« Choosing linguistic autarchy | Main | My Thanksgiving present from Yana »
Non (Anti?) Sequitur of the Day
Speaking on why it would be a terrible mistake to overturn Washington DC's 31-year old ban on handguns, Assistant police chief Alfred Durham said today:
The ban on handguns is a matter of life and death because 80% of the murders in DC are caused by handguns.
Posted by Alex Tabarrok on November 21, 2007 at 12:27 PM in Economics | Permalink
Comments
He's not really stupid, some b**** just set him up.
Posted by: Keith at Nov 21, 2007 12:51:28 PM
I see no non-sequitur here.
Posted by: chris at Nov 21, 2007 12:52:57 PM
Chris, it appears that the "ban" on handguns does not actually limit the number of handguns in the hands of murderers in DC.
Posted by: Alex J. at Nov 21, 2007 12:54:32 PM
I gather you think Durham's statement is a non sequitor because, despite the ban on handguns, there are still so many handgun murders. But I don't think your logic holds up. Would it be a non sequitor to maintain a ban on drunk driving even if drunk driving was a major cause of traffic fatalities in DC?
There may be constitutional arguments for overturning the ban, but the appalling number of handgun murders in DC is not an argument for putting more such guns in circulation there (although, admittedly, this is not really a problem that can be effectively addressed at the local level, since guns are so freely available in places like Virginia.)
Posted by: A student of economics at Nov 21, 2007 12:56:11 PM
In a place where most people want guns to be illegal I can't imagine that allowing firearms would do anything to deter violence. Non-criminal people in DC by a wide margin do not want guns around. Do you suppose that they will change their minds and purchase guns for self defense if they are made legal? Or maybe not allowing/making it more difficult to arrest people for having a firearm will mean that more dangerous people with hand guns are on the streets. I know gun ban means that only criminals will have guns but in a place where that is pretty much going to be true no matter what I think fewer criminals having fewer guns is a better equilibrium. If a community wants guns I have no interest in taking them away, but a community that doesn't want as many guns around should be able to decide that without people who don't have to worry about their safety trying to prove some dubious and ideologically driven point.
Posted by: Michael Foody at Nov 21, 2007 1:03:05 PM
Alex, okay, let's make handguns legal in D.C. Then let's make all violent crime legal too, since clearly laws against that don't limit anything either. Hey - and what about nuclear proliferation? - just look at North Korea - clearly that treaty limits nothing - so let's overturn that while we're at it.
Posted by: chris at Nov 21, 2007 1:03:52 PM
You could make a great graph of Alex's posts with silliness on the Y axis and the probability that they would be uttered by a contestant on the $10,000 Pyramid for the category "Things a libertarian might say" on the X axis. Such a graph would form a straight line with a positive slope.
Posted by: Michael Foody at Nov 21, 2007 1:09:25 PM
Clearly, it's a non-sequitur. What some of you are arguing is that it's not an "anti-sequitur" but that's why anti is followed by a question mark.
Posted by: Alex Tabarrok at Nov 21, 2007 1:10:23 PM
Student,
Legal drunken driving would not protect you from illegal drunken drivers in the same way that legally arming yourself could possibly protect you from those who do so illegally, or rather those who do so for illegal means. I'm not sure if your comparison works.
Posted by: G.ira at Nov 21, 2007 1:12:02 PM
Student,
Legal drunken driving would not protect you from illegal drunken drivers in the same way that legally arming yourself could possibly protect you from those who do so illegally, or rather those who do so for illegal means. I'm not sure if your comparison works.
Posted by: G.ira at Nov 21, 2007 1:13:38 PM
The drunk driving comparison is asinine to say the least. I can throw that strawman right back at you and say that driving a car completely sober causes a high number of fatalities so why don't we just ban cars! See I can play that game to.
The question that should be asked is what is causing the high number fatalities (e.g. our idiotic drug laws) and not the tool that was used to carry them out.
Posted by: tim at Nov 21, 2007 1:13:51 PM
Many commentators are being inconsistent. If 80% of the murders in DC were caused by knives, a much higher percentage than in the rest of the United States, wouldn't you take that as evidence that the handgun ban was working? I would (not 100% evidence that it was a good idea but given imperfect substitutability I would take this as evidence in favor of the law.) If so, then to be consistent the fact that the murder rate and handgun murder rate in DC are both high should incline you to believe the law has low benefits (and of course may have some costs).
Posted by: Alex Tabarrok at Nov 21, 2007 1:18:38 PM
Alex T, Google lists 11 examples of "anti-sequitur" - maybe a few more if you don't omit duplicate results. It's as good as meaningless. Having your defense of your initial post depend on it is pretty odd.
The sentence you quoted is not a non-sequitur. I think we've made our point very clearly.
Posted by: chris at Nov 21, 2007 1:22:28 PM
I think the problem is that Alex is reading the statement as a support of the ban's effectiveness and Durham was remarking on the ban's importance. If there wasn't a serious problem there would be no need of a ban. In Alex's reading what Durham is saying is silly, but in the more generous, and I think more accurate reading, there is no contridiction.
Posted by: Michael Foody at Nov 21, 2007 1:32:28 PM
Chris, let's spell it out for you. Non-sequitor means "it does not follow." Alex is correct that from "80% of the
murders in DC are caused by handguns" it does not follow that "the handgun ban
is working."
Posted by: anon at Nov 21, 2007 1:33:26 PM
I don't think the question is whether DC has an elevated rate of knife murders, or whether it has higher murder rates than other cities, because (obviously, i would hope) crime rates are a function of far more variables than the degree to which the citizenry is armed or the degree to which the criminal element is disarmed.
The correct question is "If the handgun ban were repealed, would there be more or fewer murders in DC?". Neither of the possible answers to this question imply anything about the murder rate in DC _relative_ to other cities, nor do they imply anything about the frequency with which murders are committed with guns, knives, or any other weapon.
Posted by: Mark at Nov 21, 2007 1:35:36 PM
anon, you've changed the quoted sentence from:
"The ban on handguns is a matter of life and death because 80% of the murders in DC are caused by handguns."
to:
"80% of the murders in DC are caused by handguns because the handgun ban is working."
I think you're right that the second sentence contains a non-sequitur.
Posted by: chris at Nov 21, 2007 1:40:56 PM
Definition of Non-sequitur: Conclusion does not follow from premises and assumptions.
Premise 1:
80% of murders are caused by [illegal] handguns.
Premise 2 (an implied premise or assumption):
Any increase in the availability of handguns [legal or otherwise] will result in more murders.
Premise 3 (also implied)
The current ban on handguns limits handguns for murderers.
Conclusion:
The ban on handguns is a matter of life and death and must stay in effect.
The non-sequitor:
It does not follow that an increase in [legal] handguns will necessarily result in increased murders or that the ban is even effective in its intended goal.
Posted by: Robert at Nov 21, 2007 1:52:56 PM
G.ira: I was not arguing whether banning guns is good policy. Some people think having more guns around reduces gun violence, others produce statistics that indicate more guns lead to more gun violence. Certainly, the latter case has a lot of face validity, just as banning drunk driving does (even if some people make arguments that it drives teens to binge, or hide their drinking, etc.)
The question is whether the high number of handgun murders might be a reasonable reason for banning handguns. While Alex did not reproduce the entire argument Mr. Durham was making (or was thinking of) when he made that argument, it is no stretch to say that it might be reasonable to ban something that is used for a lot of murders, even if the ban is imperfect. Certainly, as other posters point out, we do that with lots of other bad things. The continuing existence of a type of crime does not mean that we should decriminalize it, or even that its a non-sequitor to support the laws against it.
Alex: The statement is certainly not an "anti sequitor" anymore than supporting other imperfect attempts to reduce crime. Furthermore, it is not necessarily a non-sequitor to point out a particular crime problem and argue that we should keep the law aimed against it on the books. It all depends on what your implicit or explicit model is of the relationship between the ban and the rate of murders. Clearly, you and the assistant police chief have different models in mind. For many (most?) Americans, not to mention people in other nations, the model that the chief seems to have in mind is the more plausible one. In that case, the evidence he provides is part of a perfectly logical argument for maintaining the ban. It seems harsh to grab a single sentence from a minor bureaucrat and jump on him for not making a complete self-contained case with it.
Posted by: A student of economics at Nov 21, 2007 1:53:28 PM
Non-criminal people in DC by a wide margin do not want guns around. Do you suppose that they will change their minds and purchase guns for self defense if they are made legal? i>
Yes. Since they never had the option, they didn't considered it. The people who oppose guns will not change their mind, but many citizens who cannot afford to move out of high crime neighborhoods may change their mind.
I don't think the question is whether DC has an elevated rate of knife murders, or whether it has higher murder rates than other cities, because (obviously, i would hope) crime rates are a function of far more variables than the degree to which the citizenry is armed or the degree to which the criminal element is disarmed.
Correct. In some cities and countries, there are knife bans in effect or being proposed because following a gun ban, people use knives. Which sucks if you have a short reach or weigh 100lbs less than your attacker.
Posted by: 8 at Nov 21, 2007 1:55:10 PM
Robert: You define non-sequitur as "Conclusion does not follow from premises and assumptions."
This is incorrect. It is defined as ""Conclusion does not follow from premises". The "assumptions" bits is not true.
Also, you slip a necessity criterion into your definition at the end when you write:
"It does not follow that an increase in [legal] handguns will necessarily result in increased murders or that the ban is even effective in its intended goal."
So you start with a too-strong definition and then slip in a criterion of necessity. This kind of sinks your argument.
For the record, if the necessity criterion were true, then anything that is not a strict law (that is to say, a law of physics - although some doubt there are any truly strict laws whatsoever) would be a non-sequitur.
Posted by: chris at Nov 21, 2007 2:06:11 PM
I would say that he is making an unstated assumption: that the handgun ban is effective enough to reduce DC murders by at least 1, making it a "matter of life and death".
I suppose technically that might count as a non-sequiteur.
In real life though, if we stated every assumption all our conversations would take many times as long and we'd never get anything done.
I think that assumption is sufficiently widely understood (even if not agreed with) that he shouldn't have to state it explicitly.
With that assumption in place, that DC has a high proportion of handgun murders is a relevant datum supporting his argument.
Posted by: TheophileEscargot at Nov 21, 2007 2:07:16 PM
(I have no idea why that came out in italics)
Robert: I just checked with the Wikipedia entry for non-sequitur, and it looks like you are just plain wrong to argue that a lack of necessity renders something a non-sequitur. Here's a part of its description of what a non-sequitur is:
"Even if the premises and conclusion are all true, the conclusion is not a necessary consequence of the premises."
Posted by: chris at Nov 21, 2007 2:11:21 PM
According to the link Alex provides, the assistant police chief did not actually say "because", so Alex's indignant reaction to his supposedly imperfect logic seems a bit overblown. Here's what the article says:
"Assistant police chief Alfred Durham says 80 percent homicides in the district this year have been committed with firearms. He says the way the Supreme Court rules in this case is literally a matter of life and death."
Alex rephrased these two adjacent sentences into a specific, causal claim and then took issue with the logic train that he created. In fact, we don't know precisely what Mr. Durham's complete reasoning is, although he does appear to believe two perfectly reasonable factual statements.
Does it rise to the level of an "anti-sequitor" to attack an assistant police chief for a train of logic that he didn't actually state, or is it just nit picking?
Posted by: A student of economics at Nov 21, 2007 2:18:38 PM
"If there wasn't a serious problem there would be no need of a ban"
It seems like a lot of peoples' argument stems from the assumption that the the proposed solution (in this case, a ban) is the only logical or effective response to the problem. In this case, it may be the logical answer (I'm no expert). However, even if it where true, that would not change the fact that the statement is a non-sequitur. For example: "Auggie is a mammal, therefore he is a human," Is a non-sequitur. It may be true that he is, but it doesn't necessarily follow that he is indeed human just because he is a mammal. It seems to me that those who argue that the statement fails to meet the criteria of a non-sequitur fail to recognize the fact that there is no single conclusion to be drawn from the first observation.
Just as Auggie could be any one of 5,400 species based on the mere fact he is a mammal, the fact that gun violence accounts for 80% of murders in Washington could lend itself to a myriad of possible responses. If he went on to describe how the A = B, that'd be more fair, I suppose.
Posted by: G.Ira at Nov 21, 2007 2:20:26 PM
