From: John Lott
Sent: Monday, January 20, 2003 1:58 PM
To: Eugene Volokh
[Critical Commentary by Tim Lambert.
This is a copy of a January 20 email by Lott to the firearmsregprof list. My comments appear in italics like this.]

Response to Lindgren’s January 17th posting:

Lindgren’s role in this process seems to be that of a prosecutor. [This is not correct. Lindgren is careful to present any evidence that supports Lott. ] He claims that he “never heard of a professor doing anything of that size with no funding . . . .” Well, I did the 2002 survey paying students $10 per hour out of my pocket. I have also done many, many projects spending hundreds or thousands of dollars out of my own pocket. The copy of my family’s 1997 tax returns that I forwarded to people on Sunday was from the year that survey occurred and showed that at least $8,750 was spent that year on research assistants. In 1996, I spent $31,000 to finance a special issue of the Journal of Law and Economics and another large amount on research assistants. Lindgren is also disturbed that I supposedly did not consult with experts about “sampling” before doing the survey.

Lindgren’s demands for documentation are ways excessive. [All he was looking for was some documentary evidence that the survey had been conducted. There is none.] As to the allegedly unusual circumstances about financial records, I do not consider it unusual that people do not keep payment records for six years, computer crash or not. Is there an established rule of how long research records should be kept? Is 10 or 20 years perhaps the standard, or what? My wife (who is the one in the family who deals with bank records and taxes) is in the habit of throwing out supporting documents after the 3-year period required by the IRS.

Now, I agree totally with Lindgren that it is generally wise to keep documentation for tax purposes (employee expenses or business expenses), a fact that my wife keeps on nagging me about, too. Over the years I have sometimes paid by check (if I am able to find one handy in my messy office), especially for larger amounts.

Lindgren suggests that it is almost without precedent for people not to keep “phone reimbursements, or records” for six years. If Lindgren has his personal telephone bills from six years ago, I will be impressed.

[ In fact, what Lindgren wrote was:

I had never heard of a professor doing anything of that size with no funding, paid support, paid staff, phone reimbursements, or records (though there are probably precedents for such an unusual method).

]

Many of the problems noted below might have been avoided if Lindgren had sent me his draft findings before he posted them simply to see if my recollections of the conversations were the same as his. I think that it would have also been useful if Lindgren had asked for my reaction to claims made by others. Instead, Lindgren never asked me for comments before postings and sometimes even posted his findings well before forwarding them to me. [This is not true. The report was forwarded to Lott at the same time that it was posted.]

Response to “Comments on John Lott’s . . .”

1) “John Lott called me shortly after I posted a notice to the FirearmsRegProf discussion list on September 15, 2002.”

I contacted Jim Lindgren after he e-mailed me, not because he posted a notice on the FireArmsRegProf discussion list as he claims in “Personal Note, January 17, 2003.” This can be verified in several ways. (1) I have an e-mail directly from Lindgren to me on September 14th (see appendix below). [It is clear from the email in the appendix that Lindgren's message was both posted to firearmsregprof and emailed to Lott. Lindgren's statement does contain a small error—his posting was made on September 14, not September 15. The most likely reason for the error is that his posting was a same-day response to a September 15 posting by myself. However, my posting crossed the International Date Line, so Lindgren received it on September 14. ] (2) I would guess that it is possible for Eugene Volokh to verify that I have not I subscribed to the discussion list. (3) I called Lindgren the night that I received his e-mail. While I don’t have the AEI telephone records handy, my guess is that since the call just took place last fall it should be relatively easy for me to have it retrieved. While Lindgren claims to have a very clear memory of this telephone conversation, he claims that I called him shortly after he posted his discussion on September 15th. Yet, I called him the night of the 14th when I got the e-mail.

This point might seem pretty nitpicky, but in the context it is not, for Lindgren creates the misimpression that I only respond to public postings attacking me. [The fact is that Lott did not respond to the emails and letters from Duncan and that is why the matter had to be raised publicly. The posting that Lott responded to on September 14 was both posted to firearmregprof and emailed to Lott. ] To the contrary, I have responded to virtually anyone who has contacted me with questions and I have tried to be forthcoming (examples noted in previous posts include Glenn Reynolds, Clayton Cramer and Dan Polsby). (Lindgren also gave a similar misimpression in his earlier comments regarding when I responded to Duncan.)

The crucial point here is that Lindgren’s memory is not perfect, despite many statements that he “carefully listened.” Getting things correct — such as why I contacted him or when the contact occurred — are especially serious lapses when he is trying to play the role of prosecutor. [Lindgren was correct that Lott's phone call followed Lindgren's posting to firearmsregprof. His only error was not a memory failure but his assumption that a September 15 posting would not arrive the day before it was posted.]

2) The “most surprising” claim by Lott

Lindgren’s first comment about changing statements deals with me supposedly “changing” my discussion to include the word “possible” as to whether I paid by check. To Lindgren, my original statement was a direct test of “assessing [my] credibility,” and that by changing my story I am failing that test. This interpretation would make some possible sense if after I (allegedly) said that I paid by check, I feared that further investigation would reveal that the bank could find no checks relating to the project. So by adding the word “possibly,” I would have guarded myself by allowing for the possibility of untraceable cash payments. However, there is one serious problem with such a posting. In the very same posting that I sent to Volokh and others on January 14th, where I noted the word “possibly,” I also mentioned in the third paragraph of the e-mail that my wife had already checked with the bank and found that they do not keep bank statements for longer than five years (Yvonne Macias in the book keeping department at University National Bank, (773) 684-1200). (Checks are actually kept for something less than that though my wife did not pin down that date.) Indeed, she had already checked this information the previous day before I had written Lindgren noting how I thought that he had misquoted me. In other words, I could not have had the motive to “changing” my story that Lindgren attributes to me. The only possible reason that I can see for pointing out Lindgren’s misquote was to set the record straight.

[However, Lindgren's main point was Lott's switch from a Jan 13 email where he was not certain about what he said in September, to a Jan 14 email where he was certain that Lindgren was wrong.]

3) “Two University of Chicago Students”

I told Lindgren that there were about 8 people involved (I am honestly not sure I remember the total), but that two were University of Chicago students. One I said was a senior, but the other one I was not certain of. (I did not mention the University of Illinois at Chicago because I was not certain and am still not at all certain if that is correct.) I assumed that it was obvious that not all the callers could have been University of Chicago students since I only mentioned two as coming from the UC.

4) Lunch with Tom Smith I am not going to repeat what I have previously written, but suppose there was a misunderstanding between Lindgren and myself regarding his question about “discussions with any samplers.” One logical reaction is to check to see whether what I said was true, and see if it provides any additional evidence for my explanations. I would have little reason to make it up since Tom Smith and I probably do not agree on these issues and he would have no reason to pretend that we had such a lunch. (Attached in the appendix is an e-mail that I received from Smith verifying the lunch, though he doesn’t remember when it occurred.)

Instead, Lindgren claims that my mentioning the lunch represents evidence that I am supposedly changing my story. While my contact with Smith was not mentioned in Lindgren’s earlier postings, Lindgren’s January 17 posting now admits that I mentioned previously talking with Smith, but there is no explanation as to why. Yet, why would my discussions with Tom Smith, who directs the National Opinion Research Center, even be raised in my September discussion with Lindgren if it wasn’t related to surveys? (Presumably, I wasn’t simply listing people that I know in the country.)

[The specific question was whether he had discussed his sampling design with Smith. Obviously there are questions about surveys he could have discussed (for example, More Gun, Less Crime uses survey data on gun ownership).]

“One just doesn’t pull a national sample out of one’s head.” This is obviously true, but when I talked to Lindgren he admits that I did use a CD with people’s telephone numbers on them. Also, I conducted a similar survey last year, without consulting any true “expert” in the area at that time. (The survey questions were discussed at length with my wife and with Russell Roberts, though.) Given that I mastered the task of pulling together this type of survey once, is it not plausible that I might also have mastered it at another time? [In his 2002 survey discussion, Lott also references a paper published in 2000 that influenced his survey design.]

5) Other supposed changes in my discussion Lindgren alleges some other “changes” in my description of the circumstances surrounding the survey. However, even if we are to accept his account of our conversations, there is no motive that can be found for me changing my account. There is no gain to me “changing” these minor details.

Conclusion

I appreciate Lindgren getting the statements from such people as David Mustard, who confirms that we discussed the survey sometime in 1997, [This is not true. Mustard is not certain about when he heard about the survey.] possibly after he got to the University of Georgia in August, and who mentioned that I lost the data in my hard disk crash. Lindgren seems to place somewhat greater weight on Mustard’s statement in his January 17th note than in his earlier writings, and I appreciate that. And recent statements from Geoff Huck provide confirmation for my claim that I lost part of the book and that the hard disk crash occurred at least before the end of July (Geoff has an e-mail about this). Many other people confirm that I lost my hard drive with everything on it. [Notice that there is no problem confirming the disk crash. The survey should be much easier to confirm.]

No one’s memory is perfect, neither Lindgren’s nor mine. But if Lindgren makes documentable mistakes about something that happened four months ago and Lindgren was at the time presumably trying to make a record of things, [Lott has indentified just one mistake, and it wasn't a memory failure.] it is hard to understand why he discounts statements from Mustard, claiming that Mustard’s memory is fuzzy about the exact dates during 1997. [Because Mustard is not certain about the dates. That's what his statement says.] Events that happened six years ago are difficult to remember even without lost documentation such as hard disks.

Appendix
Begin forwarded message:

From: James Lindgren
Date: Thu Nov 14, 2002
4:44:42 PM US/Eastern
To: John Lott, Firarmsregprof
Cc: Geoffrey Huck
Subject: John Lott’s computer crash

Tim Lambert raises the possibility that John Lott fabricated a study: “it seems increasingly likely that the survey is fictional.” These are extremely serious speculations to be made in public and need to be resolved as responsibly and as quickly as possible. Fortunately, it should be fairly easy to determine whether such a possibility is even remotely plausible. We do not need another issue in which questions, which could have been easily answered if scholars had simply done their jobs, is somehow turned into an unnecessary, protracted scandal.


—— Forwarded Message From: Tom Smith
Date: Sun, 19 Jan 2003 09:49:14 -0600
To: John Lott Subject: Re: Question

I’m sorry, but while I recall the lunch, I have no idea when it was.

______________________________ Reply Separator

_________________________________

Subject: Question
Author: John Lott
Date: 01/19/2003 2:51 AM
Dear Tom:
I trust that things are going well. I was wondering if you could help confirm something for me. Do you remember exactly when in the fall of 1996 we got together for lunch at the Quadrangle Club? I realize that this might seem like a strange thing to ask, but I would appreciate it. Was it November 1996? Thanks.
Best,
John

From: Geoffrey Huck
To: Julian Sanchez
CC: John Lott
Date: Wednesday – January 15, 2003 9:57 AM
Subject: Re: John Lott Survey
Attachments Mime.822 (3600 bytes) [View] [Save As]
Dear Mr. Sanchez:

Since I no longer work for the University of Chicago Press, I don’t have access to the Press’s files. Nevertheless, I am pretty certain that there is nothing in those files that would help you. The Press generally did not keep any draft manuscripts, tables of contents, etc., after publication – there was simply not enough space to accommodate all that that would entail. I have purged all my Press files from my home computer, and I would doubt that there are any electronic files at the Press that would answer your questions. That leaves my archived home e-mail files (some of which I still have) and my memory (which isn’t all that reliable). As to the latter, I have a vague recollection of a chapter or a section or sections of a chapter that had to be scrapped because of the computer crash, but I don’t at this stage remember the subject of it (or them). At the time, we were talking about a variety of things John could do (e.g., including a chapter on mass public shootings). As to my e-mail archives, there are a couple of brief mentions in John’s and my exchanges about the crash and loss of data, though I have found nothing explicitly about the defensive use of handguns in them (which doesn’t mean anything in itself, since we were mostly talking on the phone and there must have been all kinds of things that were lost in the crash that we didn’t discuss in the archived e-mails I still have). There is one message from July of 1997 that gives some sense of the magnitude of the problem we were dealing with, though. We fortunately did have an earlier draft of the manuscript in hard copy at the time of the crash and a disk with some – but not all – of the tables. My assistant Theresa let me know what was missing. I don’t at this point know what these tables are – you might want to ask John about that.

>Date: Fri, 25 Jul 1997 16:13:03 +0000
>To: lott
>From: Geoffrey Huck
>Subject: More about tables
> >John: Theresa tells me that she has on the disk you gave her all the
>tables from 1 to 30 except 2, 5, 16, 21, 24, and 25, and 2 appendix
>tables. I asked her to e-mail you the contents of the disk. Hope this helps.
>Geoff There is also one e-mail message I have from February 1997 which contains an attachment that I’m not at this point able to decode. If that file turns out to be decodable and has anything of relevance to add to this question, I (or John) will let you know.

I hope you find this information useful. Cordially, Geoff