What is Erisology?

“Erisology” is a made up word for a made up academic discipline I think should exist. Built from the name Eris, the greek goddess of discord, it refers to the study of disagreement. I introduced the term in a post in January 2016, but this is a condensed version, intended to give an overview to those not interested in reading a meandering essay.

“Disagreement” can mean many things, but this is what I have in mind: A lot of online discourse is hostile and often needlessly adversarial (I trust no one needs to be convinced of this). I’ve got a lot of experience wasting time online and a lot of that experience is reading people arguing. Over the years I’ve come to the conclusion that the severe dysfunction so much of online discussion exhibits is at least partly the result of a limited set of pitfalls that people tend to fall into time and time and time again. The same things happen in meatspace and traditional media, but the dawn of the internet and social media turned it all up to 12.

Erisology is the study of this dysfunction and, theoretically, the attempt to fix some of it by making people more aware of how it happens and how it doesn’t always need to happen.

To be extremely brief: for many reasons we don’t understand each other nearly as well as we think we do.

The field would be extremely cross-disciplinary, because I think you really need to bring together many of bodies of knowledge to make sense of something as thorny as “online verbal conflict”. It’s a funny chicken-and-egg problem that constructing the field of erisology would require some high-level erisology itself (understanding and harmonizing very different perspectives and paradigms).

Some examples off the top of my head:

The study of cognitive biases and how they affect our thinking, breaking it in particular predictable ways.

Traditional philosophy and its discussion of the nature of categories, objects and properties; a staggering proportion of online verbal conflict concern, at its core, some variety of the question of what category something belongs in. The pitfall here is that people act as though such questions have true answers when in most cases they don’t – making it possible for two people to both be right while contradicting each other.

Data analysis and its understanding of the relationships between models and data, clusters and categories, axes and properties. Statistical modeling and interpretation issues mirror a lot of the problems that arise when people use their own particular experiences to build models of how the world works (which later clash with those of others).

Cognitive and perceptual psychology, for insights in how we form concepts in the brain and how they affect our perceptions and interpretations of what we see, giving rise to differences between people we have a tough time understanding because they are so fundamental to our mental function they slip out of awareness. Also useful is how attitudes and opinions are sometimes the downstream result of ultimately physiological differences in perception and emotion.

Personality psychology, for differences between people that may create hard-to-comprehend, subconscious divisions.

Poststructuralist theory and its conception of language as being inherently slippery and devoid of ultimate, definitive meaning. Our intuitive blindness to this causes us to misinterpret things other people say without realizing it.

Rhetoric, the art of persuasion, is partly useful because it’s a practice more than a science and as far as I know lack theory that grounds “what works and what doesn’t” in human psychology.

Anthropology and its examination of how many things we take for granted in our societies are non-obvious and somewhat arbitrary.

Literary theory and its treatment of narratives, their interpretation and how they cannot be definitive or claim absolute truth.

Epistemology, and how people take for granted different approaches to knowledge. I’m not talking so much about explicit differences like “personal revelation vs. scientific study” but more underlying differences like the balance between personal experience and statistics, empirical data vs. theoretical considerations. These differences are sometimes discussed but are present as an important factor in way more contexts than they are explicitly talked about.

Sociology and history and their theories of social construction, which are very useful when not overstated and used as a political bludgeon. They also have valuable insights about how the design of technology and institutions shape behavior.

Evolutionary psychology and social instincts, especially those related to intra- and intergroup conflict like argumentation, rivalry, social status, identity and dehumanization. An important aspect here is also recognizing that modern large-scale societies is an extremely unnatural social structure for humans and this gives rise to all kinds of weird effects.

Computer science, specifically insights from attempts to create artificial intelligence and the difficulties of modeling human concepts. Writing software also give you good habits, since it often makes you understand that accurately modeling reality is way more complicated that you first thought.

Not the purview of any particular field, but understanding reductionism and its discontents are important to a lot of erisology covering the often disappointing interactions between academic disciplines. Differing attitudes to reductionism vs. inherent semantics makes people find different kinds of explanations satisfactory.

I could go on, but that’s quite enough (I expect there are more useful bodies of knowledge that I’ve overlooked or don’t know about). In short, a lot of different research paradigms and philosophical frameworks are in play when people talk about anything even remotely abstract and/or ambiguous. And behind disagreement on even the most concrete things there is often one or several undissolved philosophical issues being discussed by proxy, and at the same time the discussion process itself is disturbed by all kinds of corrupting psychological and social influences.

An erisology research program would try to integrate basic insights and models from all these fields. It would involve describing and cataloging the kinds of issues that hide under the surface in dysfunctional discourse and the processes that make us unaware of them, contributing to the problem. Ultimately and hypothetically the goal would be to improve discourse by creating and spreading ideas and mental tools that work to defuse unnecessary conflict before it occurs, as well as clarify necessary conflict so we know what it’s really about.

For the original article, look here.