Re: "Natural Law" (was "META: Rhetoric")

From: Anders Sandberg (asa@nada.kth.se)
Date: Wed Jun 11 2003 - 14:45:50 MDT

  • Next message: Robin Hanson: "Re: Investing"

    On Wed, Jun 11, 2003 at 06:01:28AM -0500, Greg Burch wrote:
    >
    > In a limited sense, I do. (As usual) without the time to be complete in
    > explaining myself, I think that some elements of social interactions are
    > universal and derive from the very nature of (thus, "natural")
    > intentional beings interacting with each other. Almost all of what I
    > would call "natural law" I ascribe to the logic discernable in game
    > theory, especially the prisoners' dilemma, most especially the iterated
    > prisoners' dilemma.

    I think "the natural laws of rhetoric" need to be based on psychology as
    well as game theory. A group of agents exchanging pieces of information
    does only explain a small subset of the interactions of rhetoric. These
    abstract agents simply have no (defined) emotionality and motivations,
    making them too general to say much about human rhetoric (they can still
    have interesting disagreements and ethical interactions). They have the
    logos and likely ethos of the old "Logos, ethos, pathos" formula, but
    lack the pathos part.

    If we add some emotional hard/soft-ware to them, things become
    interesting. A simple model would be that humans have an arousal level
    that determines the likeliehood of using a previously used or learned
    response rather than finding a new or trying something different (in
    reinforcement learning this would correspond to the temperature of the
    softmax action selection). Arousal have a clear evolutionary benefit,
    since in a crisis situation (core values such as own life in danger)
    usually fast well-trained action rather than slow introspection is
    rewarded in nature (at least *our* nature - might make an interesting sf
    story to set up an ecosystem with an opposing situation).

    This carries over into the discussions of the agents: when their core
    values are threatened, arousal goes up and their responses become more
    predictable and stimulus-response rather than creative. This means that
    overall information production goes down, and this is in itself bad
    enough. But as an additional psychological assumption, if we add
    aggression (the deliberate threat of important values) as something that
    is made more likely by high (negative) arousal, then it is clear why we
    also get a vicious circle in the discussion when the arousal goes up.
    The only way to avoid it if the basic arousal/aggression mechanism is
    unchangeable is to use cognitive tricks to insulate the discussion from
    arousal increases.

    Other laws of rhetoric likely stem from cognitive effects, such as
    primacy and recency (putting emphasis on the beginning and end pays), as
    well as the von Restorff effect showing why occasional surprises are
    well recalled and increase attention at the expense of surrounding
    material.

    -- 
    -----------------------------------------------------------------------
    Anders Sandberg                                      Towards Ascension!
    asa@nada.kth.se                            http://www.nada.kth.se/~asa/
    GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y
    


    This archive was generated by hypermail 2.1.5 : Wed Jun 11 2003 - 14:51:12 MDT