A Series of Unfortunate Events
As finem respice has acquired a copy of "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," [156kb .pdf] the study by Dr. Janet Rosenbaum recently the center of much attention in the wake of last week's awful events in the United States, it seems prudent at this point to evaluate the claims it purports to address, and the various sources of authority it cites in support of these claims. The results are not going to be pretty and, after careful review, finem respice believes there are grounds to think that Dr. Rosenbaum has violated at latest one code of academic integrity in connection with this research. As such, finem respice now publicly presents this analysis of Dr. Rosenbaum's work:
The attempted assassination of a US Representative in Arizona in 2011, killing six bystanders, renewed interest in the question whether gun control improves or reduces public safety. Gun control advocates claim permissive gun laws such as Arizona’s increase the likelihood of such high-casualty violent events, as well as homicides and suicides within the households of gun owners. Gun advocates claim gun ownership deters and mitigates the impact of violent events by increasing the likelihood of intervention by armed bystanders, without increasing firearm morbidity and mortality. They cite Switzerland and Israel as countries where permissive gun control and widespread gun ownership have improved public health and safety and make three primary claims. First, Swiss and Israeli gun control laws restrict gun ownership only minimally or not at all. Second, gun ownership and licensing rates in these countries are at least as high as in the United States. Third, these governments encourage citizens to own guns for crime and terrorism prevention and to carry private guns in public for personal defense.
This article evaluates these claims with survey-weighted analysis of four waves of International Crime Victimization Survey (ICVS) data, literature search, and translation of Israeli and Swiss gun laws. It will be useful for readers to keep in mind the populations of each of the countries: Israel: approx. 7,418,000; Switzerland: approx. 7,664,000; United States: approx. 310,384,000.
We will give Dr. Rosenbaum a pass on her low approximation of Swiss population. She was, after all, using data from the Department of Economic and Social Affairs, Population Division, Population Estimates and Projections Section of the United Nations,1 and what's a couple hundred thousand people between friends?
Dr. Rosenbaum continues:
The literature search included the following terms: Switzerland, Israel, Swiss, Israeli, guns, firearms, injury, suicide, homicide, regulations, legislation, military, civilian guard, civilian policing, and their Hebrew, French, and German translations. The databases included Proquest, Pubmed/Medline, JSTOR, Lexis/Nexis, and the Harvard University library catalog.
Swiss gun laws were obtained from the Swiss Consulate of Boston (420 Broadway, Cambridge, MA) in French and German languages and translated by the author. Israeli gun laws were obtained from the Israeli Ministry of the Interior website (moin.gov.il) in Hebrew language and translated by the author.
Data and analysis
Gun ownership data are from Israeli media reports, the Small Arms Surveys, and the ICVS. The ICVS, a nationally representative sample of United States and Swiss citizens, yields estimates on the prevalence and Gun access in Israel and Switzerland and reasons for gun ownership in 1989, 1996, 2000, 2004–2005. The ICVS questions were:
‘Do you or anyone in your household own a handgun, shotgun, rifle, or air rifle?’
‘Could you tell me which sort of guns you own?’ K ‘For what reason do you own the gun(s)?’
Survey-weighted ICVS data were analyzed in Stata. Plots used Andrew Gelman’s ‘secret weapon’-plotting method19 and were created using the plotCI function in R.
Well that all seems to be entirely above boa- uh... wait just a moment....
Swiss gun laws were obtained from the Swiss Consulate of Boston (420 Broadway, Cambridge, MA) in French and German languages and translated by the author.
It is just possible that we have begun to understand the source of the misunderstanding in this matter. Let's just see:
Results and discussion
Gun control laws
This section assesses gun control opponents’ claim Switzerland and Israel have permissive gun control laws.
Gun control laws in Switzerland
The Swiss federal government requires gun permit applicants to demonstrate need for protection against a specific risk and pass weapons safety and firearm use regulation tests (Swiss code RS 514.54, Ch. 6, art. 27 (1997)). Permit holders may own only one handgun for 6 months, and must subsequently renew permits every 3 months (Swiss code RS 514.54, Ch. 2, y1, art. 8 (1997)).
The prose to describe what a gross misrepresentation of Swiss law Dr. Rosembaum inflicts on her readers here simply escapes one. Literally not a single material assertion in this paragraph is correct. Of course, finem respice addressed these concerns and endeavored to provide a comprehensive understanding of the "Bundesgesetz über Waffen, Waffenzubehör und Munition" ("Swiss Federal Law on Weapons, Weapon Accessories and Ammunition" yesterday.2
Prevalence of gun ownership
This section assesses gun control opponents’ claims about the numbers of guns and gun owners in Switzerland and Israel.
Gun ownership in Switzerland
Poe called Switzerland ‘the most heavily armed nation on earth, per capita’, with 2 million guns. At the time of his writing, Poe was not correct: the 2002 Small Arms Survey estimated 1.2 million civilian firearms in Switzerland, or 16 per 100 residents versus 83–97 civilian firearms per 100 residents in the United States for the same period. Swiss firearm ownership increased between the 2002 and 2007 Small Arms Surveys because military rifles were released to the public due to drastic army size reductions. In 2007, Switzerland had 31–60 total firearms per 100 residents, about the same as Finland, and less than the estimated 83–97 per 100 in the United States and 29–81 per 100 in Yemen.
This section is problematic, to say the least. First, as finem respice indicated, all figures for civilian gun ownership are estimates because not all firearms are registered. The very Small Arms Survey Dr. Rosembaum refers to quips:
Swiss gun owners can point to a long and distinctive tradition of gun ownership, with roots in an armed neutrality. But unlike many countries with widespread firearms ownership, the Swiss are comparatively private about their firearms. The Swiss federal government does not keep comprehensive statistics on private gun ownership.3
Further, Dr. Rosembaum indicates that:
Swiss firearm ownership increased between the 2002 and 2007 Small Arms Surveys because military rifles were released to the public due to drastic army size reductions.
This is baffling given that the Small Arms Survey for 2002 that Dr. Rosenbaum cites in support of this proposition doesn't pretend to know the size of Swiss firearm ownership with any reliability. Quite the contrary, the Small Arms Survey 2002 notes:
The Swiss example shows how vital it is to include militiamen and reservists when evaluating the small arms of a national military establishment. The information released here offers insights into a distinctive military structure. As revealing as they are, though, these figures do not include police weapons or privately purchased firearms. Until data on those categories is collected, the total number of small arms in the country cannot be established.4(Emphasis added).
Later, we find the Small Arms Survey of 2002 suggesting that:
Several countries are gradually developing estimates for their civilian weapons inventories. In Switzerland, for example, a picture is emerging through a combination of official disclosures and public polling. Swiss federalism makes total national firearms ownership difficult to ascertain. As noted above, the Swiss military controls 695,000 automatic rifles and service pistols. Many of these are stored in the homes of the nation’s seven million people (see Box 2.3; Clasmann, 2001). The scale of total private gun ownership in the country was suggested by a recent survey. This reported that 35 per cent of all Swiss households have at least one gun (Swiss Crime Surveys, 2001). While this is not sufficient to establish the total number of civilian firearms in the country, it ranks Swiss public gun ownership higher than other well-armed countries like Australia and Canada, in which one out of four households has a gun, but below the United States, where over 40 per cent of households have at least one gun.5
Perhaps there is some obscure passage in the Small Arms Survey for 2002 that has been omitted from the OCR layer and is therefore not revealed by text searches. Perhaps this passage corrects a publishing error in the Small Arms Survey for 2002 and provides highly certain figures for Switzerland. Failing this, what "Swiss firearm ownership" figures is Dr. Rosenbaum comparing the Small Arms Survey 2007 numbers with, exactly? It is impossible to find a direct reference to this given the citation Dr. Rosenbaum provides:
Karp, A. (2002) Small Arms Survey 2002: Counting the Human Cost. New York: Oxford University Press.6
How is it remotely possible to justify citing such figures when the source itself admits that its data on Switzerland is "...not sufficient to establish the total number of civilian firearms in the country..." and elsewhere that "...the total number of small arms in the country cannot be established."?
Returning again to this proposition:
Swiss firearm ownership increased between the 2002 and 2007 Small Arms Surveys because military rifles were released to the public due to drastic army size reductions.
The SwissInfo article provides in part:
But when military service is over, even this restriction is lifted: a form is filled in and the soldier can keep his weapon for the rest of his life. The only safety measure is that the magazine is removed.
A survey carried out by the central barracks in Zurich revealed that 57 per cent of those who have completed their active service keep their rifles, while more than 75 per cent hold on to their pistols.
But the army is changing. From a strength of almost half a million men, the Swiss army is being reduced to little more than 100,000 active soldiers.
All militiamen up to the class of 1974 will be retired by the end of 2005. Last year, almost 120,000 soldiers left the service.
This means that, in just three years, at least 150,000 assault rifles and pistols will have become the private property of Swiss citizens; the official figure is a total of 324,484 weapons.9
Presumably, this article is meant to mitigate apparently rising levels of per capita firearms in Switzerland by arguing that such gains are not a product of a Swiss government that encourages firearms ownership (one of the claims Dr. Rosenbaum purports to rebut in the instant research) but rather some "off balance sheet" transaction. As if privatizing 150,000 military weapons somehow counts not as encouragement, but a deep and passive aggressive ambivalence by the Federal authorities with respect to the private ownership of firearms by Swiss citizens.
Note also the tendency of Dr. Rosenbaum to conflate figures. Firearms per capita (which may or may not include military firearms), civilian owned firearms per capita, and households reporting at least one firearm (having the effect of looking much smaller than firearms per capita figures in Switzerland) are all interchanged, mixed and matched to fit the premise of the moment. In addition, Dr. Rosenbaum is prone to subtract out those military rifles that are at home in the possession of civilian soldiers as if these were somehow distinct and less "risky" from a privately held SIG 550 in the home of a Swiss female who had performed no military service. Why should the latter be counted but not the former? Do the rifles know if they are active-issue militia weapons?
The more of Dr. Rosembaum's citations and research methods one examines, the more concerning the picture becomes.
Apparently, one must acquire a taste for mystery and the suspension of disbelief to fully appreciate the latent genius of Dr. Rosenbaum's research work. It becomes difficult, in this context, not to accuse Dr. Rosembaum of serious and negligent research if not outright academic fraud. And we are just beginning.
Next, having purported to invalidate (or simply muddle) the data from the Small Arms Survey, Dr. Rosembaum resorts to survey data from the International Crime Victimization Survey to make judgements about how many Swiss "households" have firearms. To wit:
On a per-household basis, the nationally representative ICVS data find more US households than Swiss households own handguns and at least as many own firearms (Figure 1). A quarter of Swiss households reported they own a gun for army service. Few Swiss households, fewer than 13 per cent, own a gun for non-military reasons.
Gun advocates claim the Swiss own guns due to tradition, but more than six times as many US households reported owning a gun because they have ‘always had one’ (Figure 2). Gun advocates claim Swiss own guns because shooting contests are the national sport, but only 5 per cent of Swiss households reported owning guns for sport versus 12 per cent of American households. Eight times as many American households reported owning a gun for ‘self-protection’ as did Swiss.
Leaving aside for a moment the horribly flawed methodology that has ICVS survey personnel asking intensely privacy-focused Swiss citizens questions like "Do you or anyone else in your household own a handgun, shot- gun, rifle, or, air rifle?" "Could you tell me which sort of gun or guns you own?"10 it's not clear what claim Dr. Rosenbaum is trying to make or rebut.
Moreover, some crude "gut checking" suggests some real data flaws. For instance:
Those slackers at the Bundesamt für Statistik (Federal Office of Statistics) apparently stopped downing schnapps long enough to get around to producing a report that has the cheek to suggest it knows how many "private households" are in Switzerland. For the year 2000 this figure was 3,115,400. In 2009 the slothlike work in the Bundesamt für Statistik figured that number to be 3,399,300. Taking straight-line extrapolations is dangerous (population growth isn't always linear), but splitting the middle to get figures for 2005 we get 3,257,400 private Swiss households. Using the data Dr. Rosenbaum relies on with respect to Swiss households with firearms, we get a figure of 28.6% of Swiss households reporting at least one firearm and 10.3% reporting owning a handgun for the 2004-2005 ICVS data. This would imply around 932,000 households with a firearm. If the Small Arms Survey figures are to be believed (and Dr. Rosenbaum certainly doesn't disclaim them) then the average household that has any firearms has 3.76 firearms. It doesn't take much of a standard deviation to press some Swiss households into double digit firearms ownership. This is a suspicious statistic if one is trying to minimize the scale of firearms ownership. Actually, it is a suspicious statistic period, unless households with firearms are seriously understated in the survey data.
Further, when you ask a Swiss citizen what they mean by the Swiss "tradition" of firearms ownership they mean "our tradition of armed democracy." When you ask a Swiss citizen what they mean by "sporting" use for firearms they likely aren't thinking of the national shooting competition ("Feldschiessen"). But these are cultural vagaries, and given her obviously limited command of German and French we shouldn't expect Dr. Rosenbaum to pick up on this sort of nuance. After all, only half of her paper is devoted to the study of Swiss gun culture.
And why exactly are we using 6 to 21 year old data based on field survey work and telephone interviews crunched with Stata when recent data from the Bundesamt für Statistik (Federal Office of Statistics) is readily available?
With no transition at all Dr. Rosenbaum then jumps into:
Switzerland’s limited gun access does not prevent gun violence.11
"Switzerland's limited gun access"? Where, exactly did that come from. Even Dr. Rosembaum's most recent source on privately owned firearms in Switzerland, the Small Arms Survey (2007) puts Switzerland 3rd (in the top 2%) of 178 countries for "rate of firearms ownership."
This is very jarring.
Switzerland’s limited gun access does not prevent gun violence. Greater firearm ownership predicts greater firearm suicide, homicide of females, and murder-suicide. Swiss gun owners are more likely than non-gun owners to report having seriously injured others. Respondents who owned a handgun or more than one gun reported more violence than respondents who owned long guns or just one gun. These findings imply that either owning a gun makes men more violent, or more violent men own guns, which Swiss law does not prevent.12
Or it implies that those living in violent areas (relatively in the case of Switzerland) or whom are exposed to violence more often by their geography or occupation have a greater tendency to own firearms. Rosenbaum cites a 10 year old study to support her proposition here13 and it is not clear from the abstract if corrections for baseline violence by geography were done (though it does suggest that a multivariate analysis was performed:
The role of guns and other weapons in violent acts has often been a subject of debate. The present study is based on a sample of 21,314 valid interviews with 20-year-old Swiss men, representing more than 70% of this cohort. The results show a much higher frequency of violence among owners of handguns and other weapons, but not of rifles. Gun owners also have been injured more often and they suffer more often from psychiatric symptoms. A considerable proportion of violent gun owners had previous police contacts and court appearances, suggesting that policies designed to confiscate guns would be feasible. In a multivariate model, which considered a great number of conventional criminological variables (such as delinquent friends) and indicators of psychopathology, ownership of handguns and other weapons (but not rifles) turned out to be a very important factor in explaining violence leading to bodily injury.14
Still, it is difficult to understand what the point of these observations are, particularly given this inconvenient detail in the abstract:
...ownership of handguns and other weapons (but not rifles) turned out to be a very important factor in explaining violence leading to bodily injury. (Emphasis added).
Let's review and notice that Dr. Rosenbaum conflates these results:
Swiss gun owners are more likely than non-gun owners to report having seriously injured others. Respondents who owned a handgun or more than one gun reported more violence than respondents who owned long guns or just one gun. These findings imply that either owning a gun makes men more violent, or more violent men own guns, which Swiss law does not prevent.
Is this mere sloppiness? One can only hope.
It goes on:
Government gun-related programs
This section assesses gun control opponents’ claims that the governments of Switzerland and Israel increase civilian gun ownership for security purposes.
Government gun-related programs in Switzerland
Gun advocates cite the Swiss Army as a Swiss government program to encourage citizens to use guns for self-protection. Swiss men serve in the army until age 35 (age 50 for officers), yet the ICVS data find only a quarter of Swiss households reported having a gun in their home due to Army service (Figure 2). The media report large numbers of households keep their guns, but the ICVS data find only 2 per cent of households reported keeping an army gun post-service. Some cantons allow reservists to keep their service-issued guns in local gun depots and unit arsenals rather than inside their homes, but reservists in cantons without local depots were required to keep their guns at home.15
This is almost certainly a flaw in the ICVS data when you compare to data on post-militia gun registrations from the not-particularly-industrious officials of the Bundesamt für Statistik who seem to be laboring under the impression that 900,000 former militia weapons are in private hands. If only 67,986 housholds have kept their rifles, does Dr. Rosenbaum suggest that 832,014 rifles are unaccounted for, on the black market or have been destroyed? And what does it matter if the original militia member keeps his weapon or passes it on via inheritance or otherwise transfers it? The Swiss Federal Government just took a military weapon and made it available for civilian ownership. It takes logical contortions of the highest order to somehow suggest that this is anything but encouragement.
Dr. Rosenbaum then tries to hedge by conflating "encourage ownership" with "encourage citizens to use guns for self-protection,"16 with the claim she initially purported to refute, that Switzerland and Israel "...encourage citizens to own guns for crime and terrorism prevention...."17
Is it almost over?
Swiss and Israeli gun ownership is rare, regulated stringently such as by putting the burden of proof on permit applicants to demonstrate a specific need for a gun, and neither country encourages gun ownership. The extensive gun control in both countries do not prevent guns from being associated with violent deaths, but increased gun control in the Israeli army may have reduced gun suicide. The firearm mortality evidence from Switzerland and Israel seems to concur with the public health literature.
Let's take these in turn:
Swiss and Israeli gun ownership is rare...
In the case of Switzerland this is simply impossible to state in the absence of some very creative categorizations. Sources ranging from the Swiss Federal Office of Statistics to the Small Arms Survey show Switzerland has one of the highest per capita civilian firearm ownership figures in the world. It requires highly pessimistic estimates of private firearms ownership to press Switzerland down "merely" the top 5% in the world on this measure.
Swiss and Israeli gun ownership is rare, regulated stringently such as by putting the burden of proof on permit applicants to demonstrate a specific need for a gun...
With respect to "regulated stringently" this is totally, demonstrably false in the case of Switzerland. There is no such burden of proof in Swiss law or in actual practice. The argument that Dr. Rosenbaum is referring to "carry" permits here (which some industrious apologist for her will doubtless try to make) is complicated by the inclusion of the phrase "gun ownership" in the first clause. This argument will be made anyway.
...and neither country encourages gun ownership.
Well, except, in the case of Switzerland, by providing a select fire military rifle to every serving male and then permitting him to take possession of it permanently thereafter, and, you know, that whole thing with the national shooting festival that 200,000 residents participate in every year and which is jointly funded by the kantons and the federal government and where juvenile teams are encouraged to compete for bragging rights for their kanton, school, or private association and free ammunition is often provided. Other than that, not at all. Exactly.
The extensive gun control in both countries...
Which effectively mirrors the law in Connecticut at the moment? That "extensive gun control"?
Earlier this week finem respice started this exploration into Dr. Rosenbaum's work under the assumption that sloppy research, laziness, and other forms of academic misfeasance were at work. Over the course of the last 48 hours, however, it is difficult not to come to the conclusion that Dr. Rosenbaum's work has been tainted by academic malfeasance.
The instant study appears to be laden with pervasive and material flaws, some of which are difficult to explain in the absence of malice. In the case of her interpretation of Swiss firearms law these defects appear particularly egregious. Further, the conclusions to which Dr. Rosenbaum leaps and her propensity to publicize them by seeking out exposure through mass media outlets are concerning. Coupled with her public advocacy statements on e.g., her Twitter stream one is given the impression that Dr. Rosenbaum is engaged in advocacy, not in science. Certainly, it is a strain to view her as an unbiased researcher in light of these impressions.
Regular readers will be aware that finem respice is not normally a formal publication. Be this as it may, in this case it seems important to adopt a more formal role. It is entirely possible that finem respice has overlooked some piece of information or mitigating circumstance that would explain what appear to be severe flaws in this body of work. If such exist, finem respice has not been made aware of them- and would like to be. Accordingly, and because her research is being increasingly relied upon to shape public debates on the subject, finem respice calls on Dr. Rosenbaum to issue a prominent public correction of the flaws identified in her work, explain why they are not actually flaws, or to fully retract her paper: "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland."
In the event Dr. Rosenbaum does not address these concerns in the next 10 days, finem respice will formulate a formal report regarding academic dishonesty under the University of Maryland's Code of Academic Integrity, University of Maryland being the institution Dr. Rosenbaum was affiliated with when the instant research was conducted, the equivalent standards code at the State University of New York, where Dr. Rosenbaum is currently an Assistant Professor, and the Journal of Public Health Policy, which published the study.
- 1. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, p. 47 (2012) apparently citing: "Switzerland Population (Thousands), Medium variant," Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Population Prospects (2010).
- 2. Private, Equity, "Mark to Myth," finem respice (December 17, 2012).
- 3. "Small Arms Survey, Chapter 2, p. 78, (2002). [2.17mb .pdf]
- 4. "Small Arms Survey, Chapter 2, p. 78, (2002). [2.17mb .pdf]
- 5. "Small Arms Survey, Chapter 2, p. 79-80, (2002). [2.17mb .pdf]
- 6. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, footnote 29 (2012).
- 7. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, footnote 31 (2012).
- 8. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, footnote 30 (2012).
- 9. Papacella, Daniele, "Sale of Army Weapons Triggers Heated Debate," SwissInfo (October 14, 2004).
- 10. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, p. 50 (2012).
- 11. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, p. 50 (2012).
- 12. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, p. 51 (2012).
- 13. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, footnote 35 (2012).
- 14. Killias, M. and Haas, H., "The role of weapons in violent acts: Some results of a Swiss national cohort study," Journal of Interpersonal Violence 17(1): 14–32. (2002).
- 15. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, footnote 35 (2012).
- 16. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, p 43 (2012).
- 17. Rosenbaum, Janet, "Gun Utopias? Firearm Access and Ownership in Israel and Switzerland," Journal of Public Health Policy 33, p 47 (2012).