Deltoid

Don B Kates, Jr. writes:

In vol. 62 # 3 (1995) of the TN Law Rev, Henry Schaffer, three professors at Harvard and Columbia Medical Schools, and I have an article evaluating the medical/public health literature on firearms. Our general conclusion goes beyond simple negativity. We conclude that it is not just methodologically incompetent, but an ideologically based “literature of deceit.” In almost 90 pages and with over 360 footnotes we document that the literature meets the specification of a model based on the law of actionable fraud, including overt misrepresentations, partial statements used to deceive by failure to reveal the entire truth, and fraudulent non-disclosure of material facts. Among the practices we document in wearisome detail are: a) the statement of falsehoods and false statistics “supported” by references to non-existent reference sources or sources that do not address the issue they are cited as proving; and b) reference to “supporting” references which do address the issue but say the OPPOSITE of what they are cited as saying.

Dear Mr Kates, I have read your paper and I have some questions that I hope you will be able to answer. In Section XIII “A Critique of Overt Mendacity” you accuse several authors not just of being wrong but of actually lying. It seems to me that it is much harder to establish that someone is lying than it is to establish that they are in error since you must prove that knew what they said was wrong as well as showing that it was wrong.

The only support I found for the charge of lying (as opposed to the charge of being wrong) was on page 578 where you state that “innocent mistakes tend to be random and to balance each other rather than all erring in favor of the position for which they are presented”. I feel that if you observe such a pattern it does not prove that an author is lying, since it can easily result from unconscious bias on the part of the author, where data that supports the author’s position is not examined as carefully as data that does not. Such a pattern can also result from unconscious bias on the part of the reader, where the reader only notices those errors that support the author’s position, and where the reader may have counted as errors things where there is room for honest disagreement and things where the reader has misunderstood what the author was trying to say.

When I read section XIII and checked it against the sources I found what appeared to me to be ten errors that all happened to support your position, and only one correctly identified such error by the authors that you say were lying. I am hoping we can determine which of the possible explanations I mentioned in the previous paragraph applies.

Here are the details of what I found (all your statements come from pages 577 and 578 of your paper):

A. Dianne Schetky.

You state:

1. that she claimed handguns account for only 20% of the firearms in use today, but they are involved in the majority of criminal and unintentional firearm injuries.

2. that she claimed handguns comprise only 20% of all guns, yet are involved in 90% of gun accidents and crime.

3. that she falsified a citation for 2.

4. that she claimed that 90% of gun murders are committed with handguns.

I could not find any of the above claims in her paper. Not one. The closest she comes is a statement that handguns are involved in 50% of homicides.

Have I misunderstood something? Can you explain please?

B. Gordon Smith and Henry Falk.

You state:

1. that they claimed handguns account for only 20% of the firearms in use today, but they are involved in the majority of criminal and unintentional firearm injuries.

2. that they claimed handguns comprise only 20% of all guns, yet are involved in 90% of gun accidents and crime.

3. that they falsified a citation for 2.

Smith and Falk do not make claim 2. They do make claim 1.

You state that claim 1 is “false in every respect”. That is, each of these claims are false:
1a. handguns account for 20% of the firearms in use today and
1b. handguns are involved in a majority of criminal firearm injuries.
1c. handguns are involved in a majority of unintentional firearm injuries.

I am puzzled as to why you say this. According to the data you present elsewhere in your paper, the only one that is false is 1a.

1a: According to your table 3, they should have said “handguns account for 33% of the firearms in use today”, ie. 1a is false.

1b: In footnote 268 you give figures that show handguns are used for 70-75% of firearm homicides. In footnote 179 you estimate that long gun injuries are 4-11.4 times as deadly as handguns. If these estimates are correct, it follows that handguns are involved in 90-97% of criminal firearm injuries, ie. 1b is true.

1c: In table 2 you give figures that show handguns are involved in 41% of fatal gun accidents. In footnote 179 you estimate that long gun injuries are 4-11.4 times as deadly as handguns. If these estimates are correct, it follows that handguns are involved in 74-89% of unintentional firearm injuries, ie. 1c is true.

I am also puzzled as to why you describe the citation that Smith and Falk gave for claim 1 as a “fabrication”. The citation only supports part 1b of the claim and not parts 1a and 1c so you could certainly criticize them for not fully supporting their claim, but calling their citation a “fabrication” seems unjustified.

Have I misunderstood something? Can you explain please?

C. Sloan et al

You state

1. that Sloan claimed that the 1978 Canadian gun law caused Vancouver to have less homicide than Seattle and

2. that Sloan lied when he claimed that the “intent of our article was not to evaluate the effect of the 1978 Canadian gun law”

Now, I don’t believe Sloan’s article shows that much, (since a correlation based on two points is not going to be statistically significant) but when I read it, they seemed to be stating that Vancouver’s more restrictive laws are associated with lower rates of homicide, not that the 1978 law was specifically responsible. In five earlier pages of discussion of Sloan’s paper you never mention the 1978 law. You also state that Centerwall’s data in his paper “Homicide and the Prevalence of Handguns: Canada and the United States, 1976 to 1980″ contradicts Sloan. As is evident from the title, Centerwall uses pre-1978 data, so if Sloan really was making the claim that the 1978 law was responsible, Centerwall cannot be used to contradict him.

The quotation in 2. is given correctly. Right afterwards Sloan et al state: “[The 1978 law] was just one step in a continuum of national policy over the past few decades moving towards stricter control of handguns.” I am puzzled as to why you did not quote this or comment on it.

Have I misunderstood something? Can you explain please?

[Writing to Don Kates] You asserted that handguns are involved in less than 50% of criminal firearm injuries. You dismissed my calculation that the data in your paper implied that the percentage was 90-97% as some sort of trick. Could you please tell me what you consider the correct value of this percentage to be?

Don B Kates, Jr. writes:

I answer: The correct value is determinable only from actual statistics. So far as I know, no statistics are available on the percentage of injuries involving handgun versus long gun crime. (Conceivably, the NCVS have such data, but I am not aware of it. The Uniform Crime Reports do not report it.)

Table 5.7 of “Point Blank” gives the fraction of NCVS measured injuries inflicted with handguns (.021) and the fraction with other guns (.004). This implies that about 85% (4/25) of criminal firearm injuries are inflicted with handguns.

Schetky purported to provide a value and Smith & Falk provided a less exact one. They both FABRICATED references supposedly supporting their figures. That was my primary concern.

Schetky made no claims about criminal firearm injuries (but rather about criminal firearm misuse). Smith & Falk’s claim that handguns are involved in the majority of criminal firearm injuries is supported by the reference they cite: you state this gives an estimate of handgun use in 70-75% of firearm homicides. I’m not aware of anyone who believes that handgun wounds are on average more lethal than long gun wounds, so it follows that handguns are used to inflict at least 70% of firearm injuries, which is certainly a majority.

I do not think “fabrication” is the correct characterization of what Smith and Falk did.

On the other hand, you did not give any reference to support your (false) counterclaim that handguns are NOT involved in the majority of criminal firearm injuries, and now I found that your counterclaim was not based on any facts at all.

I think, perhaps, that “fabrication” might the appropriate word to describe your counterclaim.

As far as I can tell, you have completely ignored what I wrote about the Sloan article. Is it your assertion that prior to 1977 Vancouver’s gun laws were identical to those of Seattle?”

I answer: The Sloan study is dedicated to showing that severe control over the possession of handguns greatly reduces homicide. Until 1977 Canadian handgun control was little more severe than American. The 1977 law classified handguns as “restricted weapon” and required for simple possession that the owner have applied for and been issued a Restricted Weapon Registration Certificate. See generally, the Kopel article “Canadian Gun Control,” 5 TEMPLE INT’L. & COMP. L. J. 1, 9-18 (1991). This is the type of legislation the Sloan article addresses and exalts — according not just to us, but to Sloan (as we quoted) and other anti-gun public health advocates whom we quoted describing the Sloan article (whom we also quoted).

As you admit, prior to 1977 Vancouver’s gun laws were not identical to those of Seattle. You consider those differences to be insignificant, Sloan does not. You seem to believe that it is impossible for anyone to honestly hold a different opinion from you on this issue.

Furthermore, I note that while Centerwall (whom you cite approvingly) disagrees with Sloan on the efficacy of Canadian gun laws he appears to agree with Sloan that there were significant differences in the gun laws before the 1977 law, since he uses pre-1977 data in his comparison.

A number of other “public health” students of firearms policy (whom we also directly quote) also say that they take Sloane et al to have been evaluating the effect of the 1977 law. Indeed, how could they not have been? The hypothesis is: the law has some sort of effect on peoples’ behavior and hence on the statistics that reflect firearms mortality. Is it now claimed that a rising homicide rate and a more restrictive gun control law are somehow consistent with the premise of gun control laws?

The two references that you cite in your paper on the effect of the 1977 law on homicides (in footnote 144, Mundt and Mauser) both found that the law was associated with a decline in the homicide rate.

If that is indeed the case, one is bound to suspect that further discussion is futile, as those who hate to think of firearms in private hands would evidently then be in possession of a theory so powerful that it is substantiated by any possible evidence and is consistent with every possible state of the world. Theories that powerful are religious, not scientific.

Mundt and Mauser both felt that the decline was caused by other factors. Does this imply that they are in possession of a religious “gun control does not work” theory?