I received a minor bit of castigation from Bob Fertik regarding the piece I wrote on the Pew poll and the public's acceptance of Bush's warrant-free wiretapping. (Scroll down.) I wrote: "Much of the public also believes that the wiretapping scandal is no scandal at all."

Not so, says Mr. Fertik. And he has numbers which, he claims, prove his point.

Zogby, it seems, has conducted another poll on this question: "If President Bush wiretapped American citizens without the approval of a judge, do you agree or disagree that Congress should consider holding him accountable through impeachment?" A 52% majority answered "yes"; 43% said "no."

This is the poll Fertik likes -- and it would seem to conflict with the Pew results, which recorded a 51% approval of Bush's actions. What accounts for the disparity in results? Is there a disparity?

Granted, some of the conflict may be due to methodology. But the real issue is one of wording.

The Pew question comes down to this: "Do you agree with what Bush did?" Quite a few of the people who approve of Bush's actions are under the impression that he functioned within the law. As I wrote in my earlier piece, folks in this category are probably beyond the reach of reasonable debate -- if they haven't yet been informed of what really occurred, they will probably never get the news.

The Zogby question is very different, because it is based on an "if-then" formulation: "IF Bush broke the wiretapping law, THEN should he be impeached?" I'm not shocked to learn that a majority answered yes. I'm more shocked to learn that a whopping 43% of the populace believes that Bush should be above the law even if he has behaved criminally.

I raised the same objections back when my fellow progressives were crowing about another poll, in which a majority expressed a desire to see Bush impeached IF he lied us into war. When a later poll asked the impeachment question without the "if" qualifier, the number of people supporting impeachment dropped sharply.

Let's look at the matter objectively. Let's remove politics from the equation. Presume that a pollster asks: "If Mr. X broke the law, do you think he should be punished?" Now presume that a pollster asks a simpler question: "Do you think Mr. X should be punished?" Common sense tells you that the first question will always find a higher level of agreement, no matter who X is or what he did.

Bob Fertik deserves enormous respect and admiration, so I hope he won't take these words incorrectly. But I would advise readers not to give much weight to "if-then" poll questions. And -- sadly -- I see no reason to revise my original statement.