To review the bidding for those who came in late:

In answer to a question prompted by the DC-area sniper problem, Ari Fleischer rejected out of hand the proposal to insitute a national ballistic signature system, on technical and privacy grounds, and using the “Why would a killer obey a gun law?” argument.

When it comes to criminal behavior and people who use guns to commit murder, there is no amount of laws that will stop these people from committing these depraved crimes. The issue is their morality. The issue is their values. They have broken the law, they will break the law, new laws do not stop people like this. (See here.)

Congressman Jim Moran asked why the White House wanted to protect the privacy of people “who would be shooting other people.” See here and here. Glenn Reynolds called that a “blood libel.” Various pro-gun bloggers backed up Fleischer in various ways, claiming (1) such a system would cost too much; (2) it would produce too many false positives, leading to unwelcome police visits to buyers of guns with look-alike signatures; (3) a gun’s signature naturally changes with time, and can also be deliberately disguised; (4) getting signatures of new guns wouldn’t do much good due to the large stock of existing weapons, and there’s no feasible way to get signatures from weapons already in private hands; (5) the resulting database would amount to natinal firearms registration, with confiscation as a possible next step.

Flesicher later backed off somewhat, saying the technical issues ought to be studied.

OK. Deep breath. There are two issues here, one operational and one rhetorical. Operations first.

Here’s the proposal. Each firearm would be test-fired (as almost all now are), the peculiarities of the marks it leaves on the bullet and shell casing recorded in some electronic form, and the data stored in a national database along with the serial number of the gun. The serial number, in turn, allows a gun to be traced to the federally licensed firearms dealer [“FFL”] who sells it.

That “gun tracing” process is already (imperfectly) in place for guns found at crime scenes. Each FFL is required to make sure that each purchaser is not disqualified from buying (not a convicted felon, for example), and to retain a record of that check, including the name of the buyer. So, given the identify of the gun, it is possible to trace it to its first retail purchaser. A ballistic signature database would make it possible to start the tracing process with a bullet or shell casing, rather than the gun itself.

Because sales in the secondary market (when one non-dealer owner sells to another) do not have to be background-checked, the trail frequently runs cold at that point. The gun is traced to the retailer and then to the purchaser, but he sold it through an ad in the newspaper to someone whose name he doesn’t recall. Still, tracing statistics can be useful in identifying patterns of illegal gun trafficking.

[Note: Many of us think that it would be useful to eliminate the private-sale loophole and require all weapons transfers to be background-checked, thus making it possible to trace each gun to its last lawful purchaser (and incidentally making it marginally harder for ineligible buyers to acquire guns). Since many secondary-market sales go on at or near gun shows, there is a superstition among the gun controllers, which they have managed to communicate to the press, that there exists some sort of “gun show loophole.” False. Nothing is legal at a gun show that wouldn’t be legal anywhere; the loophole is for private sales generally. But the current trace system still has investigative benefits, despite the problem of the trail running cold.]

There is some dispute about how good current ballistic signature technology is; Eugene Volokh links to a post from Clayton Cramer citing a skeptical view from a California law enforcement working group, though that report is about hanguns rather than rifles. But Moore’s Law is still working hard, so anything that a computer can almost-sorta-kinda do today will probably work pretty well ten years from now. That suggests that whatever system we put in place now ought to record as much data as possible — perhaps a photograph — of the bullet and casing markings, in addition to whatever classification is now done, in order to allow upgrading the in the future.

A system with a high false-positive rate will be too expensive to use routinely; on the other hand, if the Maryland State Police had a list of 200 rifles, with one chance in three that the sniper’s gun is on that list, they’d be running down those leads.

Yes, markings change with use, and can be deliberately altered. But here again advances in image-processing techniques are likely to make such techniques decreasingly useful, as people who thought they had erased audio tapes or deleted files from computer storage media have discovered.

But the “What do we do about old guns?” argument actually destroys the “Technology isn’t ready yet” argument. Whatever its technical capacity, the system won’t be much use the day it starts, because it won’t have many signatures in it. Over time, the proportion of guns whose signatures are recorded will rise, and the usefulness of the system will rise with it. That means that the time to start taking signatures is before the comparison system is fully mature — now, for example — so that when the computers get smart enough to do high-accuracy automated comparisons they’ll have some data to do them on.

The “gun registration” argument is pure red herring, though Eugene takes it seriously as does the Right-Thinking Blogger. Given the serial number of a gun, it’s already possible to trace it to its first retail purchaser, though the records are decentralized in the FFL’s rather than kept in a national database. Taking ballistic signatures won’t change that in any way; such a system only means that, given a crime-scene bullet or shell casing, it might be possible to identify it with a gun — even if the gun isn’t available — and thus with a purchaser. Gun confiscation will be exactly as feasible technically the day after the system is in place as it was before. (Exactly as feasible politically, too, which is to say, not feasible at all.)

The “We have enough laws already” business is herring of an even deeper red. No one is asking killers to obey anything. We’re trying to make it easier to catch them: trying that is, to better enforce the laws already on the books, which is precisely the mantra of the NRA and its tame politicians.

None of this answers the question about whether compiling a database of ballistic signatures is currently worth doing; some experts (David Kennedy, among others) think so, as does the Bureau of Alcohol, Tobacco, and Firearms, but they could be wrong. Still, the alacrity with which the White House press secretary (acting on instructions, of course) dismissed the idea on obviously fallacious grounds suggests that the politics of guns currently trumps even domestic-security concerns at 1600 Pennsylvania. If terrorist threats are now a semi-permanent feature of the American landscape, making investments now that pay anti-terror dividends in five years is only prudent. (Fleischer’s quick retreat suggests that, at least this close to election day, even the NRA isn’t all-powerful.)

Now back to Mr. Moran. If he really meant, or meant his hearers to understand, that Ari Fleischer and his bosses don’t want to do ballistic tracing because they want serial killers not be be caught, that would have been an obviously false and defamatory thing to say. But it seems implausible that he meant that, or that he meant to be misunderstood in that way. He was indulging a fairly standard rhetorical device: stating an unintended consequence of one’s opponent’s position and then demanding why anyone would desire it, eliding the fact that the position so attacked has advantages as well as disadvantages. [Moran seems to be saying that only criminals, rather than target shooters or hunters, will have their guns “matched” by the database, which would be true only if the technology were perfect; in the real world, a certain number of innocent people are going to find the police on their doorsteps asking questions, not necessarily politely.]

That’s not a rhetorical trick I approve of, because it tends to conceal, rather than revealing, what’s at stake in a dispute. But to call it a “blood libel” seems to me … well, libellous.

The original, or copyright, Blood Libel was the accusation that Jews made ritual use of the blood of Christian babies. (In the most common version, the blood was used in baking matzoh for Passover.) The first time I heard it used in political discourse to describe a false charge of non-ritual killing was by Ariel Sharon and his friends, who called it a “blood libel” that people held Sharon responsible for a mass killing undertaken by his Christian Lebanese allies against their Palestinian opponents, with Sharon’s troops nearby and not intervening. I thought it then, and think it now, a skilful attempt to portray Sharon as a victim of anti-Semitism rather than as someone who had provided a good reason to be ashamed of being a Jew. But the logic from Sharon’s side was clear: Jews had been falsely accused of murder in the past, and this accusation against a Jew was simply a reversion to that unsavory practice. (Who says the left has a copyright on victim-mongering?)

But all of that was a long way from Congressman Moran. When the President said, in reference to the Senate Democrats’ unwillingess to make the proposed Homeland Security Agency a patronage dump for Karl Rove (see how easy it is to make a difficult problem seem easy if you’re willing to be nasty enough?), that the Democrats preferred the interests of public-employee unions to the security of the nation, was that a blood libel? When Attorney General Ashcroft claimed that opposition to military tribunals was aiding and abetting terrorism, was that a blood libel?

The seeming willingness of this administration to sacrifice privacy in every other domain — not just in the domestic-security realm, but also with respect to things such as medical records — makes it hard to swallow the claim that its willingness to forgo possible investigative advantage in the case of guns is due to its solicitude for privacy rights. (This isn’t the first instance of preferring gun owners’ privacy to public safety; we’ve already had the Attorney General’s announcement that the FBI would be required to instantly destroy records from the National Instant Check System, thus making it harder to prosecute some violations of the felon-in-possession law.) The decision to oppose ballistic signatures without looking at their merits was a decision to please the gun-rights crowd rather than give (future) law enforcement an extra edge.

It’s not libel to say so. Truth is a defense.

[UPDATE: There turn out to be several technical errors in the above. The system under development works on shell casings, not bullets. It’s not just images that would be stored, but the actual casings themselves. And the point about false positives apparently isn’t right: comparing two casings can provide a definitive match. Thanks to Susan Ginsburg for the corrections.]

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact: