Republic, Lost_ How Money Corrupts Congress--And a Plan to Stop It - Lawrence Lessig [14]
This same principle is one that should lead judges not to accept courtesies like railroad passes from persons or companies frequently litigants in their courts. It is not that such courtesies would really influence them to decide a case in favor of such litigants when justice required a different result; but the possible evil is that if the defeated litigant learns of the extension of such courtesy to the judge or the court by his opponent he cannot be convinced that his cause was heard by an indifferent tribunal, and it weakens the authority and the general standing of the court.24
The legal system thus avoids that chance. Or at least it takes the smallest chances it can. In this sense, following Professor Dennis Thompson, we can say that the “appearance standard identifies a distinct wrong, independent of and no less serious than the wrong of which it is an appearance”—because of this effect.25
But there’s another side to this “impartiality might reasonably be questioned” standard that people often miss: the word reasonably. The question isn’t whether any crazy person might wonder if a judge were biased. (“Your Honor, I notice you have the same birthday as the plaintiff, and I am concerned that might mean you are biased against Capricorns.”) The question is what a “reasonable” person might think.26 And so a reasonable question might be: Why stop at “reasonable”? If the objective is to protect the system, why not require recusal whenever someone in good faith at least worries that the judge is biased?
I learned about this side of the recusal rules the hard way. On December 11, 1997, the judge in the Microsoft antitrust trial appointed me a “special master” in that case. That meant I was to be a quasi, temporary, mini-judge, charged with understanding, and then making understandable, a complex technical question about how Windows was “bundled” with Internet Explorer. Microsoft didn’t want a special master in the case, or at least they didn’t want me. So almost immediately after the appointment, they launched a fairly aggressive campaign, in the courts and in the press, to get me removed. Their opening bid was that I used a Mac (on the theory that a neutral master would use Windows). It went downhill from there.
My first reaction to this firestorm (coward that I am) was to flee. To resign. I didn’t need the anger. I certainly didn’t need the hate mail (and there was tons of that). But when I spoke to a couple of friends who were federal judges, they insisted that it would be wrong for me to resign. If a party could dump a judge merely by complaining, then parties could simply dial through all the judges until they found the one they liked best. The test, as I was told, was not whether a party could question my impartiality. The question was whether my “impartiality might reasonably be questioned.” In their view, given the facts, it could not.
This story will help us understand the dynamic I described earlier in this chapter. In both cases, there was a factual question at stake: Is BPA, or are cell phones, safe? In both of those cases, there was a process by which that question was answered: scientific studies that presumably applied scientific standards to reach their results. But in both cases, there was also an influence present when conducting those studies that made at least some of us wonder. Why—except bias, one way or the other—would 72 percent of industry-funded studies find no danger from cell phones when 67 percent of independent studies found danger? Why would 100 percent of industry-funded studies find no harm from BPA while 86 percent of independently funded studies found some harm? And is it reasonable that someone would wonder about this scientific integrity