Wednesday, April 12, 2017

Why Creedal Statements Annoy Me

I'll admit - I have a bias against creedal statements - especially moralized ones coming from academic departments.

Here is the main reason why.

The first is a classic case of what a lot of people feared from the March for (intersectional) Science - the community would be heavily divided, bifurcating along creedal lines.  One camp values the pursuit of cold objectivity, the other values embracing intersectionality and "equitable" outcomes.

(By the way Tara Smith is an associate professor of infectious diseases and so is certainly aware of many scientific process debates.  She was also up on Science Blogs.)

Like most things, there's value to both teloses.  I'd suggest the intersectional camp facilitates religious like dynamics more than the objective camp, but as the New Atheists showed, quasi-religious dynamics never go away.  Once groups coalesce, the chances of them stumbling into adaptive group resonances and then into quasi-religious dynamics increase.

What's the danger?  "We don't debate hate speech and hate facts!"  It's the inevitable polarization that (stochastically) occurs once you remove the boundaries against decohering polarization.  Only certain experts are allowed to moralize.

Because morality is usually dogmatic, debates tend to utilitarianism.  However, the utilitarianism is tribalized.  It is not what is "good" for all (at least directly), it is what is primarily good for identifiable groups which is secondarily assumed to be good for all.  Identity politics rears its cancerous head, and while I may be naive, it seems that marginalization of "privilege" based on identifiable characteristics is inevitable.  This sets the stage for severe group competition.

While acquiescence is certainly possible, it is very unlikely except in certain exploratory groups who are playing the odds that loss of individual fitness will be offset by gains emerging from group-oriented altruism.  But the only way to protect the group from cancerous within-group competition via identification (oppression olympics) ad infinitum is to have some very strong boundary norms about where divineness stops.  Hence the value of clear confessional statements and strong in-group out-group gradients (via norms, costly commitment displays, shared rituals, behavioural/dress demarkers, etc.).

So why do I hate creedal statements in academia?  Because I am very tempted to support the Shermer approach to science, sign what ever creedal statement they eventually come up with and start sticking it to all those medieval retrogrades!  You see, if we position it so that intersectionality is a sin as egregious as modern racism, because, well it is classically racist, and our side is sure these acts will decohere society as much as any well-intentioned political communism ever could, then I'd be fully justified.

So why not just let the issues of intersectionality stay under the cover and ignore them?  For me, ignoring problems is just stupid.  But formalizing them in a way that invites their resonance (by some) into weaponized inter-group competition weapons is equally silly.  

Listen to voices and deal with issues at a personal rather than group level.  Group dynamics, especially around moral issues are very different from what happens at the individual level.

I fear Western society will re-learn this "fact" as society continues to bifurcate between the quasi-religious political left (ctrl-left) and the far / alt-right.

Monday, April 10, 2017

Creedal Confessions in Academia

Creedal religious confessions seem to be a structural feature of universalizing religions.  This might range from fairly basic "there is no god but Allah and Mohammed is his prophet" to technical Nicene-like statements. 

Recently a statement of confession and commitment was produced by Christian faculty members from across the United States.  In the tribalized trajectory of Western society, what might this portend and what light does cultural-evolution theory shed on the general dynamics of groups evolving to formal-creedal practice?

Tribal religions "just are".  In highly implicit cultures where religion, group membership and culture have no boundaries, technically precise definitions of minimal belief standards are non-sensical.  Beliefs aren’t aren’t designed to make sense to outsiders and they certainly aren’t optimized for proselytization. 

However, religion and governance (Norezayan, 2013) have co-evolved.  Some multi-level selection researchers view this as a series punctuated co-evolutionary steps (Turchin, Norezayan, Atran).  Basically human groups bounce around an evolution-transition-landscape* which due to between-group competition selects for highly cohesive highly coordinated small groups or less continuously cohesive less continuously coordinated large groups.

Under some conditions large groups can outcompete smaller more efficient groups.  In other conditions they can’t.  Tensions between within-group selection and between-group selection makes the whole scenario rather complex (in the technical sense of the word).  But, evidence strongly suggest a directional arrow to the products of selection*: Increasingly larger groups facilitated by better conflict minimization tools, better coordinating tools and sufficient within-group dependencies.

The emergence and stabilization of universalizing religion as a higher-level of selection was, in part facilitated by overt belief structures and easier in-group immigration paths.

A small-group "coherence" approach minimizes within-group freeloading at the expense of size.  A large-group "size" approach threads the omnipresent freeloading needle by making essential norms very overt and by leveraging various natural tendencies for group-to-individual biasing.  Obviously religion and quasi-religious dynamics are very effective in this regard.  Atran's “In God's We Trust” provides a pretty good base in this regard.  Norezayan's “Big Gods” makes a compelling evolutionary argument why these dynamics are selected for (albeit only making tangential use of formal multi-level selection tools).

The Big Question
This multi-level lens leads to some interesting speculation about the dynamical landscape of formal belief (creedal) statements within the intersection of academia and todays increasingly tribablized quasi-religious political environment.

Speculation will avoid value-based analyses.  While I favour the conflict minization that comes with higher levels of selection, I’m also pragmatic enough to realize some structures just aren’t yet stable enough (selection amongst component actors hasn’t stabilized the conditions required for an evolutionary “transition”).  Unfortunately this situates conversation in the world of utilitarianism.  As a na├»ve non-philosopher, I see not way around this.  So, this argument is obviously subject to all the faults and critiques inherit with utilitarianism.  Alternative suggestions are welcome.

I’ll talk about the dynamics of creeds in religious-secular intersections by situating the formalization of belief in a multi-level selection context.  This will be done by showing how creedal statements function as a higher-level of selection for an individual and a lower-level of selection for a nation-state. Next, I’ll frame this as a classic example of tension between adjacent levels of selection.  This brings in complex oscillatory dynamics.  When expressed in a sufficiently large population we should therefore see multiple (drift) paths expressed.  Gene-culture co-evolution suggests which paths are (stochastic) wells. The cultural phenotypes of these wells will be explored and compared to some recent examples of quasi-religious secular social movements who have also tended to semi-formal creedal-like confessions.  I’ll then speculate, using cultural evolutionary insights, what difference the explicitness of formalized creeds might make.  Finally, I’ll remind you that just because formalization may facilitate between-group competition does not mean that creeds are either good or bad.  Rather it means that good intentions can have bad outcomes and bad intentions can have good outcomes.  I’ll tackle this by coming back to the issue of time scales and units of analysis.

Creeds & Levels of Selection
The academic “statement of confession and commitment” presents an interesting case study in levels of selection.  It brings together a number of Christian faculty from across the country.  Its pretty likely that signatories will “have each other’s backs”.  In other words, the signing of this confession increases the probability of mutual support, especially as related to attacks against the morality espoused therein, and the ability to espouse such morality within academic settings.  While some signatories may view this creed as simply a “good-enough” statement of their own values, the group nature of this statement makes omission of  a functional analysis naive. 

Functionally, why a joint statement?  Why a statement at all?  A levels of selection analysis looks at these functional question in terms of whether it brings people together as a group or separates them into smaller groups. 

This question is not simple.  Human acts are nested in multiple strong and weak overlapping group hierarchies.  For instance, what is the functional effect of this type of creed on a University as a whole? On a pluralistic political society? On a nation torn by questions of open or closed borders? On a lone individual in an isolated university department?

In this regard delving into a deconstruction of the creed itself may just muddy the waters.  Some might find the creed unifying.  Some may find it divisive.  Do we arbitrate who is right?

My particular approach is to simply reference classic studies, especially those in social identification theory, which suggest identity politics – well intentioned or not, tends to lead to clearly identifiable groups which then (probabilistically of course) increase the likelihood of between-group competition. 

But, in this regard trajectory is key.  Identity politics may facilitate a move up a level of selection; individuals coalesce as small groups which can then coalesce as groups of small groups (medium groups), and so on…  But identity politics may also facilitate tribalization.  Large group cohesion is replaced by a focus on smaller group ties.  Instead of nationalism, we get various levels of tribalism.  Does nationalism disappear?  Of course not.  Rather the tension between adjacent levels of selection is tweaked.

Which way do things go? In most cases, we just don’t know.  Speculation exceeds error bounds by quite a bit.  We do know that as the fitness between adjacent levels of selection become equivalent the degree of complexity in cycling between adjacent levels increases commensurately (Okasha, 2006).  But, not knowing which way things will go doesn’t mean we’re blind.

For instance, a recent Nature co-authored by Nowak (who has done a lot of simulation work in the field of evolutionary altruism) suggests pairwise bonds facilitate within-group altruism.  Too many connections limit the conditions necessary for reciprocal altruism.  Too few connections and altruism can’t spread.  Cordes and Richerson (among lots of others) have also done good technical work in this field.  Unfortunately in a world with multiple level of selections, “pairwise” is always relative…  is it between individuals?  Between small groups?  Medium groups, etc.  The math is agnostic on scale.

A random walk approach (aka. the default drift hypothesis previously mentioned) is quite useful in large populations.  Most every possible walk eventually gets expressed.  What then matters then is comparing potential paths to natural evolutionary wells.  In other words, what proximate causes influence (stochastically) the relative cultural fitness of different paths?  This doesn’t imply what will happen.  It simply informs what is likely to happen if the population is large and our gene-culture co-evolution information is accurate.

Gene-Culture Co-evolution
For a primer on gene-co-evolution basics watch a classic presentation on lactase persistence or look at the classic Cavali-Sforza book or the Boyd Richerson version or any of the modern stuff by Gintis, Mesoudi, Henrich and others.  Darwin’s Cathedral is a good intro as well. Although it doesn’t get into the technicalities of dual inheritance much, it’s a classic work on multi-level selection insights into religion, quasi-religion and adaptive group resonances.

Some of the most salient applications can be summarized by combining D.S. Wilson’s and Scott Atran’s works.  Adaptive groups resonate around morality.  As List & Pettit (2011) show, quasi-real group agents emerge in non-deterministic judgment aggregation problems via a process of inferred moral ascription.  Both Wilson and Atran suggest hypsentitive agent detection heuristics do this, as does a strong fitness advantage for being able to predict other people’s actions (in a way where false positive are much less problematic than false negatives).

Once groups start getting moral, the tensions between within-group selection and between-group selection (i.e. freeloading vs. altruism) are well handled by religious / quasi-religious processes.  Atran’s “In Gods We Trust” is excellent here.  Religion is fit for a good reason.  Sacrilized politics is similarly fit for the same reasons.  The different degree of embodiment of “moral Big Brothers” and relative levels of supernaturalism, is, in my mind, of secondary importance.  Human dynamics are similar at the first order.

As per Atran, processes which enhance group adaptiveness include:

  • Shared rituals
  • Moral Big Brothers
  • Slightly counter-intuitive (and hence highly memorable) memes
  • Costly commitment displays
  • Norm enforcement
  • Hard to fake beliefs & actions

At extreme levels of morality identify fusion processes come into play.  However, it seems like these are mainly expressed during periods of severe group competition or as “loss leaders” amongst population tails who facilitate group expansion via slightly counter-intuitive costly commitment displays (think of the first Christian’s seemingly maladaptive martyr fixations) or “don’t mess with us” dynamics.

But enough theorizing, what insights does this give us with regard to creedal dynamics?

This might be best seen by looking at quasi-religious secular social movements who have also tended to semi-formal creedal-like confessions.  Here are some obviously non-random examples.

We are intentional about amplifying the particular experience of state and gendered violence that Black queer, trans, gender nonconforming, women and intersex people face… we know that patriarchy, exploitative capitalism, militarism, and white supremacy know no borders. We stand in solidarity with our international family against the ravages of global capitalism and anti-Black racism, human-made climate change, war, and exploitation

No formal creedal statements are readily available.  Morality is situated around attacking/resisting authoritarianism especially as represented by neo-nazi’s and far right politics.  There are obvious intersections between anti-patriarchy, anti-capitalism, and minority empowerment.  Allowing the normalization of authoritarianism and repression are clearly seen as sins.

Donald Trump is building a broad coalition as he makes an historic run for the White House. We were among the first to recognize the LGBT community had a place in his campaign. As Americans from all walks of life listen to Trump's message of economic nationalism, American exceptionalism, and limited government, those of us in the LGBT community should start paying close attention. Now that America has entered a post-Marriage Equality era, it is time for the LGBT community to stop viewing politics through the narrow lens of the culture war and start engaging the whole political spectrum. For too long LGBT folks were told they had to be Democrats to be for equality. Well, those times have changed and it is time for us to unshackle ourselves from the ideology of the past and embrace the ideas of the future.

Inclusion, diversity, equity, and accessibility are central to the mission and principles of the March for Science. Scientists and people who care about science are an intersectional group, embodying a diverse range of races, sexual orientations, gender identities, abilities, religions, ages, socioeconomic and immigration statuses. We, the march organizers, represent and stand in solidarity with historically underrepresented scientists and science advocates. We are united by our passion to pursue and share knowledge.

We acknowledge that society and scientific institutions often fail to include and value the contributions of scientists from underrepresented groups. Systems of privilege influence who becomes a part of the science community, what topics we study, and how we apply our work in creating new technologies and crafting policy. We recognize that, historically and today, some scientific endeavors have been used to harm and oppress marginalized communities. Political actions -- such as gag orders for government science agencies, funding freezes, immigration bans, and policy changes blocking action on climate change -- lead to greater damage for vulnerable populations. Science itself can drive our conversations about the importance of diversity, as it provides us with the data to understand how systemic bias and discrimination impact our communities and how best to change it

Given that, since 1948, hundreds of UN resolutions have condemned Israel's colonial and discriminatory policies as illegal and called for immediate, adequate and effective remedies; and

Given that all forms of international intervention and peace-making have until now failed to convince or force Israel to comply with humanitarian law, to respect fundamental human rights and to end its occupation and oppression of the people of Palestine; and

In view of the fact that people of conscience in the international community have historically shouldered the moral responsibility to fight injustice, as exemplified in the struggle to abolish apartheid in South Africa through diverse forms of boycott, divestment and sanctions; and Inspired by the struggle of South Africans against apartheid and in the spirit of international solidarity, moral consistency and resistance to injustice and oppression;

We, representatives of Palestinian civil society, call upon international civil society organizations and people of conscience all over the world to impose broad boycotts and implement divestment initiatives against Israel similar to those applied to South Africa in the apartheid era. We appeal to you to pressure your respective states to impose embargoes and sanctions against Israel. We also invite conscientious Israelis to support this Call, for the sake of justice and genuine peace.

But how is any of this different from the vision statement of business and other institutions?  The evolution of intersectionality provides one interesting data point. From McKibbin et. al (2015)

Carbin and Edenheim argue that intersectionality has shifted from being a metaphor grounded in structuralist ontology to being an overarching feminist theory which makes explicit an ontology of neither the subject nor power. Intersectionality is no longer defined as a metaphor for the way in which intersecting systems of oppression impact on women’s subjectivities, but is referred to in the literature variously as a methodology, a tool for data analysis, a nodal point in feminist theory, a feminist project or platform, and a framework for social policy development.

But is this academic parsing of categories or belief-based tribablization?  I would argue it can be all of these, with the key being the nodal aspect of bifurcations.  This gets back to the value of a random walk approach where various paths are analyzed according to likely gene-culture wells.

I would also argue that for some, perhaps many groups, intersectionality has become a creedal like confession.  For example the distinction between second wave and third wave feminism revolves around intersectionality and the targeting of resources/concern.  A case in point is whether Beyonce’s sexualization can be considered “feminist”. 

White feminism is a set of beliefs that allows for the exclusion of issues that specifically affect women of color. It is ‘one size-fits all’ feminism, where middle class White women are the mold that others must fit. It is a method of practicing feminism, not an indictment of every individual White feminist, everywhere, always.”

It is usually not that overt, and most White feminists would deny that this is what’s being said or done, but you notice it in more subtle comments like “Why do you have to divide us by bringing up race?” or “Are trans women really women? There should be a distinction.”

In the face of calls for a more intersectional feminism, there are even White feminists who claim the whole concept of intersectionality is just academic jargon that doesn’t connect with the real world.
Yet the irony seems lost on some feminists who make these claims while staunchly opposing the language of “humanism” in place of “feminism.”

Simply put, it’s not those who are calling for a feminism that is responsive to the specific issues they face that are being divisive. It’s those of us who refuse to acknowledge the need for an intersectional ethic in feminism.

A certain type of belief is required for membership in certain types of groups. This belief is moral.  In business is making money a morality?  Perhaps for some wall-streeters.  But for most employees it is a transactional good.  Sacred values attachment is minimal. This is what distinguishes a business value statement that may or may not be moral and may or may not be morally interpreted from moralized creedal statements like the Christian faculty’s Statement of Confession and Commitment.   In-group actions are communicated as moral and out-group actions as immoral (implied or overt). 

The Difference of Explicitness
There’s a bit of complication in the fact that many of the cited creed-like statements don’t come right out and call those who don’t ascribe to their central tenets sinners.  For example in the Statement of Confession and Commitment there is little doubt that any Trump supporting or closed border supporter is a sinner someway.  But they don’t come right out and say this.  Other cited movements do (if you don’t support intersectionality and acknowledge your privilege you’re a racist).  I think my brother reference Derrida’s “Force of Law” work for his understanding of this general issue.

What we have is a bifurcation point.  Formalized belief statements don’t cause the viewing of out-groups as sinners.  Such direct causation is farcical.  But the trend to formalized belief statements with explicit morality in zones where few hitherto existed does change the cultural-evolutionary landscape.

The main cultural tools you use in this regard are:

  • horizontal transfer
  • oblique transfer
  • prestige bias
  • content bias
  • conformity

Each has a different growth curve, which is unfortunately beyond my time to analyze with respect to the evolution of formalized-belief-statement based groups.  

Explicitness enables high fidelity copying.  Memetic fitness is optimized with a sprinkling of slightly counter-intuitive beliefs.  This also resonates with the characteristics associated with adaptive group formation (costly commitment displays, hard-to-fake beliefs/actions, norm detection & policing).  This gives us two things to look for: 1) what cultural transmission resonates with high fidelity copying and 2) what beliefs statements are memetically fit in their formalized state?

According to Mesoudi’s interpretation of Boyd & Richerson,

Guided variation is where people individually modify acquired cultural traits according to their own individual learning biases.  Content biases, like other forms of cultural selection, occur when people preferentially choose among existing traits found in the population without changing those traits.  Guided variation is an individual process, content bias is a population process

So if the interpretational range of formalized belief statements is loose and it is used to move up levels of selection, then guided variation is a possibility and log-like expansion should be expected.  But the fidelity of tightly written belief statements suggests a content bias.  The question then becomes about conformity levels.

Group-to individual feedback mechanisms are examples of conformity biases.  Group-to-individual feedback in morally impregnated landscapes is well studied and is significant.  One you become part of a group and make a commitment as simply as signing a statement, the probability of expressing group-to-individual feedback increases significantly.

Prestige is also a powerful conforming bias.  Prestige bias involves mimicking observed behaviours of successful individuals (i.e. buying the same shoes as Michael Jordan). As per Mesoudi (pp. 74-75)

Boyd and Richerson constructed models to explore this intuition more formally.  They confirmed that a general prestige bias is indeed a good way of acquiring adaptive behavior compared to individual learning and unbiased transmission (random copying).  However, this depends on the extent to which indicator of success correlate with the traits that are copied.

Other experiments support Boyd and Richerson’s specific prediction that prestige bias is broad and not necessarily always adaptive.

Prestige bias can also lead to a runaway “arms race” between the markers of prestige and the copied traits.  To illustrate this, Boyd and Richerson drew an analogy with sexual selection in biological evolution.

Perhaps one of the most informative articles is The Herding Behaviour in Heterogeneous Populations.  Here’s the abstract.

Here we study the emergence of spontaneous leadership in large populations. In standard models of opinion dynamics, herding behavior is only obeyed at the local scale due to the interaction of single agents with their neighbors; while at the global scale, such models are governed by purely diffusive processes. Surprisingly, in this paper we show that the combination of a strong separation of time scales within the population and a hierarchical organization of the influences of some agents on the others induces a phase transition between a purely diffusive phase, as in the standard case, and a herding phase where a fraction of the agents self-organize and lead the global opinion of the whole population

Multi-agent models often describe populations segregated either in the physical space, i.e. subdivided in metapopulations, or in the ecology of opinions, i.e. partitioned in echo chambers. Here we show how both kinds of segregation can emerge from the interplay between homophily and social influence in a simple model of mobile agents endowed with a continuous opinion variable. In the model, physical proximity determines a progressive convergence of opinions but differing opinions result in agents moving away from each others. This feedback between mobility and social dynamics determines the onset of a stable dynamical metapopulation scenario where physically separated groups of like-minded individuals interact with each other through the exchange of agents. The further introduction of confirmation bias in social interactions, defined as the tendency of an individual to favor opinions that match his own, leads to the emergence of echo chambers where different opinions coexist also within the same group.
So formalization sews the seeds of hierarchy (the original draftees are high status, as are early adopters and high commitment expressers).  It also facilitates content bias plus conformity growth curves (people can pick which creed to sign, but once they do they are biased by in-group dynamics).  These growth curves clearly have the dynamics associated with cultural phase changes.

Formalized belief statements are subject to gene-culture co-evolutionary forces.  Cultural selection resonates with genetic dispositions.  In large populations random walk drifts ensure all cultural pathways are expressed.  Those that are most fit (genetically and culturally) are of the most interest.  Groups which express adaptive group traits are selected for under conditions of weak and strong between-group selection.  It is hard not to see current political and social turmoil as anything other than a case of mounting group selection pressure.

Academic belief statements with strong moral components have a probabilistic chance of resonating with individuals who will leverage them adaptively.  Explicit moral statements facilitate this process.  Such statements enable in-group out group gradients.  Are they simple statement of belief?  Yes, and no.  As soon as it becomes a group enterprise group dynamics ensue.

Between Group Competition
Coming in another part***….


* Remember, evolutionary transitions require: 
1  in-group conflict suppression mechanism  - belief formalization enables this by delineating belief lines enabling clear distinctions between in-group and out-groups. (and no, it doesn't just minimize in-group conflict by defining away anyone in conflict as an out-group....)
2  coordination - the morality (implied or overt) of belief statements aids this
3  extreme dependency - belief statement facilitate this only in as much as they facilitate the creation of adaptive groups (other signatories have your back).  But extreme dependency is very unlikely except in momentary times of severe selection pressure (such as violent schisms).

** Note: a directional arrow does not imply a Fukuyama “end to history” nor any kind of directed process.  It especially does not imply a stage-based theory to societal evolution.  The default hypothesis for any evolutionary process is drift.  In some environments smaller groups are more fit.  In other environments they are not.  What you see is expression of both phenotypical tails.  Migration (tied to within-group competition) and between-group competition play important roles in selecting whether or not the tails are or are not cut off.  Tribal sized groups who are unable to coordinate with others as needed are certainly much less dominant than the were 10,000 years ago.  So the default hypothesis is drift with selection pressure against uncoordinatable small groups.   A number of authors take a stronger stance and suggest that war is in fact an active selector for large groups.  While I personally favour the active selection stance, for this argument the weaker drift hypothesis suffices.

*** This particular instance of group competition has some tangential connection to social media swarm politics I talked about last year (not that it is causative in a one-to-one way, but rather that it facilitates a landscape which produces such dynamics under the right conditions)