Thursday, February 8, 2018

Group Agents & the Stupid Ad Hominem Signal Argument

Pascal Boyer has a good article trying to understand the evolutionary purpose of over-the top mischaracterizations of political foes.  The case study is the infamous Jordan Peterson Cathy Newman debate.  One of the specific points is the now classic meme "so you're saying people should live like lobsters"?

Why is there an intentional engagement with a political foe via the most patently stupid interpretation of an idea that is possible?  The suggested answer is coalitional signalling.  But of course, as Boyer says, this just begs the question of why such signalling works or how it is adaptive.


Group Agents
Perhaps group agent theory formalizes the theoretical framework a bit.  Pettit & List suggest group agents emerge (quasi-real Big Brothers) when there is

  • an agent who has representational states that depict how things are in the environment
  • an agent who has motivational states that specify how it requires things to be in the environment
  • the capacity to process representational and motivational states to change the environment.
This of course requires feedback from the group to the individual (the group is able to bias the individual) and a judgement aggregation function.  

For a group agent to be stable (or more technically to have robust group rationality), you need holistic supervenience (group attitudes are determined by individual attitudes) and "robust group rationality" (individual attitudes are determined by the group).  Basically there needs to complete two-way feedback between the group and individuals.

In practice individuals vary from the group.  Variance may be significant or insignificant.  Of a long duration or fleeting.  What seems to matter most is that there is enough bias from the group to the individual that the individual conforms their behaviour to a group agent's inferred expectation.  The more variance is clawed back, the more "real" a group agent appears.

Highly moral groups have more significant biasing effects than do weakly moral or transactional groups.


Conclusion
The general reaction to Boyer's proposition that apparently stupid and patently political ad hominem attacks are a coalition signal seem to make perfect sense.  It is one of many components and signals for group robustness.  I'd even be tempted to say it is a signal that a group is adaptive precisely because it can afford to be counter or quasi-factual.

But, to go deeper, I'd suggest the Group Agent lens adds additional depth about what is happening with the group.

Patently stupid ad hominem ruts signal that there really isn't another way to view the issue being discussed.  While it is certainly theoretically possible, it is not practically possible.  This really sets group boundaries.  A between-group competition landscape is set.  But more than this, these artificial quasi or counter factual boundaries reflect a moralized group landscape.  Moralized group landscapes facilitate the emergence of group agents / quasi-real big brothers.

Hence what the Newman dynamic probably shows is that her group is getting to the point where deviance away from group morality is likely to be so pernicious that one is even tempted to deny that the sky is blue.  While its tempting to think of this as a proximate expression of adaptive group dynamics, I suspect it is more precise to say it serves as a way to energize or reflect the emergence of a moral big brother / moral group agent.

So, what you are really seeing is a concrete example of how religious-land quasi-real group agents/moral big brothers emerge.

No comments:

Post a Comment