Can Advocacy Research Involve Abolitionists?

What a recent Ecorazzi article gets wrong, and why

I recently wrote an in-depth essay arguing that much of the research in “effective animal activism” is not only deeply flawed but pseudoscientific, adding to earlier such criticism by Casey Taft and Wayne Hsiung. Yesterday, abolitionists joined the conversation, with an essay by Ben Frost claiming that the Faunalytics Animal Tracker Survey lacks any validity whatsoever, before diving into a litany of complaints about the researchers’ biases.

Unfortunately, Frost’s analysis reveals a lack of understanding of survey methods, and hardly justifies his vitriol. His main concern appears to be that the survey’s response rate was 26%, whence he concludes that “…the results can’t reliably tell us anything. All they tell us are the opinions (subject to numerous limitations) of the 26% respondents who actually bothered to engage with the survey in some way.”[n1] Of course, the mere possibility of non-response bias does not invalidate the results, and political pollsters often work with far lower response rates.

If we look beyond this, Frost largely criticizes the survey for its perceived irrelevance to the concerns of abolitionists, and the bias in the interpretation of its results. For example, he lambastes the questions about how knowledgeable respondents feel of animals in various situations, writing

What does this show? Absolutely nothing. It doesn’t make a blind bit of difference how “knowledgeable” the person you’re educating is about the particulars of greyhound racing or slaughterhouse procedure; the only thing that matters is whether or not they have moral concern. … Such compartmentalisation of issues suggests that these issues are not related, but separate entities of their own; that there’s somehow a difference between exploitation X and Y, one that is relevant for the purpose of “eliminating suffering” in that particular case.

This type of criticism does not suggest the results are inaccurate, but rather that they are useless to abolitionists who abhor single issue campaigns. However, it’s not clear that abolitionists have nothing to gain from this information. Gary Francione has frequently used the Michael Vick dogfighting case as the starting point of an argument for veganism as a moral imperative. Potentially, knowing what issues already register most with the public could inform abolitionists about which ones to refer to when making similar arguments. While these questions were apparently designed with single issue campaigns in mind, it certainly doesn’t hurt abolitionists to have access to the information. More broadly, though, Frost expresses skepticism of the idea that educating the public on these matters is even desirable, suggesting that a focus on specific problems may distract from the fundamental issue of animal rights:

Without abolitionist vegan education, having more “knowledge” on single animal issues is worse than useless. All it does is ensure that people are kept looking through cracks in the wall as opposed to knocking the wall down and seeing the entire picture; it perpetuates the myth that animal issues exist separately and that there is a morally relevant divide between each of them.

I would not like to suggest that informing the public about the plight of animals is a bad thing, nor that research should take care to omit information which abolitionists might find irrelevant. However, there are some good reasons for abolitionists to be disappointed with Effective Altruism oriented research. Generally, the incrementalist researchers fail to consider a range of perspectives and are hasty to offer strategy proposals which do not quite follow from the findings. The Animal Tracker Survey found high levels of “cognitive dissonance,” meaning that respondents expressed views consistent with animal rights that were far out of alignment with their behavior. There are several ways to interpret that. For Frost, it is evidence that the public is primed for a clear animal rights message which entails that veganism is morally obligatory. For Matt Ball, it probably means that we are “already on the same page” with most of the public, so we don’t have to focus on changing their attitudes or promoting animal rights. For Faunalytics, it means that “It’s not enough to speak the truth and then expect people to internalize it and change their behavior. Persuasion is a more nuanced process….” And for Francione, it probably just reflects the welfarist ideology which has been dominant in the West since the late 19th century.

The questions are also tailored to an incrementalist point of view, most notably the one which asks “How important to you is the protection of animals when it comes to making the following personal choices?” and lists activities such as “going hunting or fishing” and “attending circuses or rodeos.” This is poorly worded, apparently confusing abstinence from such activities with concern for humane standards —many hunters say they are concerned about the animals they hunt. The survey could likely be improved by efforts to differentiate between welfare and rights positions.

I titled this post Can advocacy research involve abolitionists? Any non-abolitionist reader of Frost’s essay is probably not excited about the prospects. However, I believe there are good reasons why researchers can and should seek ideological diversity. The use of evidence is not inherently at odds with abolitionism, and people with fundamentally different perspectives on what is likely to be effective advocacy probably can offer insights or point out problems in study design that incrementalists would overlook. Effective Altruists often act as if it is obvious that incrementalist approaches are the best ones, but this is far from the case, and the evidence for this belief is lacking. Moreover, the extreme belligerence of the conflict between abolitionists and incrementalists is cause for real concern that tribal thinking may impact the judgment of either side. As the goal of effective advocacy research is (allegedly) to find objective evidence for the effectiveness of a particular approach, a collaboration, if it could be arranged, would be a desirable step toward eliminating such ideological bias.

One Humane League study, of the type I have criticized as pseudoscientific, is particularly illustrative of the problems which may arise when incrementalists aim to evaluate abolitionists’ methods, as well as some of the reasons for abolitionists’ hostility toward Effective Altruism. In this one, titled Report: Is Animal Cruelty or Purity (“Abolitionist”) Messaging More Effective?, online survey takers were shown three-paragraph messages and asked how much they intended to change their diet, and whether they would like a free Vegan Starter Kit. 18% of the “anti-cruelty” group and 19% of the “abolitionist” group ordered a Vegan Starter Kit. However, finding that an “anti-cruelty” message prevented 350 “days of suffering” compared to the 241 “days of suffering” prevented by the “abolitionist” message, The Humane League concluded that their message was better.[n2] By this time, The Humane League was well aware that controlled studies had never shown any statistically significant positive effect in response to vegan or vegetarian messaging of any kind, and that self-reporting could not be relied upon to reflect the reality of people’s dietary habits, but this didn’t stop them from interpreting the result as supporting their view.

Importantly, this study was designed in such a way that it couldn’t possibly have proved the abolitionists wrong. Abolitionists very specifically don’t believe that people can be convinced to undergo significant lifestyle changes in response to three paragraphs of text, and their materials are not intended to be used in this way. The leaflet THL used as a basis for the “abolitionist” message came from the International Vegan Association, a Boston-based group which closely follows the recommendations of Gary Francione. But this group only distributes its leaflet as a supplement to lectures and tabling, and frowns upon indiscriminate leafleting. Additionally, the study only attempted to measure behavior change, which is The Humane League’s priority based on its focus on short-term suffering reduction, and not attitudinal change, which is abolitionists’ priority based on their focus on a long-term effort to establish fundamental rights for sentient animals. And even the title is polemical: abolitionists find the suggestion that their philosophy is based on “purity” to be derogatory.

A more useful study might try to compare humane education talks to similar abolitionist talks. This would have to be done with the cooperation of abolitionist lecturers, who would establish a standard script. It would also necessarily involve finding common ground in determining what to measure, since abolitionists are simply not interested in “days of suffering prevented,” nor is that a realistically measurable quantity. Hopefully, it would involve gathering qualitative, not just quantitative information. Most importantly, it would be conducted by impartial researchers, since we cannot accept studies by organizations on their own methods. Such a project would be expensive and complicated, but could also be tremendously informative.

While no empirical research can possibly bridge the fundamental divide over basic principles, it’s imperative to ensure that advocacy research doesn’t simply aim to prove one position right. Because of the strong feelings involved on both sides, ideological diversity, and serious consideration of differing opinions, are probably essential to maintaining objectivity.

[n1] ^ Interestingly, he also writes “The results clearly demonstrate a dire need for the embrace of abolitionist education, yet the opposite is promoted.”

[n2] ^ These numbers should not be taken seriously as, due to extreme methodological flaws, the study provided no evidence of a genuine effect.