Today Manhattan U.S. Attorney Preet Bharara will ask for rehearing of a key insider trading case in which a three-judge federal appellate panel dealt him a stinging setback in December.

According to a spokesman for Bharara’s office, he will ask for rehearing both by the panel that handed down the original decision, and by the full U.S. Court of Appeals for the Second Circuit, in a procedure known as an en banc hearing.

[Update: The petition, now filed, argues that the earlier ruling “breaks with Supreme Court and Second Circuit precedent, conflicts with decisions of other circuits, and threatens the effective enforcement of the securities laws.”

Attorney Gregory Morvillo, representing defendant Anthony Morvillo, one of the beneficiaries of the original ruling, said in a statement Friday that “the unanimous decision … unequivocally confirmed Mr. Chiasson’s innocence under the law and we have great confidence in the opinion because it is well rooted in Supreme Court and Second Circuit precedent.”]

En banc hearings are rare, and especially so in cases where the original panel ruled unanimously, as was the case in this instance, so the request might not be granted.

The panel ruling of December 10, known as United States v. Newman, overturned the convictions of hedge fund portfolio managers Todd Newman and Anthony Chiasson, of Diamondback Capital and Level Global Investors, respectively, and then dismissed all charges against them.

The reasoning of the decision, if left in place, makes it extremely difficult to criminally prosecute what are known as “remote tippees” in insider-trading cases—that is, people who learn about material, nonpublic information through intermediaries, and not directly from the “insider” who originally leaked the information.

In the Newman and Chiasson cases, the original “tippers”—corporate officials at Dell and Nvidia—leaked material nonpublic information to members of a “ring” of stock analysts, who passed the information along to each other, who then passed it up the line to their portfolio managers.

The problem was, the panel found, that a 1983 Supreme Court precedent makes remote tippees, like Newman and Chiasson, liable only if they know that the original tipper received a “personal benefit” from the original tippee—something that wasn’t proven in Newman’s or Chiasson’s cases. So the court threw out the cases.

To many nonlawyers, the ruling is extremely nonintuitive, to say the least. So long as the defendants knew they were trading on material nonpublic information, who cares whether the tipper got a “personal benefit”?

Moreover, the ruling seems to make it easy for a crooked portfolio manager to insulate himself from liability by simply placing several ranks of underlings between himself and the improper leaks.

Just yesterday, for instance, relying on the Newman ruling, another Manhattan judge allowed four individuals to retract their guilty pleas to what once looked like rather routine, shave-and-a-haircut insider trading charges. They had each admitted knowingly trading on material nonpublic information in 2009 relating to IBM’s impending purchase of software company SPPS.

But Judge Andrew Carter Jr. found that the Newman ruling cast doubt on their guilt. Again, they were remote tippees, and though there was evidence that they had acted furtively, tried to conceal what they were up to, lied about what they’d done when confronted, and wrote emails to each other expressing anxiety about “going to jail,” there was no evidence that the original tipper had received a personal benefit.

Though the controversial Newman ruling was met with outrage from some quarters—by those who think insider trading prohibitions should be about preventing market insiders from making easy riches on sure-thing information that most of us have no hope of ever gaining access to—others found it unsurprising, clearly mandated by the Supreme Court’s precedents, and an appropriate place to draw the line.

Though we often think of insider-trading laws as forbidding any trading on material inside information—and many Securities and Exchange Commission staffers and U.S. attorneys think they the laws ought to do that—the U.S. Supreme Court has repeatedly said that they do not sweep so broadly.

Certain kinds of information advantage are actually legal, and that’s why the lines get murky. I wrote a cover feature about this for the September 2, 2013 issue of Fortune, “The Gray Art of Not Quite Insider Trading.”

A key source of difficulty is that there is no actual statute specifically addressed at “insider trading.” Instead, courts have interpreted the general law against “securities fraud” to prohibit some forms of insider trading. Many forms of “unfairness,” however, don’t rise to the level of “fraud,” and the Supreme Court has repeatedly insisted that only “fraud” is forbidden by the statute.

Lurking behind that legalistic barrier is a policy consideration. The Court—at least by the 5-4 margin of the 1983 precedent that Newman relied upon, known as Dirks v. S.E.C.—wants to encourage stock analysts continue to dig hard to learn as much information as they can about the stocks they cover. Such research ensures that stocks are valued accurately and that, in turn, promotes market efficiency. The fact that these analysts gain informational advantages over others in the process of unearthing such information is not, then, necessarily illegal. Sometimes, in fact, it benefits society, the Court believes.

Further, it is also true that, the more remote a tippee gets, the more chance there is he may really have not known or even suspected that the information was illegally obtained.

Still, some observers think the outcome of the Newman ruling, though defensible as a matter of precedent, is wrong as a policy matter.

“The real problem is that Congress should bite the bullet and pass a statute that defines insider trading law,” says Thomas C. Hazen of the University of North Carolina School of Law. “That would end this problem.”

In the event that Newman remains the law, and no statute is passed, it’s probably also worth advising all the remote tippees in our audience that there are still reasons not to rush out and trade on what you know to be material nonpublic information.

According to Donald Langevoort, a professor at Georgetown University Law Center, there are strong arguments that the Newman case should only govern in criminal cases.

“The Dirks court says the tippee has to know, or should know, of the personal benefit,” says Langevoort. “That ‘should know’ was ignored by the court in Newman. Its language does not track the actual language of Dirks.”

While the prosecution in a criminal case might be held to a higher standard, he says, he does not believe the Securities and Exchange Commission should be held to that standard in a civil case.