APSA 2024 - Ethics, Law, and Civilian Casualties

Discussant Comments

Aidan Milliff

Dill, Howlett, Müller-Crepon, and Revkin

Paper Summary

Argument

  • “Fated traits” strongly shape attitudes about collaboration/reintegration in Ukraine
  • There are strong “conditional” effects by ethnicity

Empirical Strategy

  • Nationally-rep. conjoint in Ukraine (wow), with vignettes about collaboration and remorse/punishment/apology

What’s Exciting

  • Obvious importance and strong connection to real empirical setting
  • Interactions are fascinating, the fact that identity makes collaboration behavior a “one way gate” is really insightful

Suggestions/Questions

  • A lot going on in here

  • Fated traits and choices aren’t “competing” explanations, conceptually, are they? Lots more possibility in the “interactions” section. Stereotype?

  • Conjoint choice combinations have some funny outcomes, which I think explains some oddity in Figure 4 (slide 12)

    • Area controlled by Ukraine; reluctantly decided to; fight alongside Russian forces
    • Area controlled by Russia; reluctantly decided to; fight alongside Russian forces

  • What do you gain from “appropriate” level of punishment component?

  • Migration lurks below the surface of all these choices; may not be “symmetrical”

Dill, Myers, and Schubiger

Paper Summary

Argument

  • Consent form the intended beneficiaries of a military intervention is an important determinant of public support for intervention (good!)

  • This mostly works at the “collective” level, not the individual level

Empirical Strategy

  • large representative online survey of Americans
  • Forced choice conjoint

What’s Exciting

  • Empirical philosophy! It’s really cool to see a paper focused on understanding the moral reasoning that shapes public attitudes

  • Tells us something about a justification we hear a lot.

Suggestions/Questions

  • A LOT going on in here

  • Help a simple empiricist (me) out w/r/t aggregation. Examples from a conflict?

  • The “individual” vignette read as quite odd. What might be the implications of that? Do you have any debrief?

  • Fig 2 (slide 18); seems like support from “less than half” is indistinguishable from “support from all”

  • Conjoints force attention to the variables you chose; Consent “loses out” to other factors – bystander and beneficiary harm AMCEs are sometimes larger (Fig 5, slide 20)

Voytas and Lyall

Paper Summary

Argument

  • Citizens in the US and Pakistan largely support/expect solatia payments … and over-estimate how often they are made

  • Support for these payments is remarkably un-conditional

Empirical Strategy

  • Parallel online experiments with relatively high SES samples in US and Pakistan
  • Conjoint!

What’s Exciting

  • This is a very meaningful descriptive point, even if you read as (mostly) un-conditional

  • Observing the “omitted variable” tells us some unsettling things about expectations of collateral damage in US

Suggestions/Questions

  • Give the descriptive results much more spotlight

  • Sample is a little funky, but in a way you can make an advantage

  • One of the only sig. findings in study 1 feels a little mechanical - apology is part of the outcome index as well as the treatment

  • Study 3 could be part of a conclusion section; seems like its answering a different question

  • For future: Casualties are not the only harm civilians experience from drone strikes

Lushenko and Carter

Paper Summary

Argument

  • Chaplains assess the legitimacy of drone strikes depending strongly on “battlefield” vs “OTH” setting

  • Many Chaplains who consider strikes “legitimate” still don’t approve of them (!?s)

Empirical Strategy

  • Survey of US Army Chaplains
  • Not a conjoint ;)

What’s Exciting

  • An important and cool population

  • Interesting to get a sense of the moral attitudes toward drones; often a lawyer-centric discussion

Suggestions/Questions

  • Don’t lead with the arguments about why chaplains don’t matter!

  • Why are we not primarily comparing arm vs arm? What is the pure control and why do we care about it?

  • Some comparisons almost definitely underpowered; just consider that when deciding what to focus on

  • Microfoundations - are attitudes causes of attitudes, and what’s going on in the sentiment analysis?

  • Please tell us more about sources of the legitimacy – support gap

Thanks! Great Papers!

And I really do mean it.