Muki Haklay’s (2022) response, “Is Convergent Realism an Appropriate Method for Evaluating Ethics?”, to my article “The Method of Convergent Realism” (2022) did not criticize the method’s application to the natural sciences or mathematics, but raised three challenges to the suggestion that the method should guide the evolution of research ethics practices. I believe that some of Haklay’s points did not actually challenge my essay, but invited useful clarification which I would like to make herein. On some other points, I would like to invite clarification from Haklay. This response discusses Haklay’s challenges in reverse order, saving the biggest for last … [please read below the rest of the article].

Image credit: A.Davey via Flickr / Creative Commons

(3) Technology to support the proposed research ethics practices is not yet realized. 

Regarding Haklay’s third challenge, I agree that the full potential of technology to support the method of convergent realism is not yet realized (as is the nature of technological progress), and I appreciate the opportunity to clarify that one of the ways the method guides us is to guide the development of that technology. Haklay specifically pointed-out that engineers are still striving to reduce the bias, opacity, and risks to personal information of search engines and other recommender systems.

The method of convergent realism requires reasoning to be communicated, and search engines help accomplish that by making reasoning more discoverable, so the method of convergent realism gives us reason to expect that the success of these engineers would accelerate convergence towards truth in research ethics (as well as in other fields). Thus, Haklay’s example is one in which the method succeeds in providing guidance, rather than one in which the method falls short; however, Haklay has a legitimate concern to raise about convergent processes and enabling technologies in general.

Scholarly work often includes the development of enabling technologies, but, if a convergent process were to guide its practitioners to pursue an enabling technology that they think cannot be achieved, such as a time-machine, would it not be fair to abandon the process until the technology has been proven possible? And would it not be fair to reject the process, even when suspecting that its demands can be met, simply because our suspicion might be overly optimistic? The trap that convergent processes can fall into by demanding the impossible is known as “the problem of local maxima,” and is a problem for some convergent processes but not for others.

What puts the method of convergent realism in the latter category is that the method of convergent realism is parallel, leaving some scholars working to improve search engines, for example, while other scholars follow other approaches likewise justified by the method. The “unit of convergence” discussed at the end of the essay is a community that does not become stymied as a whole over any specific technology. The only specific technology required by the method of convergent realism is the method itself.

(2) The cited example, Belleville Research Ethics Committee (BREC), might not be viable.

Haklay’s second challenge was skepticism about the viability of the Belleville Research Ethics Committee (BREC), the design of which was frequently cited as an example of the application of the method of convergent realism. I appreciate this opportunity to clarify that BREC was cited as evidence that every step of the method of convergent realism can be applied and can yield novel results, rather than as evidence that it should be applied. Had I cited any success story as evidence that the method should be applied, then my argument would have been fallacious in the way criticized by Laudan (1981). To put it briefly, only time will tell which stories are truly successful. Therefore, the way in which my essay argued that the method should be applied was by citing the risks that would arise (and have arisen) were a given step of the method to be skipped.

Meanwhile, just so readers do not get the wrong idea about BREC, I should clear-up some apparent misunderstandings. Unlike a jury, selection of BREC members was not at all random; members were carefully selected to maximize its credibility within the village of Belleville. Researchers seek review from BREC because they propose to conduct experiments in Belleville on the people of Belleville. BREC is designed to enable researchers to see their research plans from the perspectives which those people would most want taken into account.

For example, the local physician on BREC happens to be an academic, but she was selected because of the local trust she has earned by managing the health of our community. Even better than a review by a doctor, we prefer a review by our doctor, so I consider the selection process of BREC-style committees a relative strength. Furthermore, BREC is more reliable than conventional research ethics committees at including relevant expertise because its procedures oblige it to bring-in any missing relevant expertise via consultants (whereas conventional committees likewise cannot include every relevant expertise among their members, yet have caseloads too large to permit a two-hour-long consultation for each review).

(1) There is “no clear truth” on which ethics evaluation can converge.

Haklay’s first and “foremost” challenge was that “there is no clear truth towards which the activities of these multiple [Research Ethics Committees] can progress” (33). I would like to invite Haklay to clarify how this is a challenge to the suggestion that the method of convergent realism should guide the evolution of research ethics practices.

Haklay cited evidence of disagreement between research ethics committees. However, every convergent process comes with a history of disagreement, so this evidence does not seem to raise any challenge to the essay. Haklay further claimed that “ethics codes continuously evolve, are an intensely social process, and need continuous interpretation in order to make them operational” (30). That claim likewise seems consistent with a convergent process. Haklay also referred to his response as written from a “social constructionist position, mixed with critical realism” (29) and referred to the method of convergent realism as “relevant for natural science research projects … [where we might] assume that there is a single truth …” (29). Haklay acknowledged that a series of revisable theories arise in research ethics, as in natural science, but then incorrectly supposed that the method of convergent realism might aim to converge on these “temporary truths” instead of on the real moral truths that the theories aim to describe. That makes me wonder whether his foremost challenge is simply to agree with about 40% of philosophers who deny that any real moral truths exist (Weinberg 2021).

But even if real truth existed only in the natural sciences and not in ethics, how would that be a challenge to the suggestion that the method of convergent realism should guide the evolution of research ethics practices? Even if no moral truths existed, moral realists do exist. Moral realism is such a basic perspective that it would come back, even if Thanos eliminated all current moral realists with a snap of his fingers. Moral realists do aim at a method (even if it converges on something other than real truth), and the rest of the world grants moral realists some influence over the evolution of research ethics practices, such that even non-realists become required to submit their moral claims to independent review (as though moral errors are possible). Would you propose to revoke this kind of participation by moral realists in the evolution of research ethics? If not, then are you agreeing that the method of convergent realism should continue to guide the evolution of research ethics practices? Is there some non-realist method that you think should also have been mentioned?

I hope the appearance of the term “realism” in the name “the method of convergent realism” will not cause non-realists to interpret my essay as a manifesto designed to unite moral realists against everyone else. Moral realists bring a passion that can make social divisions deadly for entire communities (e.g. via war), so resolving disagreements between moral realists would not necessarily be “against everyone else.” Better clarity about the method that realists aim to follow could be good for bystanders too. The way in which disagreements get resolved matters, of course, and I am truly interested in any concerns non-realists may raise about my attempt to better-articulate the method of convergent realism because I realize that the way in which realists understand their method impacts non-realists–the method belongs to everyone in that sense.

Summary

I interpreted Haklay’s last two challenges as invitations to clarify subtle points. We might not actually disagree about anything in these areas (please correct me, if I’m wrong about that). Regarding his first and foremost challenge, I think Haklay is the one with the subtle point to make. Rather than risk misrepresenting him, I’d rather ask him to clarify what he means and what is the challenge to the suggestion that the method of convergent realism should guide the evolution of research ethics practices.

References

Haklay, Muki. 2022. “Is Convergent Realism an Appropriate Method for Evaluating Ethics?” Social Epistemology Review and Reply Collective 11 (10): 29-34.

Laudan, Larry. 1981. “A Confutation of Convergent Realism.” Philosophy of Science 48: 19-48.

Santos-Lang, Chris. 2022. “The Method of Convergent Realism.” Social Epistemology Review and Reply Collective 11 (1): 33-49.

Weinberg, Justin. 2021.”What Philosophers Believe: Results from the 2020 PhilPapers Survey.” Daily Nous November 1. https://dailynous.com/2021/11/01/what-philosophers-believe-results-from-the-2020-philpapers-survey/.

Categories: Critical Replies

Tags: Belleville Research Ethics Committee, Chris Santos-Lang, convergent realism, Larry Laudan, Muki Haklay, research ethics