common, powerful genre of employment discrimination class action that revolves around the claim that a company gives its managers excessive discretion in making pay and promotion decisions, allowing those managers to engage in unconscious discrimination. In the past, similar suits have been brought against the likes of Costco (COST), Home Depot (HD), and FedEx (FDX), and a group of other large corporations, including Altria (MO), Bank of America (BAC), and Hewlett-Packard (HPQ), had filed amicus briefs in which they admitted feeling vulnerable to suits brought on the Wal-Mart (WMT) template.

Category of class action. The Court unanimously agreed that this case — and possibly any class action seeking monetary compensation — cannot be brought (as this one was) using the lenient and minimal procedural safeguards that the Federal Rules of Civil Procedure require of suits seeking only injunctive or declaratory relief. This is a ruling of broad significance, not confined to the employment discrimination context.

Commonality. The majority decided that for plaintiffs to win the right to proceed as a class, they must demonstrate a relatively high degree of “commonality” between their claims. “Commonality requires the plaintiff to demonstrate that the class members have suffered the same injury,” Justice Scalia wrote. “This does not mean merely that they have all suffered a violation of the same provision of law. . . . What matters . . . is not the raising of common ‘questions’ . . . but rather the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation.”

Individual hearings. The Court unanimously agreed that the lower courts’ attempts to streamline the adjudication process in this case to accommodate the litigation of more than a million claims at once deprived litigants — not just Wal-Mart, but also absent class members — of needed statutory safeguards. The lower court had let the case to go forward on the theory that it could be decided based on elaborate computer models, but without hearings ever being held to determine whether individual class members were entitled to relief. The Court unanimously rejected such a “Trial by Formula,” as Justice Scalia dubbed it.

The judge’s fact-finding role. The majority said that in performing its gatekeeping function in ensuring that classes are composed of plaintiffs whose legal situations share a sufficient minimum degree of “commonality,” judges are permitted to make some findings about disputed questions of fact — a role that many judges had thought had to be reserved for later resolution by a jury. In addition, the majority suggested that judges at the class certification stage can probably also make gatekeeping decisions about whether expert witnesses are engaging in junk science.

Victimization ratio. The majority noted in passing that while the Court had, in a previous case, permitted plaintiffs to allege a “pattern and practice” of discrimination by an employer where one in eight class members claimed victimization, in the Dukes case the plaintiffs had only demonstrated that one in 12,500 class members had complained of wrongdoing. The court implied that such a sparse showing was insufficient. Though the majority insisted that it was not setting up any new statistical threshold for making “pattern and practice” claims, the Court’s citation of these ratios clearly invites such inferences in the future by lower court judges, who crave bright-line rules of thumb of this type.

To recap the basics, in 2004 a San Francisco federal judge allowed six female Wal-Mart employees to sue on behalf of every one of the nearly 1.5 million female employees who then worked, or had worked, at any of Wal-Mart’s 3,400-plus stores nationwide since December 26, 1998. The suit alleged gender discrimination with respect to promotions and pay. (Had the case been allowed to proceed, it would actually have been much bigger than what the Court described — Wal-Mart now has more than 4,300 stores nationwide, for instance — but the Court cites statistics that were accurate when the lower court record on class certification was created in 2004.) In April 2010, the Ninth Circuit pared the class very slightly, but generally approved the lower court’s ruling.

The plaintiffs alleged that Wal-Mart allowed individual store managers to make pay and promotion decisions based on excessively subjective criteria. As a consequence, these store managers (who are more often than not men)
unconsciously
tended to choose people like themselves (i.e., other men) to receive career advancements, it was claimed. Over time, Wal-Mart’s failure to curb store-manager discretion in the face of continuing statistical gender disparities in pay and promotion rates was then said to amount to intentional discrimination by the company.

Since individual store managers made most promotion and pay decisions at Wal-Mart, one obvious question was whether the alleged abuse of discretion by certain store managers at a small percentage of stores could be extrapolated to establish abuse of discretion at all Wal-Mart stores. In response, the plaintiffs argued that Wal-Mart’s strong corporate culture led store managers to unconsciously abuse the discretion they were granted in uniform ways.

At oral argument in March, Justice Scalia had protested that he felt whipsawed by that argument, commenting: “On the one hand, you say the problem is that [the decisions] were utterly subjective, and on the other hand you say there is a strong corporate culture that guides all of this. Well, which is it?”

In today’s ruling he rejected almost every aspect of this theory, including the core notion that excessive subjectivity in personnel decisions could amount to a “general policy of discrimination” susceptible to a class-action remedy. Scalia stressed that while Wal-Mart’s expert on organizational sociology, William Bielby, said he believed Wal-Mart’s procedures were “vulnerable” to discrimination, Bielby admitted that he could not say whether 0.5% or 95% of store manager employment decisions were actually motivated by improperly stereotyped thinking. “Whether 0.5 percent or 95 percent of the employment decisions at Wal-Mart might be determined by stereotyped thinking is the essential question on which respondents’ theory of commonality depends,” Scalia wrote. “If Bielby admitted he has no answer to that question, we can safely disregard what he has to say.”

Though some will interpret today’s majority ruling as reflecting the conservative majority’s pro-business tilt, an alternative explanation is more likely. There was always a thermonuclear issue lurking just beneath the surface of this case, though it was not one of the specific technical issues the Court asked the parties to brief.

Conservatives view lawsuits like this one as coming very close to permitting gender (or race) discrimination to be proven on the basis of little more than statistical disparities in the workforce that are extremely widespread in our society and which might simply result from a stew of complex, innocent, cultural causes. Many conservatives fear that if employers have to avoid statistical disparities to avoid getting sued, they will feel pressured to adopt secret quotas, which are illegal.