HYPOTHESIS LAB · SEPTEMBER 30, 2026
Separation, Contact & Conflict
An interesting hypothesis earns an investigation, not a conclusion.
1. Observation
Watching The Road Between Us, I kept noticing fences. They protected communities from people outside them—but once breached, they could also restrict movement.
That made me wonder what physical separation does to conflict beyond its immediate security purpose.
It also made me wonder: Did Palestinians in Gaza have the same infrastructure to protect themselves during attacks?
The question was no longer simply whether a fence protects people. I began noticing the architecture surrounding two neighboring populations:
Who has a protected room? Who has a communal shelter? Who controls movement? Who can leave? Who can enter? Who has access to construction materials? Who encounters the people on the other side—and in what role?
Those questions became the starting point for the investigation.
2. First Hypothesis
Fences can help facilitate the conditions that lead to war.
The hypothesis is not that building a fence automatically causes conflict, or that fences have no legitimate security purpose.
The question is whether prolonged physical separation can become one component of a larger system that creates or reinforces conditions associated with conflict.
The fence may protect people from an immediate threat while also changing:
- who can interact with whom;
- who can move freely and who cannot;
- what kinds of interactions remain possible;
- how much ordinary exposure each population has to the other;
- how each population experiences safety, restriction, and inequality; and
- what each population comes to believe about the people on the other side.
The hypothesis therefore isn’t simply “a fence causes war.”
At this stage, those mechanisms are questions to test, not established conclusions.
3. What Would Have to Be True?
For this hypothesis to work, there would need to be mechanisms connecting physical separation to human behavior.
Who do we encounter?
Does physical separation reduce ordinary interactions between populations?
If those interactions disappear, what interactions remain?
If someone’s remaining encounters with another population disproportionately occur during situations involving security, authority, fear, or violence, does that change what they learn about the population as a whole?
What does ordinary contact change?
Does personally knowing people from another population make it harder to reduce that population to a single category based on nationality, religion, ethnicity, language, or another characteristic?
When ordinary exposure disappears, does unfamiliarity increase fear, stereotyping, or perceived threat?
Who controls access?
Who controls the fence? Who decides who crosses? Who can leave? Who can enter?
Does each population experience the restriction equally?
Does separation change access to employment, education, healthcare, businesses, family, land, transportation, or other communities?
And does unequal access create additional grievance independently of whatever caused the original conflict?
Who gets protection?
Who receives the security benefit? Who bears the restrictions created by that security system?
Do people on both sides have comparable infrastructure for protecting themselves when violence occurs?
If they don’t, how does unequal access to protection affect perceptions of safety, power, vulnerability, inequality, and the other population?
When does grievance change behavior?
How much prejudice, discrimination, humiliation, inequality, restriction, violence, or loss can humans experience before their behavior changes?
There may not be a universal threshold. So what determines whether someone responds through political participation, protest, civil disobedience, nonviolent resistance, migration, armed resistance, violence against combatants, or violence against civilians?
Does experiencing harm personally differ from watching it happen repeatedly to members of your family or community?
Does believing peaceful methods have failed change people’s willingness to consider violence? Does being unable to leave matter?
Does inequality matter differently when people can directly observe another population possessing freedoms, protections, resources, or opportunities they cannot access?
Who gets called a terrorist?
Where are the boundaries between protest, civil disobedience, resistance, insurgency, guerrilla warfare, political violence, and terrorism? These are categories to examine, not a necessary progression.
Are those categories applied consistently? Does the identity of the actor change the terminology? Does the identity of the victim?
Does targeting civilians distinguish terrorism from other political violence, or are additional conditions necessary?
Are comparable acts described differently when committed by a government, a military, an insurgency, a resistance movement, a non-state armed organization, or an individual?
Can labeling particular actors ‘terrorists’ cause fear to generalize from those actors to the larger population with which they’re associated?
If so, does that become another feedback mechanism?
An American comparison: January 6
January 6 offers a domestic comparison for examining perceived grievance, political belief, peaceful protest, civil disobedience, illegal disruption, and political violence. These behaviors must be distinguished; they do not form an inevitable sequence.
When different groups engage in comparable political behavior or violence, do we describe them using the same terminology?
Instead of deciding beforehand whether January 6 was equivalent to another event, we can compare behaviors, targets, objectives, methods, casualties, legal treatment, media terminology, and public perception.
That gives us a possible way to test whether terms such as terrorist, insurgent, rioter, protester, militant, extremist, and resistance fighter are being applied according to consistent behavioral definitions or partly according to the identity and perceived legitimacy of the actors.
The Proposed Feedback Loop
So far, one possible mechanism has emerged. Each step below is a proposed link to test.
Feedback Loop
- 01
Violence / fear / grievance
- 02
Physical and social separation
- 03
Fewer mundane human interactions
- 04
Greater proportion of adversarial or unequal interactions
- 05
Limited experiential sample of the other population
- 06
Greater perceived difference or threat
- 07
Greater acceptance of separation and security measures
- 08
Additional restrictions and fewer ordinary interactions
- 09
Potential additional grievance
- 10
Some people become increasingly receptive to confrontation or violence
- 11
New violence
- 12
‘This proves we needed the separation.’