bout 80 percent of decisions issued by the federal appeals courts are tickets good for one ride: they decide only the particular case, and they do not establish binding precedents.

In many parts of the country it is unlawful even to mention these one-time rulings in legal papers submitted in later cases, and judges have been very resistant to change the policies.

"We may have decided this question the opposite way yesterday," Richard S. Arnold, a federal appeals court judge in Arkansas, wrote in describing the current system, "but this does not bind us today, and, what's more, you cannot even tell us what we did yesterday."

But the prohibitions may soon be easing. On Jan. 1, the United States Court of Appeals for the District of Columbia Circuit and the Texas Supreme Court will reverse their restrictions on citing these so-called unpublished decisions. Systemwide change seems to be on the horizon, too.

Last month, an influential committee of the administrative arm of the federal courts proposed a rule that would require all federal appeals courts to allow citation of unpublished decisions.

"It's a major, major development," Arthur D. Hellman, a law professor at the University of Pittsburgh, said of the proposal. "There has been so much resistance to this at the judicial level for more than 20 years now."

Stephen R. Barnett, a law professor at the University of California, said developments in the individual courts more generally reflected a growing "aversion to secret law."

"It has reached a tipping point," Professor Barnett said.

Supporters of the current rules say, however, that any reliance on unpublished decisions, which are generally brief and may not be carefully reasoned, will corrupt the system.

"If flipping a coin that you know is bad is better than nothing," said Judge Alex Kozinski of the United States Court of Appeals for the Ninth Circuit, "it's better than nothing."

The rule against citing a court's own decisions back to the court is in many ways odd.

"It is difficult to justify a system that permits parties to bring to a court's attention virtually every written or spoken word in existence except those contained in a court's own nonprecedential opinions," wrote Patrick J. Schiltz in the draft minutes of the November meeting of the Advisory Committee on Appellate Rules, which is proposing the systemwide change.

"Nonprecedential" is an unlovely word, but it is more accurate than "unpublished," which technology has turned into a misnomer. Not long ago, most unpublished decisions were not only absent from the law books but were also truly generally unavailable. With the advent of legal databases and courts' own Web sites, almost everything issued by appeals courts is widely and almost instantly available.

And since September 2001, "unpublished" decisions have been collected and published in the literal sense, in books called West's Federal Appendix.

The general availability of unpublished decisions has eliminated a crucial objection to allowing them to be cited. Before technology leveled the playing field, it was considered unfair for institutional litigants, like corporations repeatedly sued on similar claims, to be able to collect and selectively cite from a body of law unavailable to their adversaries.

The Justice Department proposed the rule adopted in principle by the court system's committee last month. It was put forward in the Clinton administration, but Solicitor General Theodore B. Olson said the government continued to support it.

"When courts decide things, they don't just disappear and become a hidden trove of law," Mr. Olson said.

But when it came time to vote, Mr. Olson's representative on the committee abstained. According to the draft minutes, the abstention followed a phone call from Judge Kozinski to Mr. Olson.

Mr. Olson said the concerns of several judges on the Ninth Circuit, which covers nine Western states, prompted him to move cautiously.

"The immense size of the circuit, the number of judges there and the huge volume of work they handle make it very difficult for them to monitor all of the decisions of their colleagues," he said, describing the judges' objections.

Judge Kozinski said the huge size of the circuit was one aspect of the problem. But he emphasized that unpublished decisions, often prepared by staff lawyers, were fair to the parties but were "worth nothing at all" as precedent.

"They simply don't get vetted the way that binding opinions do," he said. "Having them available as a source of wisdom is simply a fraud."

The current rules are a patchwork.

Five of the 13 federal appeals courts, covering the Northeast, the West and much of the Midwest, ban citation of unpublished opinions, but the First Circuit, which covers most of New England, is considering a change. The Fifth and 11th Circuits, which cover six Southeastern states, allow citation of unpublished opinions but do not make them generally available, making the question largely moot. The remaining circuits allow but generally disfavor such citations.

The large majority of state courts restrict the citation of unpublished cases.

In the California state court system, 94 percent of all appellate decisions are unpublished, and their citation is forbidden.

In Texas, the appeals court in Dallas did not publish 97 percent of its cases in recent years, said Charles L. Babcock, the chairman of an advisory committee to the Texas Supreme Court. The appeals court in Beaumont, by contrast, Mr. Babcock said, published about half of its decisions.

"The law is not more weighty and interesting in Beaumont," Mr. Babcock said, "yet it was having a much bigger impact than the Dallas court's decisions."

The arguments about whether citation of unpublished decisions should be allowed is one of three related issues involving such opinions. The second is whether courts are bound to follow such decisions as precedent. And the third is whether litigants are even entitled to a statement of the appeals court's reasoning. About 1,350 of the 29,000 decisions issued by the federal appeals courts last year consisted of a word or a phrase (usually, "affirmed") and contained no reasoning.

Though allowing citation of unpublished opinions would not mean that they have the force of binding precedent, opponents say that it would inevitably push judges in one of two unacceptable directions.

They would either work harder on unpublished decisions, at the expense of published decisions or of the efficiency of the judicial system generally. Or they would dispense with reasoning entirely.

Judge Kozinski said he had already altered his approach to unpublished decisions.

"I have now started saying less and less to the parties," he said.

Writing an opinion binding on future courts, on the other hand, Judge Kozinski wrote last year, "is an exacting and extremely time-consuming task" that sometimes requires 70 or 80 drafts over several months.

Some scholars say there are worse things than exerting some pressure on judges to spend a little less time on some of their longer decisions.

"While courts should be thorough in major cases," Professor Hellman said, "they don't have to treat every aspect of the case with 20 footnotes."