Friday, July 5, 2019

US Civil War - Picking a Model

In the first post in this series I laid out a couple of scenarios I see as plausible for inflection into a US civil war.

Part 1
Part 2
Part 3

Scenario 1

A second Trump term from 2020-2024  causes increasing low level left-wing extremist violence rationalized by hyperbolic "Trump will never give up the presidency" hysteria which then justifies a "resist at any cost" ethos. Violence levels should get going in 2021 -2024 via gradual ramp-ups.

I view this type of conflict as likely, but of insufficient magnitude for the "civil war" term.

Scenario 2: Revolt against Democratic Authoritarianism

A second Trump term is followed by a reactionary 2024-2028 Democratic president who pushes hard on authoritarian solutions such as gun control, border dismantlement and abortion defense. Censorship and quasi-criminalization of "hate" (political) speech triggers right wing extremists. Violence emerges as the right starts protests and starts becoming overtly violent against "counter-protesters". This may be justified as a reactionary response to perceived existential threats against most types of conservative public sphere participation.

I view this conflict vector as moderately likely. It should peak around 2026. Total violence levels should be moderate (probably similar in magnitude to 1960's / 1970's violence levels, albeit with more spatially concentrated outbreaks). The dynamics however, are more civil war like than  1960's/1970's political violence. I think this is the lower-bound Turchin envisions with his Structural Demographic intra-elite competition theory.

Scenario 3: Politicide by a Reactionary Right

After a 2020-2024 Trump presidency, things simmer during a Democratic presidency of 2024-2028, but things don't explode. In 2028 an oscillation back to an even more reactionary (populist/authoritarian) Republican president leads to significant radical left-wing violence which infects decent portions of what has hitherto been the center-left. There is a strong over-correction from the authoritarian right who justifies extra-legality as necessary and tacitly unenforceable by foes and an all-but gone political center.

I view this type of conflict as the most likely.  It is likely to have enough system energy to spiral into  "civil war" like dynamics. Violence occurs in 2028-2029, largely emerging from radical left wingers, but egged on and resonated by emboldened and frustrated right-wing extremists.


Only Scenario 2 and 3 warrant civil war analysis.  Both vectors seem to resemble ethnic based civil war conflicts (where ethnicity is interpreted broadly and includes identity fused politico-moral ideologies with strong adaptive group dynamics).

Scenario 2 Analysis

Revolt against Democratic Authoritarianism

Scenario 2 seems to resemble the gradual violence ramp ups Sambanis sees characterized in Nigeria. Democratic rule of law has a hard time cracking down on instigators due to its inherent penchant for equality under-the-law. An inevitable spiral into one-dimensional identity based criminality will worsen rather than weaken this basin trap. I think Tommy Robinson's recent conviction for contempt of court, based largely on his identity and its "proof" of intent, demonstrate this attractor basin.

This type of civil war will therefore largely involve confrontation over messaging mediums. This is akin to the low level civil war which preceded US independence. Journalists, outspoken B-celebrities like Kaepernick, Alyssa Milano, Steven Crowder, Candace Owens, Ana Navarro, etc. will be the main assassination targets. Social media offices may get a few rage-based suiciders.

Open conflict will largely be limited to protests. This fits in with a Sambanis riot vector. Government officials will usually be one step behind protest violence. This may be purposeful (e.g. Portland) or just due to incompetence and a few nefarious oversights (e.g. Charlotte). Families and communities will polarize. Commerce will polarize and identity will increasingly be a key to public sphere participation (e.g. Raverly, Master Card & Bank of America).

Dynamics are  "ethnic" in nature, but are more akin to Sambanis' anti-colonial or nationalistic strife vector and a polticide human dynamic.

Pluralism returns when:
  • people get sick of the limitations this causes on commerce and public sphere participation
  • the US settles on a mid-sized polity solution where a new balance of power is maintained via semi-formal group of states alliances, somewhat similar to pre-civil war North-South dynamics.
I would guess pluralism would return somewhere around mid 2030's

Scenario 3 Analysis

Politicide by a Reactionary Right

Scenario 3 violence inflection is around 2028-2029, and emerges from a violent reactionary right. The vector is less a series of riot based ramp ups and more a concerted nationalistic strife. Fighting is largely about "if you don't like this country - get out" and "fascist resistance". Because I suspect this would involve fewer years of ramp up, I suspect it will either be a momentary blip of violence, or a more serious political purge using the force of law backing up mob-based political oppression.

Outcomes are much more chaotic (in the technical sense of the word). But, if things do explode, they're much more likely to involve governmental force. Say, for instance, forcing California to capitulate its "illegal" sanctuary cities or potential border wall tear downs.

I would guess the ramp up and ramp down from this "civil war" would be much quicker and much more damaging to democracy as we know it.


Cederman, L. E., Wimmer, A., & Min, B. (2010). Why do ethnic groups rebel? New data and analysis. World Politics62(1), 87-119.

Lyall, J., & Wilson, I. (2009). Rage against the machines: Explaining outcomes in counterinsurgency wars. International Organization63(1), 67-106.

Lyall, J. (2009). Does indiscriminate violence incite insurgent attacks? Evidence from Chechnya. Journal of Conflict Resolution53(3), 331-362.

Sambanis, N. (2004). Using case studies to expand economic models of civil war. Perspectives on Politics2(2), 259-279. 

No comments:

Post a Comment