National Journal, June 10, 2000

Before I get on with writing about the Microsoft antitrust case, I suppose I should mention that I hate Microsoft's products. Windows is the first product I have owned since my 1969 Chevy Malibu that regards breaking down ("crashing") at critical moments as its sacred obligation. And Word -- please don't get me started on Word. I'll just say that the individual who thought up the little paper-clip guy deserves to be dipped in honey and left on an anthill in the Mojave Desert. And next time the idiot-child Office Assistant demands to help me write a letter I will wring its scrawny neck.

Well, I told you not to get me started. The point is that I have no tears to shed for Microsoft, and that U.S. District Judge Thomas Penfield Jackson's decision on Wednesday to split the company in two might prove to be a good thing, and that even if it were not a good thing it couldn't possibly happen to a nicer bunch of people.

However. It is impossible for a nonlawyer -- which regular readers of this column will have discerned I am -- to look upon the Microsoft case with anything but misgivings, bordering on worry. For a lawyer the case poses no deep problems. We have a process, and we're sticking to it. A normal human being, however, might wonder whether the legal process has the faintest notion of what it is doing. And it turns out that the normal person would be right to wonder.

To see why, I refer you to a brilliant and sometimes wickedly funny book called Jurismania: The Madness of American Law, by Paul F. Campos, a law professor at the University of Colorado (Boulder). Campos begins with a simple but important observation. The further along a civil case gets in the legal system, the closer a call it probably is. The obvious cases, where the facts and the law are plainly on one side or the other, are weeded out early, either in the first stages of litigation or without any litigation at all. As soon as it's obvious which way the case must come out, the parties settle, quit, or what have you.

Many cases that reach the trial phase are thus fairly close, in the sense that a reasonable judge could go either way. There are plausible arguments on both sides. And many cases that are decided at the appellate level -- which is where the Microsoft case will wind up -- are very close. They present difficult legal or analytical questions, or they pit core principles of law and morality against each other. Such hard cases must be decided arationally, which is not the same as irrationally. That is, reason and evidence can take you a considerable distance but cannot clinch the argument. "Faced with such legal and social contradictions," writes Campos, "we cannot decide efficiently processed legal disputes on the basis of 'reason.' We merely decide."

The Azande tribe of southern Sudan resolved intractable questions by consulting an oracle. They would feed poison to a chicken, and the oracle spoke by allowing the chicken either to live or to die. Of course, the oracle needed to be addressed in just the right manner, by just the right people; otherwise, you were merely feeding a bird. Because everyone accepted the oracle's wisdom and authority, the system resolved conflicts.

If Campos is right, modern law serves, among its other functions, a similarly oracular purpose; and this is both legitimate and necessary, within reason. If intractable questions must be decided, dressing up some fairly intelligent person in a black gown and requiring her to sit on a pedestal and conduct an elaborate ritual, which ends when she intones that The Law requires this or that, is not a bad method. The important caveat is that the elaborate ritual must not become neurotically elaborate -- as, Campos argues, has happened in America today, but that's another story.

Assuming that Campos is at least partly right, as I think he is, Jackson conducted the Microsoft antitrust trial in a fashion that was either ideal or very close to it. In early 1999, as the government wrapped up its case, I spent a couple of days covering the trial for Slate magazine. In even that brief time I thought I had no trouble discerning three things. First, Jackson was scrupulously fair on matters of procedure. Second, he understood that more procedure does not equal more fairness. He realized that 24 witnesses, as opposed to 240, would be enough to tell him what he needed to know. Third, he had already mostly made up his mind: He thought Microsoft's case was a turkey. If more judges were equally brusque and businesslike, American law wouldn't be the nutcase that it is.

A good judge is paid to make his conclusions seem obvious, dictated by no less an authority than The Law itself. Here again, Jackson excelled (who needs equivocal oracles?). His findings were ringing, without squish or ambivalence. Microsoft, he held, is a monopolist whose brutal and illegal tactics hurt consumers by quashing innovation.

But in fact the Microsoft case is not obvious. It is anything but obvious. In an antimatter universe, an anti-Jackson could perfectly well have written equally ringing holdings in Microsoft's favor. For example:

* Were consumers hurt? A few years ago, I paid $40 for Netscape's Navigator browser. Microsoft now includes a Netscape clone in Windows, for free. If I still prefer Netscape's browser (as I do), that's also now free. Microsoft gave itself a convenience edge, but it did nothing to restrict the ready availability of Navigator for anyone who wanted it. So Microsoft's nastiness gave me more choices at infinitely less cost. On the other hand, Microsoft's tactics were not charitably intended. They were meant to obliterate Netscape. So was this a predatory tactic that hurt consumers by trying to kill off competition? Or was it a boon, since objectively it made consumers better off? No law could decide that, even in principle.

* Was innovation squelched? "Consumers will be hurt in the future," said the government and the judge. "That's the whole point." Maybe, maybe not. The government's claim -- that the 800-pound Microsoft gorilla discourages new entrants -- is plausible. If Microsoft gets away with its conduct, we might have fewer Netscapes in the future. On the other hand, Microsoft is itself an innovator. If the government gets away with its conduct, we might have fewer Microsofts in the future. Which is worse, potentially fewer future Netscapes, or potentially fewer future Microsofts? And what are the odds in each direction? Anyone who claims to know is lying.

* Does Microsoft control the market, or does the market control Microsoft? For the government and the judge, Microsoft's ruthlessly aggressive competitiveness, combined with its overwhelming market share, showed that Microsoft is an implacable monopoly determined to strangle all potential rivals. For Microsoft, that same aggressive competitiveness showed that Microsoft was running scared in a fast-changing market and wasn't behaving like a monopoly at all. The evidence supports both views.

Even setting aside all of that mush, the remedy issues were mushier still. Break up Microsoft? How many pieces, two or three? (Prominent economists sharply disagreed, but they did agree, helpfully, that choosing the wrong answer -- whichever that is -- might make things worse instead of better.) Even in principle, how could we know that such radicalism is necessary? Might restraints on Microsoft's conduct, combined with requirements that broaden competitors' access to the Windows code, do the job with less risk?

Here only one thing is clear: This decision cannot be made rationally on the basis of law and currently available facts. It is all guesswork: guesswork about how the marketplace might look in the future, how hypothetical companies might behave, how technology might change.

"My own view is, the antitrust laws don't need changing," said Joel Klein, the Clinton Administration's antitrust chief, in an interview with National Journal recently. (See National Journal, May 27, 2000.) I respectfully disagree. I think that if the government cannot show that present-day consumers have in actual fact been significantly hurt, the government should not be able to go to court and seek to break up a company, even a bullying company. If you're not darn sure something is broken, don't fix it. I think it's important for businesses to have a fairly clear idea, in advance, what is legal and what is illegal. The Microsoft case, I fear, subtracts from clarity instead of adding to it.

A lot of people, beginning with the government, will disagree with me. So they should. They will say that antitrust law will always be messy and that messiness is a small price to pay. They may be right. But let them, and the rest of us, be clear, at least, about what the law is doing in the Microsoft case: whistling in the dark. The process is certainly fair, and it is possibly necessary, but it is not rational. Next time, consult a chicken.