The Affordable Care Act most likely met its fate on Friday, when the Justices of the Supreme Court met in conference, though we mere mortals won’t know what happened until nearly the Fourth of July.

So nothing written in the press at this point is likely to have any effect on the result. The inside-baseball chatter goes on, and this week it centers on two comments made from the bench by Justice Kennedy. They raise the prospect that Kennedy might decide the case for the government on the grounds that a victory for the challengers would raise constitutional problems.

Conservative supporters of the challenge seem a bit unnerved by that idea. But the spinning and counter-spinning now underway seems unlikely to have any effect. So let’s skip the internet call-and-response and instead use this dispute as a chance to explain an important but little-understood part of constitutional law: the Doctrine of Avoidance.

Ah ah ah! Don’t touch that mouse! When social conversation turns to “the avoidance doctrine,” lay people tend to stampede toward the exits. But if you’re interested in how courts deal with the Constitution, the idea is worth understanding; it surfaces over and over in important cases.

Let’s begin with one obvious but important fact: King v. Burwell isn’t a constitutional case. Only about a third of the Supreme Court’s cases, in fact, directly concern the Constitution. The rest require the Court to interpret rollicking federal statutes such as the Prison Litigation Reform Act, the Religious Land Use and Institutionalized Persons Act, or the ever-popular Federal Insecticide, Fungicide, and Rodenticide Act. Interpreting statutes, however dull they may be, is the heart of what twenty-first century courts do; they have developed an elaborate, and not entirely coherent, set of tools to do that.

So the question in King is purely statutory: Section 36(b) of the Affordable Care Act provides that subsidies are available to taxpayers who buy health insurance on an “exchange established by a state under § 1311” of the Act. Elsewhere, though, the Act says that a state may establish an exchange, but if it doesn’t, the federal government will establish “such exchange” for it. The challengers say that subsidies are available only in states that establish their own exchange (“established by a state”); the government responds that reading the entire statute makes clear that it “such exchange,” whether set up by the state or federal government, qualifies its customer for subsidies.

Neither side suggests that the Act is unconstitutional. Neither side denies that, if the challengers win, the “federal exchange” states will experience a “death spiral,” in which low- to middle-income households, without their subsidies, drop their policies, all but destroying the ability of insurers to offer coverage at reasonable prices. The challengers say that Congress thought threatening states with the “death spiral” was a nifty way to get them to set up exchanges; the government says that Congress intended every consumer to be eligible for subsidies.

So the Court must decide what these statutory phrases mean. And during oral argument, Justice Anthony Kennedy asked three times whether the challengers’ reading wasn’t constitutionally doubtful. “[I]f your argument is accepted, the states are being told either create your own exchange, or we'll send your insurance market into a death spiral,” he told Michael Carvin, the challengers’ lawyer. “It seems to me that ... there's a serious constitutional problem if we adopt your argument.” He continued, “I think the Court and the counsel for both sides should confront the proposition that your argument raises a serious constitutional question.”

Under the Court’s caselaw, Congress can offer rewards to states that join federal programs, but it may not “coerce” them, either by giving their government a direct order or by offering a choice so draconian as to amount to a gun to the head. And Kennedy clearly worried that the challengers’ reading made the statute coercive. After quizzing Carvin, he raised the same point with Solicitor General Donald Verrilli: “It does seem to me that if Petitioners' argument is correct, this is just not a rational choice for the States to make and that they're being coerced. ... And that you then have to invoke the standard of constitutional avoidance.”

What is he talking about? “Constitutional avoidance,” is a “canon,” meaning a principle of statutory reading. Here’s a summary, from Reading Law: The Interpretation of Legal Texts by Justice Antonin Scalia and legal-usage maven Bryan A. Garner: “A statute should be interpreted in a way that avoids placing its constitutionality in doubt.”

Avoidance kicks in when a statute could reasonably be read in one of two or more ways. What if under one reading the statute is clearly constitutional, but under the other it isn’t, or might not be? Justice Oliver Wendell Holmes, in a 1927 case, explained that “the rule is settled that as between two possible interpretations of a statute, by one of which it would be unconstitutional and by the other valid, our plain duty is to adopt that which will save the Act. Even to avoid a serious doubt the rule is the same.”

The “serious doubt” principle actually dates back to 1830. Its most famous enunciation was in a concurrence by Justice Louis D. Brandeis in a 1936 case called Ashwander v. Tennessee Valley Authority. As recently as last term, Chief Justice John Roberts invoked these doctrines in a case involving an anti-terrorism statute. The defendant smeared caustic chemicals on a mailbox to harm a romantic rival; the government convicted her of violating a federal prohibition on “chemical warfare.” In an opinion for the Court, Roberts wrote that “the most sweeping reading of the statute would fundamentally upset the Constitution’s balance between national and local power.” That gave the Court majority “serious reason to doubt the Government’s expansive reading of section 229, and calls for us to interpret the statute more narrowly.” The defendant walked.

There are two reasons not to bank too heavily on “avoidance” to sway the Court, or Kennedy. The first is that the “canon” is only used when a statute is ambiguous. Both sides insist that the ACA is clear—that their reading is the only one possible. The government’s view implies that the challenge is insubstantial; the challengers know that, if the Court finds the statute ambiguous, it may hesitate to throw the entire U.S. healthcare system under the bus. Most of the Justices were cagey about the ambiguituity issue; Justice Scalia, however, seemed to believe that the statute could only be read one way—the challengers’. If he’s right, then “using” avoidance wouldn’t be interpretation, but what Biblical scholars call eisegesis, inserting false meaning into a text—like someone convinced that Abbey Road meant that Paul McCartney is dead. That’s what Scalia meant when he asked Carvin, “If interpreting [a statute] one way is unconstitutional, you interpret it the other way. ... But do we have any case which says that when there is a clear provision, if it is unconstitutional, we can rewrite it?”

The other problem is that a “canon” is not a statute, not a precedent, not a rule. It is a judge-made tool, one that judges may use or ignore when it suits them. As Scalia and Garner wrote in their book: “The rule is well established. Yet it presents the difficult question: How doubtful is doubtful? ... Unsurprisingly, the cases are many in which the majority and the dissent disagree on application of the standard.”

In the end, the fate of the ACA may come down to that question—“how doubtful is doubtful?”—as it plays out in one Justice’s mind.