When the Supreme Court concluded its annual term last Tuesday, I couldn't help recalling what a student of the court wrote in 1976: "Only one thing is certain and that is that there is no assurance that the Supreme Court will ever feel obliged to adhere to any of its precedents except those enlarging its own jurisdiction." The most recent term disturbingly demonstrates that what was hyperbole then is now uncomfortably close to fact.

To be sure, three of the court's most publicized decisions this year were little more than reaffirmations or logical extensions of prior rulings. The Bob Jones case, which upheld the IRS's denial of tax exemptions to private schools practicing racial discrimination, followed a 1971 decision. The statutory invalidation of sexist pension annuities was foreshadowed in 1978, and the constitutional invalidation of local impediments to abortion was essentially foreordained by a series of cases in 1976.

The court went out of its way in the year's round of abortion cases to reject the Reagan administration's invitation to the court to "follow the election returns" and to reverse Roe v. Wade, the 1973 decision that made abortion constitutional. Justice Powell, speaking pointedly for the court, said:

"Legislative responses to the court's decision have required us on several occasions, and again today, to define the limits of a state's authority to regulate the performance of abortion. And arguments continue to be made in these cases as well, that we erred in interpreting the Constitution. Nonetheless, the doctrine of stare decisis (that courts should follow their precedents), while perhaps never entirely persuasive on a constitutional question, is a doctrine that demands respect in a society governed by the rule of law. We respect it today and reaffirm Roe v. Wade."

The court's respect for stare decisis is, however, more episodic than reverential, and that is one disturbing feature of many of the decisions issued in the term just completed.

The most significant constitutional about- face of the term involved tuition tax credits and deductions. In Mueller v. Allen, the court held, 5-4, that Minnesota's tax deductions for school expenses--public and private--did not violate the "establishment" clause of the First Amendment, despite the fact that most of the beneficiaries of the deductions were the parents of children attending sectarian schools. The court found it not necessary to follow a 1973 decision that invalidated a New York program that provided reimbursement of tuition for children in parochial schools. The court in Mueller saw nothing unusual in what it did and reported, apparently with a straight face, that it was simply following well-settled principles.

The court also dealt a death blow to the 1976 decision of National League of Cities v. Usery, which held that basic constitutional principles of federalism immunized states and municipalities from the reach of the Fair Labor Standards Act. The same principle does not apply, however, to the Age Discrimination in Employment Act. The difference is one of degree, the court declared.

Justice Powell apparently found stare decisis entirely unpersuasive with respect to the Eighth Amendment's prohibition of "cruel and unusual punishment." Writing for a 5-4 majority, he found that a life sentence without possibility of parole for a repeat offender violated that amendment, although the court had held three years ago, also by a 5-4 vote, that a life sentence (with the possibility of parole) was not unconstitutional.

The dissenters could not resist tweaking him: "While the doctrine of stare decisis does not absolutely bind the court to its prior opinion, a decent regard for the orderly development of the law and the administration of justice requires that directly controlling cases be either followed or candidly overruled."

But why display candor when you can get away without it? Although the court declined to alter the "exclusionary rule," which prohibits unconstitutionally seized evidence from being used in criminal trials, it nonetheless announced it would "abandon" nearly two decades of case law setting the standards for verifying informants' tips that support applications for search warrants. The court will hear arguments next term, for the third straight year, on whether to modify the exclusionary rule. It appears that not even candor will save the rule from a substantial modification.

Even when the court writes on a clean slate, without pesky precedent to get in the way, its decisions have an unpredictable future. In Chadha v. INS, the court utilized the deceptively simple and literal reading of Congress' powers under the Constitution to conclude that the legislative veto was unconstitutional. (During the same term, the court held that Congress' inaction on the tax exemption issue in the Bob Jones case amounted to an affirmative statement of public policy, but never mind.)

Whether the court will continue to be so matter-of-fact when the stakes are higher, such as with the War Powers Resolution of 1973, is unclear. The resolution is about the only device Congress has in an emergency to keep the president from committing troops abroad for undeclared wars.

Nor is it clear that the literal approach adopted in Chadha will have general applicability in other knotty constitutional areas. Does Congress have the power to make any "regulations and exceptions" to the court's appellate jurisdiction that it likes, as Article III seems plainly to say? Sen. Jesse Helms would be delighted. If the legislative veto is so obviously unconstitutional, what about that most complete denial of separation of powers--the administrative agency that combines law-making, law enforcement and adjudication?

The short answer to these simple questions is that they are not all that simple. Their complexity is a function not only of the ambiguous imperatives of the Constitution and of the problematic role of the court but also of the welter of precedents that the court--as ultimate interpreter of the document--has generated in nearly two centuries of operation.

A court that treats its own landmarks so casually is not unlike the clock that chimes 13: it casts doubt not only on its immediate announcement but also on everything else that it has said or is prepared to say. Or, as another Supreme Court said 150 years ago: "We must respect the solemn decisions of our predecessors and associates as we may wish that those who succeed us should respect ours, or the supreme law of the land, so far as it depends on judicial interpretation, will change with the change of the judges."

Part of the reason that the present Supreme Court's jurisprudence is so unstable, as this term vividly shows, is that in many constitutional areas (the First Amendment, criminal procedure, equal protection of the law) the court has frequently changed both its results and its method of constitutional analysis.

Where the court 20 years ago issued unequivocal tests and absolute rules, the present court measures one constitutional issue after another with a "balancing test": which side has more constitutional interest than the other, as the justices see it? The tendency to see constitutional imperatives in terms of competing "interests" not only makes decisions fact-specific and thus inherently of limited staying power. It also places the judges--either comfortably or uneasily--in the position of making the types of decisions that are usually made by legislators. If, for example, the court sticks by its new interpretation of the Eighth Amendment, the justices may feel now more than ever like a super legislature or a national sentencing review board.

At the end of the term, especially with so much new law to digest, thoughts of the Supreme Court more often turn to easier questions such as workload and what if anything to do about it. The court has encouraged this distraction of late. Congress will soon consider the chief justice's proposal to establish a national court to resolve conflicting interpretations of federal law among the circuit courts of appeal. The new court, according to the chief justice, would reduce the court's oral argument calendar by one-third.

Even if a new court is established, there is no guarantee that the Supreme Court will feel its burden lightened. Several justices would like to see--and will be able to see--that other cases are heard in place of the so-called "conflict" cases. So the innovation could result in no net change in workload.

But more important problems remain. Regardless of how many or which cases the court hears annually, the work product of the justices is of little value if all bets are off at the beginning of each new term. Congress cannot address this problem by statute or even by constitutional amendment. As Robert H. Jackson observed during the court-packing fight in 1937, "You can't amend a state of mind."

He was right, of course. Now, as then, the court needs to change its fundamental attitude with respect to its responsibilities if it is to maintain respect for its decisions and for the justices who render them.