When should people report crimes to the police, especially when the crimes are done by people who are (1) friends, (2) associates, or (3) others who inform the person in his professional capacity, and expecting confidentiality? That's a really tough question, but here is an example, from a Slate column last week, of how not to deal with it (emphasis added):

In maintaining his silence, Woodward, more than anyone else in the Plamegate scandal, has upheld the highest standards of journalistic integrity and discretion. Perhaps this is because, more than anyone else, Woodward understands the tenuous and often strained relationship between a powerful government and its citizens. (Whether Woodward—who has made his career as a journalistic watchdog, attentively patrolling executive power run amok—had some special obligation to his editor is a different question.) Woodward's critics are essentially arguing that he should have volunteered information (whether directly to the prosecutor or functionally to the prosecutor via publication) before being asked —- that is, he should have become an informant.

We have laws in this country that designate precisely when citizens are required to rat on other people. The laws, for instance, require doctors who witness injuries consistent with child sex abuse to call authorities; and social workers are obligated to snitch if they confront someone clearly about to physically harm another. Certain other professionals are also deemed by law to be "mandatory reporters." But outside these narrow confines, there is no law in our country imposing an obligation to begin or to assist in a criminal prosecution -— not in drug cases, not in mob cases, not even in murder cases.

And rightly so. America has been through McCarthyism before, and we have seen what a culture of informants can produce.

In America, it is the prosecutor's job to get information, not the citizen's to volunteer it -- and this is for good reason. Many of our other important values -- such as journalistic integrity, the right to privacy, or the right to be free from unwarranted searches and seizures -- compete directly with an obligation to volunteer information. The value of our freedom from governmental authority is invariably tested during troubled times and generally faces its greatest challenges in the context of highly charged issues. But although it is chic to be patriotic, particularly in wartime, a vogue for cooperation with prosecutors shouldn't be confused with good policy. It is true that we prize honesty and integrity in America, and certainly we expect those summoned before a tribunal to testify completely and truthfully, but this is only required when someone is questioned by federal agents or compelled to testify. We may be a nation of Honest Abes, but we are not a nation of snitches. . . .

To begin with, this is argument by epithet; instead of showing that something is bad, it just calls it names. "To rat"; "snitch" (said twice); even "informant" (said twice) -- these are pejorative terms for what might be equally well called "helping catch criminals," "protecting innocent victims," or (in some instances) "whistle-blowing." Would we really call someone who calls the police when "someone [is] clearly about to physically harm another" a "snitch"? Someone who tries to protect a child from future sex abuse a "rat"? (The column gives doctors' calling authorities about possible sex abuse as a "for instance" in the category of "rat[ting] on other people.")

I should think not. And even in more morally ambiguous situations, when the crime is less serious or less imminent, or the personal connects with the criminal are closer, denouncing the person who calls the police as a "snitch" or a "rat" (as opposed to a "public-spirited citizen," who may help advance justice at considerable personal cost, for instance when someone turns in a criminal who's a family member, as in the Unabomber case) should be the conclusion of an argument, not a premise.

But beyond this, the argument fudges a person's legal duties and his moral duties. It's true that the rule in American law is that people don't have to report crimes (at least until they're subpoenaed to do so); legal duties to report are the exception. But the rule in American law is also that people may report crimes; legal prohibitions on reporting are the exception. So appealing to the "laws in this country that designate precisely when citizens are required to rat on other people" as evidence of whether Woodward should have volunteered the information is misleading: To the extent that the laws influence our ethical obligations, they simply tell us when we must come forward (rarely) and when we must not come forward (rarely) -- they don't answer when we should come forward in those situations where we may but need not.

Of course, there are plausible reasons why journalists shouldn't voluntarily reveal their sources, even when the revelation can help investigate a crime. (There are also plausible reasons why people shouldn't voluntarily call the police about conduct that we might think shouldn't be a crime, or is at most a minor crime.) But those reasons may have to do with this particular case, and may have to do with the special role of journalists (a matter that the Slate column discusses, but in too little detail, in my view) and the special values that can be served by journalists' confidentiality.

They have little to do with the proposition -- in my view an outrageous proposition -- that the rest of us, including "doctors who witness injuries consistent with child sex abuse" and "social workers . . . [who] confront someone clearly about to physically harm another" are "rat[s]" and "snitch[es]." Nor do they have much to do with the broader proposition that people's willingness to help fight crime by calling the police when they have evidence of crime would make us a "nation of snitches."