Evaluating an Alternative Explanation for Positions on Gun Control

Kahan and Braman (2003) propose a reasonable, evidence-based argument for reframing the public conversations and debates about gun control in the United States. They note that mainstream exchanges between the media, policymakers, and other stakeholders tend to focus on consequentialist, empirical evidence for and against gun control.[1] Instead, the authors suggest that the parties involved should appeal to what they find is the best explanation available for the genesis of those beliefs: cultural worldviews. Through decades of survey data collected by the University of Chicago, they show that cultural worldviews are significantly more robust predictors of positions on gun control than an array of other individual, commonly offered explanations. The authors hope that by focusing on these worldviews, gun debate stakeholders have a better chance of shifting the debate.

I will evaluate Kahan and Braman’s argument based on its forms of reasoning, the veracity and relevance of its evidence, and the sum of its strengths and weaknesses. I will proceed as follows. In section one, I will detail the argument and gloss the less-common terms. In section two, I will describe and evaluate the evidence that Kahan and Braman provide supporting their claims. In section three, I will briefly address the additional strengths and weaknesses that I see in their evidence and argument. I will conclude that, all things considered, their argument is a valid inference to the best explanation and presents a reasonable motive for stakeholders in the gun control debate to shift their focus toward cultural worldviews.

I. The Argument, the Terms, and the Premises

Kahan and Braman (2003) offer an argument and a plea for invested parties to exercise the conclusion. Their argument can be summarized deductively as follows:

P1.    if cultural worldviews determine positions on gun control more than empirical evidence, then discussing cultural worldviews will impact the gun control debate more than empirical evidence;
P2.    cultural worldviews determine positions on gun control more than empirical evidence;
∴       discussing cultural worldviews will impact the gun control debate more than empirical evidence.

To evaluate this argument, we must first understand the less-common terms on which it relies. By cultural worldviews, Kahan and Braman (2003) mean one of primarily two outlooks and associated value sets: an individualist outlook and an egalitarian outlook. They gloss an individualist as a person who “prizes individual autonomy, celebrates free markets and other institutionalized forms of private ordering, and resents collective interference with the same” (p. 1297). They gloss an egalitarian as someone who “abhors social stratification, distrusts xthe social and political authority structures that rest on such differentiation, and favors collective action to equalize wealth, status, and power” (p. 1297). The authors did not develop these classifications but borrowed them from established theories of risk perception by other social scientists.

By empirical evidence, Kahan and Braman (2003) are generally referring to the consequentialist arguments populating the gun control debate; for example, the number of crimes that may result from the government underregulating or overregulating guns. They note that “economists and social scientists have dedicated themselves to amassing empirical data aimed at determining the net impact of gun control laws on public safety” (p. 1290).

In Kahan and Braman’s (2003) argument, premise one essentially proposes that we accept an inference to the best explanation. It is a conditional statement that can be interpreted as follows: if x is a more reliable cause of an effect than y, then altering x will more reliably impact the effect than altering y. Especially when reduced to this form, that premise is a logical, reasonable observation. The second premise contains the claim that fuels the authors’ paper. It is important to note that they do not claim that cultural worldview is the singular determinant of positions on gun control but that it seems to explain those positions better than others that have been proposed: “…the demographic clustering of gun control attitudes is suggestive, if not conclusive proof, of the impact of culture on gun-risk perceptions” (p. 1300). Premise two, in particular, requires evidentiary support.

II. Evidence and Evaluation

The most persuasive evidence in Kahan and Braman’s (2003) article is the summary of their regression analyses based on survey data. By analyzing data from the University of Chicago’s 50-year-running General Social Survey (GSS),[2] the authors found that worldview was the most statistically significant common explanation for gun control positions.[3] Since the GSS boasts over 50,000 variables of respondent data, Kahan and Braman (2003) were able to tag together several variables that indicated either an individualist or egalitarian worldview by their lights (NORC, n.d.). For example, pro-military, pro-capital-punishment, and anti-LGBTQ respondents—to name just a few of the factors—were classified as individualist/hierarchist, while respondents with opposing views were classified as egalitarian (Kahan & Braman, 2003). They compared these and several simpler variables, such as gender, race, education, and so forth, to respondents’ answers to questions about gun control. Those answers were their dependent variable. Their analyses showed that cultural worldviews accounted for “considerably more (25%) of the variation” in positions on gun control than analyses based on simple demographic measures alone (p. 1308).

It is essential to consider that evidence of a correlation is not necessarily proof of causation (Sinnott-Armstrong & Fogelin, 2014). In addition to the possibility that worldview affects positions on gun control, this correlation gives us three additional possible answers: (1) some unaccounted-for variable may be confounding the results, (2) the correlation may be a mere coincidence, or (3) perhaps it is individuals’ positions on gun control that determine their worldviews (Sinnott-Armstrong & Fogelin, 2014). Based on those possibilities, I feel confident ruling out (3) as it seems unintelligible on its face for a position on gun control to determine or even be prior to a cultural worldview. The likelihood of (2) also seems low, given that it would take an incredible coincidence for an accidental correlation to remain statistically significant across five decades of surveys. Finally, it is difficult to assent to (1) because several of the related demographics we might test as possible confounding variables—such as political orientation or affiliation—were already included in the regressions, and the authors found that “…cultural orientations are ultimately not reducible to conventional political ideologies” (Kahan & Braman, 2013, p. 1307). Moreover, Kahan and Braman’s explanation is itself a previously unconsidered, potentially confounding variable. From these analyses, it seems likely that cultural worldview is the most cogent explanation for positions on gun control.

In addition to quantitatively evidencing their claim, Kahan and Braman (2003) spend several pages explaining the potential causal interactions between worldviews and positions on gun control. Drawing on a wealth of existing research about the “cultural theory of risk,” Kahan and Braman relate that different worldviews ascribe different values to the same risks (p. 1299). They suggest that this explains the connection between worldviews and positions on gun control: those of an egalitarian worldview place greater value on the risk of harm resulting from the government underregulating guns, while those of an individualist worldview place greater value on the risk of harm resulting from the government overregulating gun. As the authors observed, these differing values of perceived risk could help explain why individuals of opposing worldviews react differently to the same piece of empirical evidence and, therefore, why empirical evidence fails to move the debate.

Additionally, Kahan and Braman (2003) offer a compelling piece of reasoning against the empirical criterion. They note that “[v]ery few members of the public possess the technical training necessary to evaluate the quality of the conflicting empirical studies…” on gun control (p. 1312). If that is the case, they continue, then “something independent of—indeed, prior to—their assessment of the data must be inclining individuals to accept one empirical claim or the other in this debate” (p. 1312-3). While evidentiary in this context, this reasoning alone is not enough to prove that cultural worldview is that other something, but it certainly helps clear a space for it.

III. Additional Strengths and Weaknesses

Perhaps the greatest weakness in this argument is its reliance on presumptions about individuals’ worldviews. As mentioned earlier, Kahan and Braman (2003) built their data analyses on the expected collective opinions of those identifying with one worldview or another. While they based these expectations on existing research by the social sciences, it remains that people are not monolithic in their opinions of things like the military or sexual orientation. We may have reason to question the veracity of information that was assembled based on potentially imperfect classifications. That said, I do not find that this weakness outweighs the overall argument, given that we are looking for the best—not perfect or singular—explanation. These worldviews seem to carry relatively consistent statistical value despite potential imperfections.

A strength that I have not yet discussed is how the argument holds up to the standards of a good explanation. Sinnott-Armstrong and Fogelin (2014) acknowledge that there is no universally established rule for such evaluation but offer a few factors to consider: explanatoriness, depth, power, falsifiability, modesty, simplicity, and conservativeness. I find cultural worldview to have greater explanatoriness and power than the other such variables, especially based on the regression analyses, but also in the narrative, risk perception terms described. The explanation has tremendous depth, detailing all that is needed to understand their argument—even an explanation of how regression analyses are intended to function. I am left with no further questions by the end of their article. It is undoubtedly falsifiable, given that one need only find a better explanation to overtake and falsify the one claimed here. I can understand an objection that the explanation is not the simplest or most modest—it does rely on some complex notions and findings from several different social sciences. Lastly, I find the cultural worldview framing conservative in that it applies existing knowledge and understanding of worldviews instead of asking us to throw out everything we know about gun control preferences and risk perception for something wholly different and new. Kahan and Braman’s (2003) performance against these general standards lends additional credibility to their claim.

The project undertaken by Kahan and Braman (2003) is both descriptive and normative. They remain agnostically descriptive in their handling of the worldviews and the competing arguments for and against gun control; however, their ultimate aim—reframing the gun control debate around a better explanation—is abjectly normative. I focused on evaluating the descriptive aspect of their argument for this paper. To that end, I simplified the authors’ argument to a deductive form, evaluated the premises, and provided the most relevant exegesis. I outlined and evaluated the more compelling evidence and noted weaknesses where I found them. Ultimately, I can conclude that the authors present a well-reasoned and compelling set of inferences to the best explanation supporting cultural worldview as the genesis of individuals’ positions on gun control. The implications of this lend themselves to Kahan and Braman’s (2003) normative and overarching goal: rather than relying on the less-impactful empirical and consequentialist explanations and arguments, it is reasonable to expect that addressing cultural worldviews is more likely to move the needle on the gun control debate.

Kahan, D. M., & Braman, D. (2003). More Statistics, Less Persuasion: A Cultural Theory of Gun-Risk Perceptions. University of Pennsylvania Law Review, 151(4), 1291. https://doi.org/10.2307/3312930

NORC (n.d.). GSS General Social Survey FAQ. The Nonpartisan and Objective Research Organization at the University of Chicago. Retrieved May 1, 2023, from https://gss.norc.org/faq

Sinnott-Armstrong, W., & Fogelin, R. J. (2014). Cengage Advantage Books: Understanding Arguments: An Introduction to Informal Logic. Cengage Learning.

[1] Due to the volume of citations, narrative references to Kahan and Braman (2003) following an initial in-text citation will omit the publication year per section 8.16 of the APA 7^th Ed. Publication Manual.

[2] The GSS has been performed annually since 1972; the population includes adults 18+ in U.S. households who speak English or Spanish; the initial target sample size of 1,500 individuals once per year was doubled to two samples of 1,500 in even years as of 1994 (NORC, n.d.).

[3] This is true apart from gender, which is an outlier that I will discuss in section three.