Argument Repair

It’s easy to blame current problems with public discourse on the brevity of on-line communication. People are moving away from blogs to image exchanges. TLDR. Limited word counts (such as on Twitter) and the combination of immediacy with anonymity lend to verbal fighting. Quick comebacks receive rewards of “likes” and “thumbs up”. We might mitigate or even reverse this momentum by directing attention to how we engage other people when we argue with them, and so I recommend we make argument repair part of our practice and our teaching.

Constructive engagement seems to require a longer format and detailed reasons. That analysis jibes with the dual process model of cognition developed and made famous by Daniel Khaneman, especially in Thinking, Fast and Slow. According to dual process theory, quick thinking tends to employ biases of various kinds, including basing views on single experiences and other problematic ways of generalizing but also basing views on social stereotypes.

The answer then seems to lie in fighting the tendency to reason quickly. In North America, Australia, and the UK, critical thinking courses that involve argument analysis and evaluation tend to be a standard part of college and university education for students in all sorts of fields. Perhaps this could be done as a more basic part of education, in secondary school, and in more post-secondary education.

Yet, we need to consider (for this and other reasons) whether the way we teaching critical thinking suffices to resist the pull of quick tussles on the internet. A strategy for slowing down reasoning and an alternative model of argumentation would help.

A model for this can be found in Richard Epstein’s notion of argument repair, which involves improving other conversants’ arguments by deleting weak premises and adding strong ones. Related techniques can be found in reasoning textbooks by Sharon Bailin and Mark Battersby and by Michael Gilbert. All of these treatments are more specific than vague appeals to charity in interpretation.

Epstein suggests that we are justified in adding a premise or conclusion if and only if all the following hold:

1. The argument becomes valid or strong
2. The premise is plausible and plausible to the other person.
3. The premise is more plausible than the conclusion.

The first and third of these we can sort out ourselves. The second requires conferring with the person whose argument seems unacceptable on the face of it. In all cases we treat other people as engaged with us in a shared project of understanding. So, argument repair builds both reasons and interpersonal relationships.

Epstein explains that argument repair implicitly involves a certain regard for other arguers. We must assume they:

1. Know about the subject under discussion.
2. Are able and willing to reason well.
3. Are not lying.

Often these may not be true of people we argue with, and certainly not all disagreements are sites for argument repair.

In fact, a certain sort of troll preys on those who adopt this pro-social attitude in the trolling technique known as “sealioning” (the term based on a comic, here). Sealions are antagonists who play at desiring respectful discussion and enlightenment while feigning ignorance. The main harm done is to waste time and annoy the other person.

It would be great to find ways to spot sealions, avoid them, and deflect their trolling, but I’m not sure there’s a simple strategy to be had. A culture of argument repair need not contribute to the problem, however. A more robust culture of on-line respect in which the limits of reasonable questioning and reasonable expectations of assistance becomes clear could provide an indirect strategy against sealioning. While argument repair provides some improvement on the principle of charity, more could be done to develop it.