"In some unknown number of cases the Court’s rule will return a killer, a rapist or other criminal to the streets and to the environment which produced him, to repeat his crime whenever it pleases him. As a consequence, there will not be a gain, but a loss, in human dignity."

Those were Justice Byron White’s words in 1966, in a forceful dissent from the Supreme Court’s 5-4 ruling in Miranda v. Arizona that police must tell arrested suspects of their rights to remain silent and consult counsel, and must cease interrogation once a suspect has requested a lawyer.

Earlier this month, however, White joined a 6-2 decision that both reaffirmed Miranda and extended it. Justice Anthony Kennedy’s majority opinion in Minnick v. Mississippi reversed a murder conviction and suppressed a confession on the ground that, after a suspect has consulted an attorney, the police may not initiate or resume questioning without the attorney present.

So Justice White’s gotten more liberal, right?

Not exactly. Take a look at his evolution on affirmative action.

In 1979, White lined up with the liberals, joining Justice William Brennan Jr.’s opinion in Steelworkers v. Weber, the first clear endorsement of racially preferential affirmative-action plans by private employers.

But by 1987, White had completed a seeming migration to the conservative, anti-affirmative-action side. Dissenting in Johnson v. Transportation Agency, he wrote, "I would overrule Weber." He complained that the Court was stretching Weber to bless reverse discrimination by any employer with a statistical imbalance in its work force, in a ”perversion” of the 1964 Civil Rights Act.

And in 1989, White’s majority opinion in Wards Cove Packing Co. v. Atonio overruled (while pretending not to) a key holding of Griggs v. Duke Power Co., a landmark victory for job discrimination plaintiffs in which White had joined in 1971. Wards Cove shifted from employers to plaintiffs the burden of persuasion as to whether job selection criteria that have a disparate impact on racial minorities are justified by business necessity.

White is not the only justice who seems to have reversed himself on big, ideologically charged issues in recent years.

Retired Justice Lewis Powell Jr. told a group of New York University law students in October that he. "probably made a mistake" in casting the crucial fifth vote for the . majority in Bowers v. Hardwick in 1986. That 5-4’deciston upheld a Georgia statute, as applied to homosexuals, that criminalized all oral and anal sex.

Powell’s new view is that the Constitution’s protection of privacy includes a right to engage in homosexual acts. This was also his view when the Court took its first, confidential vote after oral argument in Hardwick. But he changed his mind while the opinions were being prepared-and with it, the outcome of the case.

Now Powell has changed his mind again-too late for those who would like to see the decision overruled-:

Most other Supreme Court justices have also flip-flopped on issues of great consequence, often angering old friends, including the presidents who appointed them.

In 1986, Chief Justice William Rehnquist joined a dissent suggesting that the landmark 1964 ruling curbing libel suits in New York Times v. Sullivan "should be re-examined." In 1988, he embraced and extended that decision when he wrote a unanimous opinion throwing out the Rev. Jerry Falwell’s $200,000 jury award against Hustler magazine for intentional infliction of emotional distress.

Justice Harry Blackmun migrated from the conservative to the liberal side on so many issues that it contributed to his estrangement from his old friend Warren Burger, chief justice from 1969 to 1986. One example: Blackmun had provided the fifth vote for reviving constitutional protection for state sovereignty against federal regulation in National League of Cities v. Usery in 1976. He then provided the fifth vote for overruling Usery in 1985.

A more subtle evolution is suggested by Justice Sandra Day O’Connor’s opinions on abortion. In her first abortion case, in 1983, she seemed fairly itching to overrule Roe v. Wade, which, in a dissenting opinion, she called "completely unworkable" and "on a collision course with itself.”

But over the next seven years, whenever a colleague has urged that the time has come to reconsider the 1973 Roe decision, the Court’s first woman has responded, in essence, "Not so fast, buster.” While she has never embraced Roe, the smart money is betting she will never vote to overrule it.

Such switches and quasi-switches are reminders of the human changeability of the justices, the subjectivity of the interpretation process, and the crudeness of the "liberal" and "conservative" labels to which we resort to pigeonhole justices. The "liberal" Warren Court mustered only a 5-4 majority for the Miranda ruling; 24 years later, a far more "conservative" Court has voted 6-2 (with new Justice David Souter not participating) to reaffirm and extend it.

While some conservatives purport to have found the key to constitutional legitimacy and consistency in the original intent of the framers, Court nominee Robert Bork’s flip-flops at his 1987 confirmation hearings suggest – otherwise.

There is nothing wrong with a judge or anyone else changing his mind. It’s often a sign of intellectual growth. It should also foster tolerance for the perceived heresies of ideological adversaries, whose views we may someday find more reasonable.

It would be better, however, if more judges would frankly acknowledge and explain their reversals of position, as Powell did in the Hardwick case, instead of slipping them by without comment or straining to preserve a false appearance of consistency, as White did in Wards Cove.

People weighing the costs and benefits of decisions like Miranda, for example, could learn something from an explanation of the reasons for White’s evolution, which began many years before his Dec. 3 vote in Minnick v. Mississippi.

Here is a man of extraordinary intelligence and broad experience, immersed in cases showing the real-life impact of the Miranda rules and freed by life tenure from the pressures that politicians feel to dissemble on hard issues. What changed his mind?

Does he no longer believe the well-reasoned and quite plausible arguments he made in his Miranda dissent ? Why not? Has Miranda proved less of an impediment to crime fighting than White feared? Have his misgivings been worn down by shows like "Hill Street Blues" that have inculcated Miranda warnings into our culture? Or is he still convinced Miranda was a bad mistake, and is his adherence to it only a signal of his respect for precedent? If so, why would he go beyond reaffirming Miranda and vote to extend it? We can only guess at the answers.

Eugene Fidell of D.C. ‘s Feldesman, Tucker, Leifer, Fidell & Bank, a connoisseur of judicial flip-flops, wrote in Judicature magazine in 1988: "Recantation, of course, has an unsettling effect on the law, but it is inevitable, so it is just as well if at least some amusement comes in the process,"

Fidell has collected examples going back to (and beyond) the graceful acknowledgment in 1872 by English jurist George William Wilshere Bramwell, a baron of the exchequer, that he was departing from his decision in a similar case: "The matter does not appear to me now as it appears to have appeared to me then.”

Justice Robert Jackson, in a 1950 concurrence disavowing a position taken 10 years earlier as attorney general, invoked the panache with which one British lord reputedly brushed aside a barrister’s reliance on one of the lord’s previous rulings: "I can only say that I am amazed that a man of my intelligence should have been guilty of such an opinion.”

A modern classic of the genre was contributed by Justice Richard Sims III of the California Court of Appeal in 1986. "I write separately to fall on my sword," Sims began, noting that he had "inexplicably failed to discover the controlling statute" in two previous cases similar to the one at hand.

"The good coming from all this," Sims reflected, "is the knowledge that, having taken all conceivable sides on the issue, I must certainly at some point have been right. Unfortunately, it too obviously follows that at some point I must also have been wrong."