Popular in Political Science

American Politics

Keeping the Republic

Boston College Marc Landy

British Politics

Entrepreneurial thinking: the politics and practice of policy impact

University of Sheffield Matthew Flinders

Canadian Politics

Parties and Elections: An End to Canadian Exceptionalism?

St. Francis Xavier University James Bickerton

Asian Politics

The Politics of Knowledge in Development

University of Delhi Madhulika Banerjee

American Politics

Fifty years of declining confidence and increasing polarization in trust

University of California Berkeley Henry Brady

American Politics

The Opening Dilemma

Cornell University Richard Bensel

Political Economy

Wait and See?

University of Liverpool Mariaelisa Epifanio

Soldiers’ Dilemma: Foreign Military Training and Liberal Norm Conflict

View more insights in Social Sciences - Political Science - International Relations - Security Studies

View more figures from Georgetown University

View the page of Renanah Joyce

Read the Study

How do militaries respond when they confront the dilemma of conflict between the liberal norms of respect for human rights and civilian control of the military? Renanah Joyce discusses the effects of norm conflict on military attitudes among the Armed Forces of Liberia

Image courtesy of interviewee. July 8, 2024

Generating Workspace

Export as PDF

The study discusses how militaries navigate conflicts between two key liberal norms: respect for human rights and obedience to civilian authority. The presenter refers to this tension as "liberal norm conflict." When political leaders command military actions that suppress human rights, soldiers face a dilemma on which norm to prioritize.

Using the 1979 Rice riots in Liberia as a historical case, where the military was ordered to fire on protestors, the study argues that militaries often default to a third norm of cohesion—essential for collective military functioning—over the two liberal norms during crises. This video phenomenon is particularly pronounced in weakened democracies where the military confronts conflicting expectations.

The United States, a major provider of foreign military training, places significant emphasis on instilling respect for human rights and civilian authority during training programs. The study analyzes how such training influences military decision-making in situations of norm conflict. The findings suggest that, contrary to expectations, militaries under stress are more likely to prioritize cohesion rather than uphold human rights, especially when U.S. training is involved.

Key results from a survey of 270 Liberian military personnel indicate that those exposed to norm conflict showed less support for prioritizing human rights and more for cohesion. The research highlights the impact of training variations, suggesting that U.S.-trained soldiers were less likely to support liberal norms in conflict situations.

The study concludes that while training is intended to promote liberal norms, it may inadvertently encourage outcomes that favor cohesion, potentially leading to norm violations. Future recommendations include exploring how to decouple training for military capabilities from norms and strengthening institutional frameworks, like military justice systems, to manage behaviors during crises effectively.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.145 --> 00:02.845
The study that I'm talking about today explores

00:02.905 --> 00:05.525
how militaries respond to the problem of

00:05.525 --> 00:08.925
what I call liberal norm conflict between the norms

00:08.925 --> 00:10.285
of respect for human rights

00:10.505 --> 00:12.285
and obedience to civilian authority.

00:12.505 --> 00:14.725
And by norms, I mean, shared understandings

00:14.725 --> 00:18.525
of appropriate behavior and and respect for human rights

00:18.745 --> 00:20.925
and civilian authority are arguably two

00:20.925 --> 00:22.365
of the most important norms

00:22.365 --> 00:26.245
that govern liberal civil military relations between states,

00:26.435 --> 00:28.085
societies, and their militaries.

00:28.505 --> 00:30.885
But when political leaders order the military

00:30.945 --> 00:33.725
to repress human rights, it activates a conflict

00:33.725 --> 00:35.045
between these two norms.

00:35.665 --> 00:38.365
And so to illustrate what I mean by liberal norm conflict,

00:38.385 --> 00:40.165
let me take you back for a moment

00:40.265 --> 00:44.165
to the 1979 Rice riots in Monrovia, Liberia.

00:44.465 --> 00:47.565
So in 1979, the government hiked to the price of rice

00:48.215 --> 00:51.165
riots broke out, and the president personally ordered

00:51.445 --> 00:52.965
soldiers to fire on protestors.

00:53.305 --> 00:55.605
And, over 40 protestors were killed.

00:56.455 --> 00:59.755
And although the soldiers obeyed the president's orders at

00:59.755 --> 01:01.555
the time, it sparked deep resentment

01:01.735 --> 01:03.675
and contributed to a coup the next year

01:03.675 --> 01:04.915
that created the conditions

01:04.935 --> 01:08.715
for Liberia's descent into civil war between 1989

01:08.855 --> 01:10.035
and 2003.

01:10.695 --> 01:12.955
Now, when the riots broke out, the us

01:13.295 --> 01:16.075
the United States had been training the Liberian military

01:16.335 --> 01:18.515
for 18 years, and the goal was

01:18.515 --> 01:20.635
to create a disciplined democratic force,

01:20.645 --> 01:22.875
which meant training the military both

01:22.895 --> 01:25.275
to respect human rights and civilian authority,

01:25.695 --> 01:28.155
but the president's order pitted these norms against each

01:28.155 --> 01:29.955
other, presenting soldiers with a dilemma,

01:30.245 --> 01:31.475
which norm to follow.

01:32.255 --> 01:35.355
And so the main question that I address in this study is

01:35.815 --> 01:37.755
how does liberal norm conflict

01:37.855 --> 01:39.595
affect military decision making?

01:40.295 --> 01:43.835
And I argue as I'll explain over the next few minutes,

01:43.835 --> 01:47.995
that in moments of norm conflict, militaries will,

01:47.995 --> 01:52.235
instead prioritize a third norm of cohesion over either

01:52.375 --> 01:55.915
of the two liberal norms of respect for human rights,

01:55.975 --> 01:57.795
and civilian control of the military.

01:58.055 --> 02:01.475
So the existence of this soldier's dilemma, which norms

02:01.475 --> 02:03.435
to follow, poses a problem

02:03.575 --> 02:05.515
for countries like the United States

02:05.615 --> 02:09.035
and its allies who try to spread liberal norms globally

02:09.545 --> 02:11.675
through foreign military training.

02:12.495 --> 02:14.315
And I I single out the United States

02:14.315 --> 02:16.715
because it is the world's largest exporter

02:16.715 --> 02:17.995
of foreign military training.

02:18.375 --> 02:19.635
So in the last 20 years,

02:20.015 --> 02:24.235
the United States has spent over 20 $20 billion training

02:24.585 --> 02:27.715
more than 2.3 million military personnel from

02:27.715 --> 02:29.075
most countries in the world.

02:29.695 --> 02:32.395
And this training ranges all the way from professional

02:33.115 --> 02:36.315
military education, at, at the university level, down

02:36.315 --> 02:37.835
to sort of tactical,

02:37.895 --> 02:40.155
and individual unit level field training.

02:40.575 --> 02:42.275
Um, and this training is designed

02:43.055 --> 02:44.635
to impart war fighting capabilities,

02:45.015 --> 02:48.475
but it's also designed to do other things like impart us

02:48.475 --> 02:49.675
values to its partners.

02:50.095 --> 02:52.435
For example, nearly half of all states

02:52.465 --> 02:56.035
that receive US military training in a given year have

02:56.065 --> 02:58.435
promotion of respect for human rights

02:58.455 --> 03:00.965
or civilian control of the military as one

03:00.965 --> 03:03.885
of the primary foreign policy objectives of training,

03:04.195 --> 03:06.965
according to reports that the US Defense Department

03:06.965 --> 03:08.685
and State Department publish every year.

03:09.545 --> 03:12.445
Um, and the states that get this norms training typically

03:12.725 --> 03:15.125
represent places where US policy makers believe

03:15.315 --> 03:18.965
that encouraging liberal behavior in partner militaries is

03:19.065 --> 03:21.645
key to political stability and security.

03:22.145 --> 03:24.885
So in Africa, for example, nearly 90%

03:24.885 --> 03:28.445
of countries receive training on these norms each year.

03:29.025 --> 03:31.445
And states that get training on one norm typically get

03:31.605 --> 03:33.525
training on the other norm, as well.

03:33.745 --> 03:36.925
So because liberalism views human right respect

03:36.925 --> 03:38.925
for human rights and civilian authority is mutually

03:39.245 --> 03:42.005
reinforcing, the United States typically bundles these

03:42.055 --> 03:43.725
norms, as a package deal.

03:44.185 --> 03:48.205
In fact, US law as of 2017 now requires that all efforts

03:48.305 --> 03:51.525
to build capacity of foreign forces include,

03:51.845 --> 03:53.125
training on both of these norms.

03:53.825 --> 03:55.805
But this assumption doesn't always hold up.

03:56.385 --> 03:59.605
Uh, these norms often are not mutually reinforcing in the

03:59.605 --> 04:00.645
weak democracies

04:00.665 --> 04:03.445
and fragile states where this training is most prevalent,

04:03.815 --> 04:06.085
these states lack strong institutions

04:06.185 --> 04:07.845
to constrain executive leaders.

04:08.315 --> 04:11.005
They lack robust military justice systems

04:11.145 --> 04:14.125
to help Aries establish the legality of orders.

04:15.145 --> 04:19.475
And so in these places, when militaries confront situations

04:19.505 --> 04:21.675
with conflicting implications for behavior,

04:22.795 --> 04:26.475
deciding which norm to follow is not always so clear cut.

04:26.935 --> 04:28.675
And as a result, I argue

04:28.675 --> 04:31.675
that militaries will prioritize sticking together

04:32.345 --> 04:34.155
over complying with liberal norms.

04:34.465 --> 04:36.075
They will prioritize cohesion,

04:36.645 --> 04:39.745
and by cohesion I mean the bonds that enable militaries

04:39.885 --> 04:43.945
to operate in a unified group and mission oriented way.

04:44.405 --> 04:47.505
And so I treat cohesion in this, in this study as a norm

04:48.095 --> 04:50.985
that motivates soldiers to prioritize loyalty

04:51.205 --> 04:53.705
and commitment to their unit and the military.

04:54.285 --> 04:55.625
But it's a norm that's hard

04:55.625 --> 04:58.185
to fully separate from interests,

04:58.185 --> 05:01.905
because cohesion also enables the military institution's

05:01.905 --> 05:02.945
self preservation.

05:03.935 --> 05:05.035
Um, and importantly,

05:05.755 --> 05:08.595
training from liberal providers like the United States

05:09.345 --> 05:11.275
imparts more than just liberal norms.

05:11.495 --> 05:13.555
It also helps to instill cohesion.

05:14.015 --> 05:17.835
So training fosters a, a common identity skills

05:18.135 --> 05:20.835
and modes of communication that serve

05:21.135 --> 05:23.555
to help instill cohesion over time.

05:24.055 --> 05:26.515
So coming back to my main question, right?

05:26.575 --> 05:29.115
How do militaries respond to norm conflict?

05:29.735 --> 05:33.875
Now, US policy discourse rarely addresses this problem

05:34.355 --> 05:37.155
directly, but there's an implicit expectation

05:37.345 --> 05:39.315
that if civilians order the military

05:39.375 --> 05:40.595
to repress human rights,

05:40.825 --> 05:43.755
then the military will temporarily prioritize human rights

05:43.865 --> 05:45.275
over civilian authority.

05:45.935 --> 05:47.915
So the Department of Defense, likes

05:47.915 --> 05:50.715
to quote English philosopher John Locke on this point.

05:51.135 --> 05:53.595
Uh, Locke believed that that human rights,

05:53.855 --> 05:56.835
not governments came first in the natural order of things.

05:57.625 --> 05:58.765
And so this expectation

05:58.765 --> 06:01.205
that militaries will choose human rights

06:01.435 --> 06:02.645
assumes several things.

06:03.145 --> 06:04.925
So first, that well share,

06:04.925 --> 06:07.085
well-trained militaries will share this

06:07.085 --> 06:08.205
rank ordering of norms.

06:09.075 --> 06:11.895
Second, the choosing between conflicting norms is easy.

06:12.475 --> 06:14.775
Um, so the psychology literature suggests

06:14.775 --> 06:18.095
that when people have to choose between deeply held values,

06:18.805 --> 06:21.375
they can frame the problem as, as what's known

06:21.375 --> 06:22.695
as a tragic trade off

06:22.755 --> 06:26.335
and avoid damage to self-esteem or social standing.

06:27.195 --> 06:30.775
And third, that norm conflict will not weaken support

06:30.795 --> 06:32.375
for either norm, right?

06:32.375 --> 06:35.255
So if anything, that the process of reasoning

06:35.275 --> 06:38.215
and adjudicating between norms might strengthen them both.

06:39.175 --> 06:41.835
Um, and importantly, norm conflicts are not seen

06:41.835 --> 06:43.835
as problematic for cohesion, right?

06:43.835 --> 06:47.195
If anything, cohesion should make militaries,

06:47.305 --> 06:49.275
more unified in doing the right thing.

06:50.135 --> 06:51.635
Uh, but I argue that that all

06:51.635 --> 06:54.115
of these policy assumptions are flawed.

06:54.705 --> 06:56.845
And so when liberal norms clash,

06:57.055 --> 07:00.885
militaries instead fall back on this third norm of cohesion

07:00.885 --> 07:03.165
that is most consistent with group interests.

07:04.105 --> 07:05.765
And, and, and I, I think

07:05.765 --> 07:08.325
that this is the more likely outcome for, for two reasons.

07:08.865 --> 07:12.725
So first, because conflict makes the contested norms

07:12.725 --> 07:14.125
less salient, right?

07:14.145 --> 07:15.925
It eliminates the sort of taken

07:16.145 --> 07:18.285
for granted quality of norms.

07:18.665 --> 07:20.365
And so once norms are questionable,

07:20.945 --> 07:25.005
it invites cost benefit calculations that create incentives

07:25.225 --> 07:28.845
for soldiers to instrumentally follow other norms like

07:29.005 --> 07:32.235
cohesion, which are neither inherently liberal

07:32.895 --> 07:34.835
nor conflictive with their self-interest.

07:35.735 --> 07:39.315
And second, these norm conflict scenarios threaten cohesion,

07:39.335 --> 07:42.595
making it seem more important to protect so, so crises

07:42.665 --> 07:45.315
that pit political leaders against the population

07:45.545 --> 07:47.875
with the military in the middle have really fraught

07:47.875 --> 07:49.915
implications for military unity.

07:50.415 --> 07:54.195
So soldiers may be reluctant to make choices that they see

07:54.195 --> 07:55.515
as harmful to cohesion.

07:56.455 --> 07:59.075
Now, importantly, this dynamic doesn't mean

07:59.145 --> 08:02.435
that cohesion is fundamentally at odds with liberal norms,

08:02.655 --> 08:04.595
and it doesn't automatically predict

08:05.835 --> 08:08.555
specific norm violations in any given situation,

08:08.975 --> 08:10.995
but rather it creates the conditions

08:11.145 --> 08:13.355
that make norm violations possible.

08:13.895 --> 08:15.875
And so my expectation is

08:15.875 --> 08:17.955
that military decision making will be driven

08:18.095 --> 08:20.195
by whatever best serves cohesion,

08:20.365 --> 08:23.955
which may involve violating human rights, civilian control,

08:24.255 --> 08:27.515
or both to test the arguments implications.

08:27.995 --> 08:31.355
I use micro level experimental evidence from Liberia.

08:31.695 --> 08:35.825
But why Liberia? Well, Liberia represents a typical case

08:35.825 --> 08:37.585
where norm conflict might occur.

08:38.045 --> 08:40.225
Uh, it's a transitional democracy with,

08:40.225 --> 08:42.905
with weak institutions and a history of civil war,

08:43.405 --> 08:46.265
but it's not an easy case for my argument.

08:46.765 --> 08:50.545
Uh, after the Civil War ended in 2003, the United States

08:51.105 --> 08:53.705
rebuilt the Liberian military with a training program

08:53.895 --> 08:56.245
that heavily emphasized liberal norms.

08:56.905 --> 09:00.045
So if, if, if policy expectations about the effects

09:00.045 --> 09:03.405
of training are true anywhere, we should see them here.

09:04.185 --> 09:07.165
So the experimental evidence comes from a representative

09:07.225 --> 09:11.445
survey of 270 active duty armed forces of Liberia,

09:11.765 --> 09:14.605
a FL soldiers that I conducted in 2017.

09:15.585 --> 09:18.325
Now, this is a small sample, but it's a small military.

09:18.505 --> 09:21.005
So the sample represented around 15%

09:21.005 --> 09:22.565
of the force at the time of the survey.

09:23.465 --> 09:26.205
Um, still the small sample means we should be cautious

09:26.225 --> 09:27.325
in interpreting the results.

09:28.065 --> 09:30.445
And in addition to the survey, I conducted

09:30.445 --> 09:33.645
around 60 interviews with Liberian government

09:33.645 --> 09:35.805
and military officials, US officials,

09:36.425 --> 09:38.565
and many of the people who implemented

09:38.585 --> 09:40.445
and participated in the training program,

09:40.595 --> 09:42.845
both on the American and Liberian sides.

09:44.025 --> 09:47.485
Um, so to test these competing, expectations,

09:47.945 --> 09:49.245
us policy expectations

09:49.245 --> 09:51.845
that militaries will prioritize human rights

09:51.905 --> 09:55.045
and my expectation that they, that they won't,

09:55.065 --> 09:56.685
and they will prioritize cohesion.

09:56.985 --> 10:00.165
Uh, I conducted an experiment embedded in this survey.

10:00.745 --> 10:04.245
So the experiment primes soldiers who were randomly assigned

10:04.245 --> 10:07.705
to the treatment group to think about liberal norm conflict

10:07.845 --> 10:10.105
by asking them to consider a scenario

10:10.685 --> 10:12.905
in which the president orders the military

10:13.085 --> 10:14.985
to put down protests with force.

10:15.325 --> 10:17.105
And the military commander refuses.

10:17.565 --> 10:19.825
And I worked with Liberian military leadership

10:19.885 --> 10:23.185
to design the scenario so that it would provide,

10:23.365 --> 10:26.945
an experimental or stress test of competing norms

10:27.015 --> 10:29.065
that was relevant to the local context.

10:29.925 --> 10:30.945
Um, and immediately

10:30.945 --> 10:33.105
after hearing the scenario as part

10:33.105 --> 10:34.425
of this treatment condition,

10:34.775 --> 10:37.585
respondents were asked whether the military should have

10:37.865 --> 10:39.865
intervened, should not have intervened,

10:40.445 --> 10:41.985
or it doesn't matter as long

10:41.985 --> 10:43.305
as the military sticks together.

10:44.085 --> 10:46.695
Now, respondents who were randomly assigned

10:46.695 --> 10:49.695
to the control group did not hear any scenario.

10:50.235 --> 10:52.735
Um, instead they were asked a generic question

10:52.735 --> 10:55.575
that was similar in structure to the treatment condition.

10:56.075 --> 10:58.775
Uh, so they were asked whether,

10:59.375 --> 11:01.855
participation in international peacekeeping should be a

11:01.855 --> 11:04.775
primary mission for the A FL should not be a mission,

11:05.115 --> 11:06.775
or it doesn't matter as long

11:06.775 --> 11:08.415
as the military executes the mission.

11:09.275 --> 11:12.815
And afterwards, I asked all the respondents to answer,

11:12.915 --> 11:16.415
a series of outcome measure questions that were designed

11:16.475 --> 11:19.295
to gauge their support for prioritizing human rights

11:19.755 --> 11:21.255
and their support for democracy

11:21.515 --> 11:25.335
or its alternatives like army rule, which I use as proxies

11:25.335 --> 11:26.695
for liberal civilian control.

11:27.855 --> 11:30.315
And to gauge support for prioritizing cohesion,

11:30.795 --> 11:34.115
I created a dummy variable that took a value of one,

11:34.455 --> 11:38.195
if the respondent chose the third response option

11:38.195 --> 11:40.395
to the treatment or control question, right?

11:40.395 --> 11:42.755
So saying that it doesn't matter what the military does

11:42.895 --> 11:45.755
as long as it sticks together or executes the mission.

11:46.255 --> 11:48.835
Now those questions were not worded identically.

11:49.095 --> 11:51.475
So this variable only gives us a rough measure

11:51.655 --> 11:53.155
of prioritizing cohesion,

11:53.535 --> 11:56.005
which means it can't provide conclusive evidence

11:56.005 --> 11:59.085
of causality, but it offers suggestive evidence

11:59.315 --> 12:01.765
that I augment with some additional tests that,

12:01.765 --> 12:02.805
that I'll tell you,

12:02.885 --> 12:04.125
a little more about in just a minute.

12:04.425 --> 12:08.085
It, my baseline tests are difference in means tests

12:08.195 --> 12:11.565
that simply compare average responses across

12:12.145 --> 12:13.485
the control and treatment groups.

12:14.415 --> 12:17.875
And these results suggest that soldiers who are exposed

12:18.015 --> 12:20.155
to norm conflict are less likely

12:20.255 --> 12:22.315
to support prioritizing human rights

12:22.695 --> 12:25.435
and significantly more likely to prioritize cohesion.

12:25.815 --> 12:28.595
And there really isn't any difference in support for liberal

12:29.115 --> 12:31.795
civilian control across the treatment or control groups.

12:32.655 --> 12:33.755
So overall,

12:33.755 --> 12:37.395
these initial results provide some mixed support for my argument.

12:38.135 --> 12:40.955
Um, importantly, these results provide no support

12:41.375 --> 12:42.515
for the policy assumption

12:42.545 --> 12:45.075
that soldiers will prioritize human rights under pressure.

12:45.215 --> 12:46.595
If anything, it's the opposite.

12:47.615 --> 12:49.515
But it turns out that whether

12:49.615 --> 12:52.595
or not soldiers received US training affects

12:52.775 --> 12:54.715
how they respond to norm conflict.

12:55.015 --> 12:56.595
And this part is really interesting.

12:56.935 --> 13:00.595
So do soldiers with US training respond differently

13:00.695 --> 13:03.155
to norm conflict than soldiers without US training?

13:03.855 --> 13:06.275
And the Liberia case lets me answer this question

13:06.465 --> 13:09.515
because even though the United States trained the,

13:09.615 --> 13:10.795
the new military

13:10.805 --> 13:13.595
after the Civil War ended, there was variation in

13:13.615 --> 13:14.675
who got what training.

13:15.375 --> 13:18.835
Um, so recruits entered the military in eight groups,

13:19.135 --> 13:20.155
called batches.

13:20.185 --> 13:23.875
This was their term, not mine, between 2006 and 2015.

13:24.935 --> 13:27.475
And so the American program designers,

13:27.545 --> 13:28.755
told me in interviews

13:28.755 --> 13:30.875
that they put liberal norms at the

13:30.875 --> 13:32.035
center of the training program.

13:32.295 --> 13:35.595
And so the US trained, batches received weeks

13:35.695 --> 13:37.995
of intensive training on human rights,

13:38.085 --> 13:40.395
civil military relations, laws of war.

13:40.935 --> 13:45.075
But the US program only ran from 2006 to 2009.

13:45.855 --> 13:47.915
The first five groups received extension,

13:48.065 --> 13:49.435
extensive US training.

13:50.095 --> 13:52.315
The last three groups, batches six to eight,

13:52.505 --> 13:53.635
were Liberian led

13:53.655 --> 13:55.755
and only received about half as much training.

13:56.335 --> 13:58.435
And while the US training emphasized norms,

13:58.545 --> 14:00.755
this dropped off in the Liberian training due

14:00.755 --> 14:01.955
to resource constraints.

14:03.175 --> 14:06.555
So I use each bat respondent's batch number,

14:06.555 --> 14:08.235
which they shared as part of the survey,

14:08.775 --> 14:12.035
and compare respondents who joined the military in groups

14:12.035 --> 14:15.155
that got extensive US training to respondents

14:15.155 --> 14:17.715
who joined the military in groups that got no US training.

14:18.635 --> 14:20.175
And the results are surprising.

14:21.285 --> 14:24.615
They suggest that US training leads soldiers

14:24.915 --> 14:28.455
to act less in line with policy expectations

14:28.755 --> 14:32.175
and more in line with my argument when they are exposed

14:32.435 --> 14:35.055
to this experimental norm conflict scenario.

14:35.835 --> 14:39.615
So respondents with US training were significantly less

14:39.615 --> 14:43.055
likely to express willingness to prioritize human rights.

14:43.065 --> 14:45.095
After hearing the norm conflict scenario,

14:45.725 --> 14:48.335
they were somewhat less likely to express,

14:48.735 --> 14:49.855
absolute support for democracy

14:50.155 --> 14:53.525
and somewhat more likely to express support for army rule.

14:54.225 --> 14:55.885
And they were significantly more likely

14:55.945 --> 14:58.965
to prioritize cohesion when exposed to norm conflict.

14:59.705 --> 15:00.925
So even though not all

15:00.925 --> 15:03.125
of the differences are statistically significant,

15:03.985 --> 15:06.885
the pattern of coefficient is consistent across measures

15:07.145 --> 15:11.645
and suggests that US training makes soldiers less willing

15:11.705 --> 15:14.525
to prioritize human rights, less supportive

15:14.525 --> 15:15.965
of liberal civilian control,

15:16.305 --> 15:20.565
and more likely to prioritize cohesion when they are exposed

15:20.745 --> 15:22.125
to liberal norm conflict.

15:23.055 --> 15:26.515
And so, so finally, as I I, I mentioned a minute ago

15:26.815 --> 15:29.715
to probe the relationship between norm conflict

15:29.715 --> 15:33.355
and cohesion a little deeper, I leverage a series

15:33.375 --> 15:36.875
of questions in the survey that all respondents were asked

15:37.095 --> 15:39.995
before they were randomly assigned to treatment or control.

15:40.445 --> 15:44.035
These questions gauged respondent's perceptions of equality

15:44.095 --> 15:46.195
and cohesion within their units.

15:47.465 --> 15:49.645
And I found that soldiers who

15:49.905 --> 15:53.405
who perceived low cohesion in their units to begin

15:53.405 --> 15:57.725
with were much more likely to respond that it doesn't matter

15:57.725 --> 15:59.165
what the military does as long

15:59.165 --> 16:01.085
as it sticks together when exposed

16:01.085 --> 16:02.445
to the norm conflict scenario.

16:03.225 --> 16:06.765
And so this provides further evidence that soldiers

16:06.905 --> 16:09.325
who are worried about cohesion in the first place

16:10.145 --> 16:13.205
are more likely to interpret norm conflicts

16:13.225 --> 16:14.565
as posing a threat to it.

16:15.525 --> 16:17.165
Summarize my findings. I find

16:17.165 --> 16:21.565
that norm conflict weakens support for liberal norms,

16:21.565 --> 16:23.845
and strengthens support for cohesion.

16:24.505 --> 16:27.325
So contrary to US policy expectations

16:28.035 --> 16:31.645
that well-trained militaries will prioritize human rights

16:31.715 --> 16:34.805
over civilian control in a crisis, I show

16:34.805 --> 16:37.605
that norm conflict tends to reduce soldiers' willingness

16:37.605 --> 16:38.845
to prioritize human rights

16:38.945 --> 16:41.565
and increase their willingness to prioritize cohesion.

16:42.025 --> 16:44.365
And to the degree that US training shapes

16:44.665 --> 16:46.725
how soldiers respond to norm conflict,

16:46.785 --> 16:48.405
it makes these effects stronger

16:49.025 --> 16:52.205
and so taken together, these results show a new way

16:52.205 --> 16:54.885
that alternative norms like cohesion come

16:54.905 --> 16:56.965
to dominate military decision making.

16:57.955 --> 17:00.935
Now, one limitation to this work is

17:00.935 --> 17:03.735
that experiments can only be conducted in one place at a

17:03.735 --> 17:06.935
time, meaning they raise questions about external validity

17:06.995 --> 17:08.255
or generalizability.

17:09.275 --> 17:11.935
So future work should replicate

17:11.955 --> 17:14.335
and extend the experiment in other places

17:14.715 --> 17:17.895
and context to establish how these results generalize.

17:18.965 --> 17:22.145
And future research should also look at how long the effects

17:22.145 --> 17:24.025
of liberal norm conflict persist

17:24.405 --> 17:25.705
and the connections to behavior.

17:26.485 --> 17:30.225
So the experiment shows us soldiers immediate reactions,

17:30.525 --> 17:32.665
and shifts in attitudes about norms.

17:32.885 --> 17:35.585
But it doesn't show us how long those reactions last

17:36.085 --> 17:40.065
or what specific behaviors occur in a given situation.

17:40.365 --> 17:42.905
So do norms bounce back after the crises,

17:43.205 --> 17:44.865
or does norm conflict contribute

17:44.885 --> 17:46.945
to permanent weakening of these norms?

17:47.125 --> 17:49.725
So these are important areas for future research.

17:50.545 --> 17:54.005
Um, the research also has important implications for policy.

17:54.665 --> 17:57.645
So the United States and its allies use training

17:57.825 --> 18:00.565
to promote liberal norms to solve the problem

18:01.265 --> 18:04.645
of increasing military capability while maintaining civilian

18:04.715 --> 18:06.805
control and protecting human rights.

18:07.525 --> 18:10.425
But a key takeaway from the experimental evidence is

18:10.425 --> 18:12.145
that more training is not always better

18:12.845 --> 18:17.345
and may inadvertently encourage norm violating behavior, not

18:17.345 --> 18:20.265
because liberal training is is worse than other forms

18:20.365 --> 18:21.425
of training, but

18:21.425 --> 18:24.785
because training also imparts other things like cohesion

18:25.085 --> 18:26.305
and may increase.

18:26.565 --> 18:27.945
So soldiers' tendencies

18:27.965 --> 18:29.945
to prioritize cohesion under pressure.

18:31.235 --> 18:33.935
And again, not that cohesion itself is a bad thing

18:33.935 --> 18:37.215
because cohesive militaries tend to be effective militaries.

18:38.155 --> 18:39.335
But the point is that

18:39.335 --> 18:43.175
that cohesion can create the conditions under which norm

18:43.175 --> 18:45.895
violations occur if it's not regulated

18:45.895 --> 18:47.935
and directed in the right directions.

18:48.555 --> 18:50.615
And so, so the United States assumes

18:50.925 --> 18:53.095
that increasing a military's capability

18:53.235 --> 18:54.735
and increasing its commitment

18:54.735 --> 18:56.975
to democratic norms will go hand in hand,

18:57.675 --> 18:58.935
but my results suggest

18:58.935 --> 19:00.775
that this is not necessarily the case.

19:01.635 --> 19:04.535
And so instead of doubling down on more training,

19:05.045 --> 19:08.215
liberal providers should think about approaches that

19:08.215 --> 19:12.855
that decouple or sequence efforts to promote norms

19:12.955 --> 19:14.135
and impart capabilities.

19:15.275 --> 19:18.135
And instead of relying on individual level training

19:18.155 --> 19:21.295
to change beliefs, since I show that those beliefs

19:22.035 --> 19:24.575
may change again, under pressure,

19:24.935 --> 19:27.895
providers should focus on strengthening institutional

19:27.905 --> 19:31.495
guardrails such as military justice systems that help

19:31.515 --> 19:35.335
to regulate and constrain behavior even in the face

19:35.475 --> 19:37.535
of potential attitudinal shifts.

00:00 - The study that I'm talking about today explores how militaries respond to the problem of what I call liberal norm conflict between the norms of respect for human rights and obedience to civilian authority. And by norms, I mean, shared understandings

00:14 - of appropriate behavior and and respect for human rights and civilian authority are arguably two of the most important norms that govern liberal civil military relations between states, societies, and their militaries. But when political leaders order the military to repress human rights, it activates a conflict between these two norms. And so to illustrate what I mean by liberal norm conflict, let me take you back for a moment

00:40 - to the 1979 Rice riots in Monrovia, Liberia. So in 1979, the government hiked to the price of rice riots broke out, and the president personally ordered soldiers to fire on protestors. And, over 40 protestors were killed. And although the soldiers obeyed the president's orders at the time, it sparked deep resentment and contributed to a coup the next year that created the conditions for Liberia's descent into civil war between 1989

01:08 - and 2003. Now, when the riots broke out, the us the United States had been training the Liberian military for 18 years, and the goal was to create a disciplined democratic force, which meant training the military both to respect human rights and civilian authority, but the president's order pitted these norms against each other, presenting soldiers with a dilemma, which norm to follow.

01:32 - And so the main question that I address in this study is how does liberal norm conflict affect military decision making? And I argue as I'll explain over the next few minutes, that in moments of norm conflict, militaries will, instead prioritize a third norm of cohesion over either of the two liberal norms of respect for human rights, and civilian control of the military. So the existence of this soldier's dilemma, which norms to follow, poses a problem

02:03 - for countries like the United States and its allies who try to spread liberal norms globally through foreign military training. And I I single out the United States because it is the world's largest exporter of foreign military training. So in the last 20 years, the United States has spent over 20 $20 billion training more than 2.3 million military personnel from most countries in the world.

02:29 - And this training ranges all the way from professional military education, at, at the university level, down to sort of tactical, and individual unit level field training. and this training is designed to impart war fighting capabilities, but it's also designed to do other things like impart us values to its partners. For example, nearly half of all states that receive US military training in a given year have

02:56 - promotion of respect for human rights or civilian control of the military as one of the primary foreign policy objectives of training, according to reports that the US Defense Department and State Department publish every year. and the states that get this norms training typically represent places where US policy makers believe that encouraging liberal behavior in partner militaries is key to political stability and security. So in Africa, for example, nearly 90%

03:24 - of countries receive training on these norms each year. And states that get training on one norm typically get training on the other norm, as well. So because liberalism views human right respect for human rights and civilian authority is mutually reinforcing, the United States typically bundles these norms, as a package deal. In fact, US law as of 2017 now requires that all efforts to build capacity of foreign forces include, training on both of these norms.

03:53 - But this assumption doesn't always hold up. these norms often are not mutually reinforcing in the weak democracies and fragile states where this training is most prevalent, these states lack strong institutions to constrain executive leaders. They lack robust military justice systems to help Aries establish the legality of orders. And so in these places, when militaries confront situations with conflicting implications for behavior,

04:22 - deciding which norm to follow is not always so clear cut. And as a result, I argue that militaries will prioritize sticking together over complying with liberal norms. They will prioritize cohesion, and by cohesion I mean the bonds that enable militaries to operate in a unified group and mission oriented way. And so I treat cohesion in this, in this study as a norm that motivates soldiers to prioritize loyalty and commitment to their unit and the military.

04:54 - But it's a norm that's hard to fully separate from interests, because cohesion also enables the military institution's self preservation. and importantly, training from liberal providers like the United States imparts more than just liberal norms. It also helps to instill cohesion. So training fosters a, a common identity skills and modes of communication that serve

05:21 - to help instill cohesion over time. So coming back to my main question, right? How do militaries respond to norm conflict? Now, US policy discourse rarely addresses this problem directly, but there's an implicit expectation that if civilians order the military to repress human rights, then the military will temporarily prioritize human rights over civilian authority. So the Department of Defense, likes

05:47 - to quote English philosopher John Locke on this point. Locke believed that that human rights, not governments came first in the natural order of things. And so this expectation that militaries will choose human rights assumes several things. So first, that well share, well-trained militaries will share this rank ordering of norms. Second, the choosing between conflicting norms is easy.

06:12 - so the psychology literature suggests that when people have to choose between deeply held values, they can frame the problem as, as what's known as a tragic trade off and avoid damage to self-esteem or social standing. And third, that norm conflict will not weaken support for either norm, right? So if anything, that the process of reasoning and adjudicating between norms might strengthen them both. and importantly, norm conflicts are not seen

06:41 - as problematic for cohesion, right? If anything, cohesion should make militaries, more unified in doing the right thing. but I argue that that all of these policy assumptions are flawed. And so when liberal norms clash, militaries instead fall back on this third norm of cohesion that is most consistent with group interests. And, and, and I, I think that this is the more likely outcome for, for two reasons.

07:08 - So first, because conflict makes the contested norms less salient, right? It eliminates the sort of taken for granted quality of norms. And so once norms are questionable, it invites cost benefit calculations that create incentives for soldiers to instrumentally follow other norms like cohesion, which are neither inherently liberal nor conflictive with their self-interest. And second, these norm conflict scenarios threaten cohesion,

07:39 - making it seem more important to protect so, so crises that pit political leaders against the population with the military in the middle have really fraught implications for military unity. So soldiers may be reluctant to make choices that they see as harmful to cohesion. Now, importantly, this dynamic doesn't mean that cohesion is fundamentally at odds with liberal norms, and it doesn't automatically predict specific norm violations in any given situation,

08:08 - but rather it creates the conditions that make norm violations possible. And so my expectation is that military decision making will be driven by whatever best serves cohesion, which may involve violating human rights, civilian control, or both to test the arguments implications. I use micro level experimental evidence from Liberia. But why Liberia? Well, Liberia represents a typical case where norm conflict might occur.

08:38 - it's a transitional democracy with, with weak institutions and a history of civil war, but it's not an easy case for my argument. after the Civil War ended in 2003, the United States rebuilt the Liberian military with a training program that heavily emphasized liberal norms. So if, if, if policy expectations about the effects of training are true anywhere, we should see them here. So the experimental evidence comes from a representative survey of 270 active duty armed forces of Liberia,

09:11 - a FL soldiers that I conducted in 2017. Now, this is a small sample, but it's a small military. So the sample represented around 15% of the force at the time of the survey. still the small sample means we should be cautious in interpreting the results. And in addition to the survey, I conducted around 60 interviews with Liberian government and military officials, US officials, and many of the people who implemented

09:38 - and participated in the training program, both on the American and Liberian sides. so to test these competing, expectations, us policy expectations that militaries will prioritize human rights and my expectation that they, that they won't, and they will prioritize cohesion. I conducted an experiment embedded in this survey. So the experiment primes soldiers who were randomly assigned to the treatment group to think about liberal norm conflict

10:07 - by asking them to consider a scenario in which the president orders the military to put down protests with force. And the military commander refuses. And I worked with Liberian military leadership to design the scenario so that it would provide, an experimental or stress test of competing norms that was relevant to the local context. and immediately after hearing the scenario as part

10:33 - of this treatment condition, respondents were asked whether the military should have intervened, should not have intervened, or it doesn't matter as long as the military sticks together. Now, respondents who were randomly assigned to the control group did not hear any scenario. instead they were asked a generic question that was similar in structure to the treatment condition. so they were asked whether,

10:59 - participation in international peacekeeping should be a primary mission for the A FL should not be a mission, or it doesn't matter as long as the military executes the mission. And afterwards, I asked all the respondents to answer, a series of outcome measure questions that were designed to gauge their support for prioritizing human rights and their support for democracy or its alternatives like army rule, which I use as proxies for liberal civilian control.

11:27 - And to gauge support for prioritizing cohesion, I created a dummy variable that took a value of one, if the respondent chose the third response option to the treatment or control question, right? So saying that it doesn't matter what the military does as long as it sticks together or executes the mission. Now those questions were not worded identically. So this variable only gives us a rough measure of prioritizing cohesion, which means it can't provide conclusive evidence

11:56 - of causality, but it offers suggestive evidence that I augment with some additional tests that, that I'll tell you, a little more about in just a minute. It, my baseline tests are difference in means tests that simply compare average responses across the control and treatment groups. And these results suggest that soldiers who are exposed to norm conflict are less likely to support prioritizing human rights

12:22 - and significantly more likely to prioritize cohesion. And there really isn't any difference in support for liberal civilian control across the treatment or control groups. So overall, these initial results provide some mixed support for my argument. importantly, these results provide no support for the policy assumption that soldiers will prioritize human rights under pressure. If anything, it's the opposite. But it turns out that whether

12:49 - or not soldiers received US training affects how they respond to norm conflict. And this part is really interesting. So do soldiers with US training respond differently to norm conflict than soldiers without US training? And the Liberia case lets me answer this question because even though the United States trained the, the new military after the Civil War ended, there was variation in who got what training.

13:15 - so recruits entered the military in eight groups, called batches. This was their term, not mine, between 2006 and 2015. And so the American program designers, told me in interviews that they put liberal norms at the center of the training program. And so the US trained, batches received weeks of intensive training on human rights, civil military relations, laws of war.

13:40 - But the US program only ran from 2006 to 2009. The first five groups received extension, extensive US training. The last three groups, batches six to eight, were Liberian led and only received about half as much training. And while the US training emphasized norms, this dropped off in the Liberian training due to resource constraints. So I use each bat respondent's batch number,

14:06 - which they shared as part of the survey, and compare respondents who joined the military in groups that got extensive US training to respondents who joined the military in groups that got no US training. And the results are surprising. They suggest that US training leads soldiers to act less in line with policy expectations and more in line with my argument when they are exposed to this experimental norm conflict scenario. So respondents with US training were significantly less

14:39 - likely to express willingness to prioritize human rights. After hearing the norm conflict scenario, they were somewhat less likely to express, absolute support for democracy and somewhat more likely to express support for army rule. And they were significantly more likely to prioritize cohesion when exposed to norm conflict. So even though not all of the differences are statistically significant, the pattern of coefficient is consistent across measures

15:07 - and suggests that US training makes soldiers less willing to prioritize human rights, less supportive of liberal civilian control, and more likely to prioritize cohesion when they are exposed to liberal norm conflict. And so, so finally, as I I, I mentioned a minute ago to probe the relationship between norm conflict and cohesion a little deeper, I leverage a series of questions in the survey that all respondents were asked before they were randomly assigned to treatment or control.

15:40 - These questions gauged respondent's perceptions of equality and cohesion within their units. And I found that soldiers who who perceived low cohesion in their units to begin with were much more likely to respond that it doesn't matter what the military does as long as it sticks together when exposed to the norm conflict scenario. And so this provides further evidence that soldiers who are worried about cohesion in the first place

16:10 - are more likely to interpret norm conflicts as posing a threat to it. Summarize my findings. I find that norm conflict weakens support for liberal norms, and strengthens support for cohesion. So contrary to US policy expectations that well-trained militaries will prioritize human rights over civilian control in a crisis, I show that norm conflict tends to reduce soldiers' willingness to prioritize human rights

16:38 - and increase their willingness to prioritize cohesion. And to the degree that US training shapes how soldiers respond to norm conflict, it makes these effects stronger and so taken together, these results show a new way that alternative norms like cohesion come to dominate military decision making. Now, one limitation to this work is that experiments can only be conducted in one place at a time, meaning they raise questions about external validity

17:06 - or generalizability. So future work should replicate and extend the experiment in other places and context to establish how these results generalize. And future research should also look at how long the effects of liberal norm conflict persist and the connections to behavior. So the experiment shows us soldiers immediate reactions, and shifts in attitudes about norms. But it doesn't show us how long those reactions last

17:36 - or what specific behaviors occur in a given situation. So do norms bounce back after the crises, or does norm conflict contribute to permanent weakening of these norms? So these are important areas for future research. the research also has important implications for policy. So the United States and its allies use training to promote liberal norms to solve the problem of increasing military capability while maintaining civilian control and protecting human rights.

18:07 - But a key takeaway from the experimental evidence is that more training is not always better and may inadvertently encourage norm violating behavior, not because liberal training is is worse than other forms of training, but because training also imparts other things like cohesion and may increase. So soldiers' tendencies to prioritize cohesion under pressure. And again, not that cohesion itself is a bad thing

18:33 - because cohesive militaries tend to be effective militaries. But the point is that that cohesion can create the conditions under which norm violations occur if it's not regulated and directed in the right directions. And so, so the United States assumes that increasing a military's capability and increasing its commitment to democratic norms will go hand in hand, but my results suggest

18:58 - that this is not necessarily the case. And so instead of doubling down on more training, liberal providers should think about approaches that that decouple or sequence efforts to promote norms and impart capabilities. And instead of relying on individual level training to change beliefs, since I show that those beliefs may change again, under pressure, providers should focus on strengthening institutional guardrails such as military justice systems that help

19:31 - to regulate and constrain behavior even in the face of potential attitudinal shifts.

Faculti AI can make mistakes