Saturday, February 24, 2018

Quasi-factual Metaphorical Truth (QFMT)

For the last couple of months I've been exploring the role quasi-factuals expressed as metaphorical truths play in societal stability. I've mainly been looking at social capital and social coherence factors.

I come at this from a multi-level cultural evolution lens. I just don't think you can get very far on these questions without multi-level selection.  

Here's the summary of where I am at.


Quasi-factual metaphorical truths play a role in societal stability. What this role is, is of course, uncertain. As far as I know, it has been weakly explored. 

But the importance of understanding the "irrational" side of our human nature is growing in importance. The growth of popularist mob mentalities threaten the existential survival/coherence of many western countries. This may come from xenophobic reaction politics. But it may also come from equivalent dynamics expressed by quasi-religious secular progressives.

Understanding the science of the quasi-religious movements that threaten to balkanize society is important. It is also important to understand how far removed from quasi-religious thinking we can push society before it breaks. After all, it is likely religion and formal governance have co-evolved in a lead-pace relationship (Norezayan, 2013).

I posit three potential engines capable of starting & driving the evolutionary process for quasi-factual metaphorical truths (QFMT's):
  • low cost signalling
  • memetic fitness
  • orientational cycling between adjacent levels of selection.
These engines produce a space where quasi-factual metaphorical true (QFMT) slogans/moral can emerge. But it is QFMT's role as possible moral lynchpins that is most interesting.

Once these QFMT's represent the framing and interpretational lens of a spandreled moral belief, I suggest they play an evolutionary role in;
  • norm detection/enforcement costs, 
  • technocratic trap escape, and,
  • new landscape exploitation.

This is a "just so" evolutionary approach.  "Just so" approaches are extremely weak. They only bar they pass in one of plausibility. They do not even come close to necessity.  But, part of the reason why I write this blog is to see how far I can take unorthodox ideas...



Memetic Fitness
Ideas that are slightly counter-intuitive are remembered better than those that are highly counter-intuitive or completely intuitive (Norezayan 2013; Harmon-Vukic, 2012; Upal 2012, Atran, 2004).  For example here is an easy overview of things from the classic Mixing Memory blog.

In a world of between group competition, the presence of group-oriented altruists matters. But you don't need very many of them for your group to grow. You just need a seed (Wilson, 2015).  Stability of course may require frequent re-injections. This can happen via migration. But you also need a mechanism for self-organization. Altruists who make open signals are very easy to take advantage of. Hidden signals are one of many low cost/risk mechanism for self-organization.

Metaphor and analogy are possible hidden signal mechanisms. For example, why have 1984 and Animal Farm been so successful? Religious parable & fable?

A recent example by Bret Weinstein is an entertaining way to illustrate this. Are whales fish? This actually present a three-level rather than the apparent two-level analysis. It just so happens that two levels have the same answer, albeit for remarkably different reasons.
  1. Yes - they are in the ocean and swim
  2. No - a whale is a mammal
  3. Yes - their phylogenetic ancestor is a fish.
A yes answer won't get you in trouble with a intolerant low-educated mob of fishermen, but it also allows some game play where "sophisticated" people can tease each other out to see if they understand alternative interpretations. Basically you're just signalling for intellectual curiosity and non-conformity.

Orientational Cycling
As the fitness between two adjacent levels of selection becomes more equal, the degree of complexity of the cycling between these levels increases (Okasha, 2006). In practice this means human groups oscillate between tribalization and cosmopolitanism. Here's one example of what this looks like.

Starting & Driving
Slightly counter-intuitive ideas are remembered better than their alternatives. Ideas which allow multiple levels of meaning enable sub-groups to self-organize via low cost/risk signalling. As orientations cycle, memes refine. Surviving memes should have some flexibility in their ability to work for both higher and lower levels of selection (group orientation).  In other words, bifurcation of interpretation is fit solution over long time spans. The non-variant kernel is likely to be something robust for that particular landscape (physical & social). After all, it the meme is not useful, why would it survive?

I don't think any particular order is needed here. Cycling ensures iteration. 

Stable State
What might get produced is a slightly counterintuitive meme. Memetic fitness biases things for slight counter-intuitiveness. Exactly how this counter-intuitiveness looks can, of course, vary. It doesn't have to involve angels or minor miracles.  Other options include purposeful over-generalizations, allegory, and comedic double entendres.

Double meaning is an interesting solution class. It readily accommodates alternating orientations.  

But, for these memes to be fit, they need to convey some truth about the physical and social landscape. In a competition between a meme that is simply suitable for group coherence and a meme that is suitable both for group coherence and for a life lesson, the latter should win. But one does need to investigate a 2x2 matrix of solutions.

High Group Coherence Utility
Only a possible max if Coherence inversely related to Life Utility
Maximally fit
Low Group Coherence Utility
Only a possible max if Coherence inversely related to Life Utility

Low Life Lesson Utility
High Life Lesson Utility

The only way you get a mixed solution is if Group Coherence Utility and Life Lesson Utility are interdependent AND inversely related.  A plausible inverse relationship seems hard to imagine.

Selecting For Double Entendres
This is the crux of the case. And, this is where I am going to have to put in a lot more work cleaning things up and simplifying them.  But blogs are about exploration, so here goes...

The first goal is to show how the memes under selection entangle with the group's moral system. This could happen if a meme captures an efficient vision/articulation of the Group Agent's inferred morality. If the meme is an efficient short hand for morality predictions, it should, over time, bias the way other more specific morals are framed. This should increase spandrelling. This becomes more significant as group orientations cycle.

Biasing need not be anything particularly fancy. Basically all you need is a sort of fuzzy meme representing a fuzzy articulation of Group Agent morality. Some degree of coevolution is really all you probably need.  This means the meme is a “good enough” representation of Group Agent morality.  It also means the meme is entangled with the specific moralities which emerge from Group Agent interpretation or rationalization.

Archetypical Core
The second goal is to show how a core archetype sub-meme message/moral lesson emerges. This archetype-moral message is a sub-component of the meme that is time insensitive, of high utility and reflects tensions between adjacent levels of selection/orientation.

Archetypes seem like an efficient way to share inferred Group Agent morality. They are applicable to multiple ages and multiple IQ's. They also seem to be the most mimetically robust (long lasting) class of any slightly counter intuitive. Other meme types gain virility at the cost of longevity.  So over the course of multiple generations of cycling, one would expect archetypes to survive.

But archetypes may survive at a deep level while having more virulent interpretations superimposed on them. These superpositions may be context specific solutions optimized for short and/or mid shelf lives.

Imagine there are three ways of encoding a moral message that exists in tension between two poles between which orientations oscillate. 

Core Only
One option is for the entire message to be fairly invariant. It is broad enough to capture everything. This puts it under extreme tension. Group agents certainly have broad interpretational range due to judgment aggregation intractability. Interpreters can reconcile wildly contradictory actions by inferring some imaginary larger purpose. This maintains allegiance. But it comes at the cost of predictability. You can still predict that the group agent will exert control. But you can’t predict what you should do to align. Nor can you determine which moral judgment frame to apply.

But, the simplest issue is that a single invariant message may also prove too rigid to capture the full range of oscillatory orientations it has to contend with. The generalizations required for a single solution may prove too broad. Or, the things the explanation must exclude cause it to be seen as too narrowly rigid.

At the other extreme is complete bifurcation into separate explanations for each pole. This creates issues already discussed with group agents. Are you following one agent or two? How do you predict aligning moral actions? Switching between one orientational pole and another presents way too many easy options for freeloading. We have some people acting altruistically toward the large group while others are acting selfishly toward it. This destroys norm maintenance.

The middle ground involves a core sub-meme with surface features which function as double entendres. In simpler terms, it is a lynchpin on top of a robust base (core). The core is a time invariant moral. This is likely to be a universal or, as we shall soon see, a universal metaphorical truth.  It likely frames its truth partly in terms of tension between two poles.  It explains things which are invariant to either orientation.  But it is not specific enough to detail everything.

Details are largely flushed out by a contextualized interpretation.  I.e. “Here is what this looks like when you are altruistically oriented to the large group. Here is what this looks like when you are not.”

Efficiencies of Double Entendre Archetypical Core

The third goal is to show why a double entendre with an archetypical core and lynchpin surface for contextual features is optimally fit over long time periods with intra-group and inter-group competition.

Interest & Humour
Humans are drawn to double entendres. Double entendres are both interesting and humerous.  The ability of a simple perspective switch to reveal two different meanings is a universal feature of human jokes. The ability to garner multiple levels of information from a single item means the possible time spent on that item is likely to be higher than other similar items. Humans like optical illusions for much the same way. Marketers use the cognitive engagement of lynchpin phrases for the same effect.

Continuity with Costly Investments
The lynchpin feature also removes issues of change from the group agent and puts them on the individual. You often hear people who switch religions saying it wasn’t God that changed, it was my understanding of him that did. This keeps continuity with costly investments toward a moral Big Brother while enabling thorough reformulation of those very same beliefs. It isn’t the big Brother that changed, it was my erroneous interpretation of him/her that did.

Norm Enforcement Savings
One high cost problem is norm detection and enforcement.  Instead of always confronting sub-groups sometimes it is more efficient to let them self-organize and then confront them. The problem with this strategy is that the cost of delayed confrontation may be high - they may overturn the system.

The easiest solution for this is to give these sub-groups an escape valve. But at this point you have to balance the savings from letting these sub-groups split with the costs of inter-group competition between the parent and splinter.

One way to do this is to allow the split to happen over a lynchpin idea whose alternative interpretation becomes very costly due to the paradigmical shift it imposes on the entire moral system.  This means that almost all moral lessons will have to be recreated. The spandrel like nature of the original moral system makes anything else impractical. Re-interpretation leads to errors which should lead to a weakened splinter compared to its parent. The exception of course is if the landscape occupied by the splinter is different enough from that the parent operates in so as to make the mutation have a chance at higher fitness. This mutation rate would be a population level trait not a group or individual level trait. I think the literature on species level mutation is developed enough to easily support the tail end of this line of reasoning.

Technocratic Escapes
The benefit of this final approach is that it explains how populations can escape technocratic group traps. As large groups mature, they seem to become increasingly technocratic. This gradually moves them from dealing with reality-as-it-is to reality-as-it-is-imagined. In other words, the groups theories and moral lessons don't necessarily keep up to speed with environmental changes

Low Risk Outgroup Interaction
I've already discussed how low cost hidden signaling is important for altruistic self-organization. But it seems impossible to limit this tool to altruists. Hence its function as an escape valve for anti-large group actors. In this sense, the solution space you're looking for is something that enables altruistic self-organization and which will still provide as much net benefit to the group when other sub-group oriented individuals use it.  To my mind, a double entendre lynchpin works perfect here.

Because of the value of immigration and trade, interaction with out-groups provide another potential layer for moral slogan/meme benefit.  If a meme is good for all the things discussed, a meme which is also good at outgroup interaction will outcompete it. Its utility increases with minimal extra burden.

I strongly suspect immigrant utility is a function of social proximity. (Putnam, 2007)  I further suspect social proximity is non-linearly weighted towards major moral values (group meta-truths). There are lots of little things that just don't seem to matter.  A good meme will:
  • capture things that matter,
  • make things that don’t matter seem inoffensive,
  • provide easy in-group signals for which way people are interpreting things.

Out-group interactions in historic societies are often high-risk.  Two ways of lowering the risk are to 
  • increase perceived similarity (we play by the same rules)
  • project non-threatening behaviours.  
A convenient memetic solution here is to present a moral slogan which has obvious deep moral parallels but non-threatening obvious surface differences. I'd further add that surface differences should be very easy to reject. Counter-factuals work well here.

For instance, "The flying spaghetti monster says be nice to others treating them as yourself." The golden rule is fairly universal (for in-group interaction at least...).  The flying spaghetti monster is an obvious counter-factual that is easy to reject. Which out-group is going to feel threatened or morally castigated by this? You might not share your rejection with the speaker –why risk confrontation- but you are probably evaluating them more on the central tenet of the message rather than on the surface features. Compare that to "Joseph Smith says be nice to others treating them as yourself". This message while of a similar intent is going to rile a lot of Christians. It is a sensitive mark of difference. The foundational message is the same.

Messages that have a quasi-factuality to them enable outsiders to reject them. They are minimally threatening because they can be interpreted as factually untrue whenever needed. It facilitates strawman reasoning.  Thus you can go back to your own group and laugh at how stupid the other guy was. However, the meme piggybacks on this. There is a non-zero chance that one of the friends with whom you’re sharing your ridicule actually thinks the core message is a good idea and that if the lynchpin point is interpreted another way, the whole idea may not be half bad.  This is the point that I made in the Norm Enforcement Savings section, albeit from the out-group rather than in-group lens.

Three things iterate to produce quasi-factual moral memes
  • the memetic fitness of slightly counter-intuitive ideas
  • low risk signalling
  • orientational cycling.

The co-evolution of moral memes refinement with moral articulation of a group agent produces a entangled moral belief system.  Orientational cycling selects for a meme structure that has a deep archetypical core with a double entendre lynchpin surface. Formulation of the archetypical core is entangled with the larger moral belief system.  

The lynchpin structure is mimetically fit because
  • double entendres are interesting (and humorous)
  • conserves costly investments
  • lowers the cost of norm enforcement and detection
  • provides a population-level escape from groups' technocratic traps
  • it creates a low risk formulation for out-group moral interaction.

Next I will try and see where my train of logic is off.  I'll do this by selecting a couple of major moral slogans and see where things go.  Obviously my picks are going to be biased.  Part of that bias includes selecting versions that are probably most comparable to lynchpin formulation.  For openness, here is my rough list of memes.

Meme 1 - The way minorities are treated is shameful.

Core (Archetype or Metaphorical Truth)
  • Minimal
  • There is some implicit focus on equality and discrimination, but nothing overt. This does not carry a deep metaphorical message. It simply makes one think about the way minorities are treated and moralizes on that.
  • It does not reflect any major tension between two competing world views. It does set up some tension between desired treatment and actual treatment.


  • Yes, it is shameful. They should be treated better. They are not treated the same as others. Many say as much. (progressive interpretation)
  • Yes, it is shameful, everyone should be treated the same and not given special treatment (conservative interpretation)
  • There is some quasi-factuality due to overgeneralization. Has every minority individual really been treated shamefully? This seems improbable but can be interpreted in a lynchpin way.
  • There is also some slight quasi-factuality about the word shameful.  What is meant by that term? It is not well defined and is largely a function of interpretation, especially by minority individuals.  However the lynchpin throws this for a loop depending who does the interpreting.

  • Most people would agree with this statement. There is moral pressure to do so. You would have to be pretty brave to take the negative approach to this. However, people do. And they self-organize around it.
  • Those taking the "conservative" lynchpin interpretation can participate in conversations about this topic without outright lying, but participation would have to be cursory before detection occurred. But they could vociferously support the slogan with no tells whatsoever.
  • This meme / moral slogan passes lynchpin considerations. It also passes signalling considerations. 
  • It fails on archetypical truth.  It is interesting to note, however that this meme is fairly recent (say since the 70's or 80's) and is unlikely to have experienced many refining orientational changes.

Meme 2 - There is no God but Allah and Mohamed is his Prophet

From Meme 1's analysis, I would expect this one to have more of an archetypical message (it is religious in nature and is fairly old).  I would also expect the lynchpin idea to be fairly stark. Islam has, after all, a rather... severe.. proselytizing history.

    Core (Archetype or Metaphorical Truth)
    • There is a God. God speaks to people (through prophets)
    • There are leaders who have special roles and authority
    • Yes, Allah is the only God and Mohamed is his prophet.
    • Yes, there is a God, but Allah is not it and neither is Mohamed a prophet.
    • Moderate to ideal
    • It is mostly related to over specification.  It is unlikely that Allah is the only God. Non-muslims can certainly think of other examples.
    • The very idea of Gods and prophets are classic quasi-factuals (or even a-factuals) that have an optimal level of counter-intuitiveness.

    • The main signals people can play with here are some general ideas about whether or not there is a God, whether or not the idea of prophets makes sense, and whether or not Mohamed is a specific instance of prophethood.
    • There really is no ironic escape to this creed. Rejections, as just mentioned, are fairly straightforward.

    • This meme has a fairly typical archetype. It is well developed to support human organization at most any level. Human hierarchies are natural. On its core there is no displayed tension between adjacent levels of selection (orientations). However, the religion itself is very clear on tensions between good and evil.  This creed just doesn't espouse them except through indirect reference to acceptance or rejection of divinity itself.
    • The lynchpin is start but fairly weak. It is easy for out-groups to reject any part of this claim. To out-groups it's counter-factuals are easy to reject.  
    • In terms of signalling, it enables fairly clear delineations between in-group members who accept or reject its premises. Nuance can be added by including questions of whether there can be other prophets or whether Allah has to be the only God. However these additions require a fairly big risk. There are no double entendres.

    Meme 3 - The American Dream: Everyone has opportunity to succeed by hard work, initiative and their own abilities.

    Technically the original 1931 formulation was "The American Dream is that dream of a land in which life should be better and richer and fuller for everyone, with opportunity for each according to ability or achievement."

    Core (Archetype or Metaphorical Truth)

    • America is a meritocracy.  While there will be hurdles, people can overcome them to the fullest extent possible.
    • This is a pro-social perspective that encourages industry.  It presents a tension between hurdles and personal industry.


    • Like the Islamic creed, the American Dream's lynchpins involve either full or partial rejection of any of its premises.
    • It has an additional lynchpin based upon its outcomes - if someone hasn't succeeded they you can reject them based upon their lack of ability or initiative. Either way, they are not entitled to more.  If someone has succeeded it implies they are both capable and hard working.  In essence, even though it comes across as egalitarian, it facilitates social sorting (while still keeping open the potential of class migration).  This is a fairly strong lynchpin.


    • A major quasi-factual is overgeneralization (everyone can).
    • Another quasi-factual is under-specification. Are there other things that are required for success? Probably.
    • An inferred quasi-factual is whether this dream could be true for someone with debilitating physical disabilities. Is it really true they can succeed? If so in what way? The idea is Utopian and comes across as clearly non-falsifiable. This is quasi-factual.
    • There is an ironic sense of both eminent fairness and stifling sorting/judgment. 
    • There are no obvious a-factuals to reject.


    • The main signals come with whether or not people accept the degree of generalization presented.
    • Additional signals can come by inserting some irony such as "and by being born rich".


    • The meme presents a utopian archetype. It offers incentives for personal motivation. Progress (success) is possible no matter your station. Effort & talent development matter.
    • There isn 't a significant direct lynchpin. Lynchpins come from refining degree or adding some ironic elements. However, secondary lynchpins are fairly pronounced. The meme gives people the chance to judge others as either skilled/industrious or stupid/lazy.  It supports most any pre-conception - after all, how do you define success for any individuals natural skill and abilities?
    • The quasi-factuals are fairly strong. The whole idea is clearly utopian.

    In practice meme analysis according to the elements presented is clearly very loose. This is extremely problematic. It makes falsification practically impossible.

    The most problematic element was the lynchpin.  Two out of three memes had a noticeable lynchpin. One emerged directly in the meme itself. The other emerged in the application of the meme as a judgment tool.  The other meme's lynchpin only emerged in degree of acceptance or rejection. It is however, interesting to note that the newest most unrefined meme had the most pronounced lynchpin, while the oldest most refined one had the weakest most stark lynchpin.

    It would seem possible that the lynchpin idea is quite wrong. However, there is a chance that new moral memes may need more of a lynchpin to survive while older ones which are more entangled in moral belief systems can function as starker delineators with higher rejection costs.

    Quasi-factuals mainly involve over-generalizations or over-specifications.

    The coding of metaphorical truth also seems to be fairly weak. Additionally it did not directly represent tension between two competing orientational poles.

    No comments:

    Post a Comment