"Long before it's in the papers"
January 28, 2015


Good fences make good neighbors, scientists find

Sept. 13, 2007
World Science staff

Can the cold rea­son­a­ble­ness of math head off the flam­ing irra­t­ional­ity of vi­o­lent con­flict? May­be, say re­search­ers who this week re­ported that a math­e­mat­i­cal mod­el can pre­dict where eth­nic bat­tle will erupt. 

Such stud­ies could help pol­i­cy­makers de­vise so­lu­tions be­fore prob­lems get out of hand, the sci­en­tists said. More than 100 mil­lion peo­ple have died in vi­o­lent con­flict in the past cen­tu­ry, they added, of­ten be­cause of clashes be­tween eth­nic­ally or cul­tur­ally dis­tinct groups. 

The in­ves­ti­ga­tors found that eth­nic­ally mixed ar­eas with poorly de­fined bound­aries were prone to con­flict. The stu­dy, by sci­en­tists at the New Eng­land Com­plex Sys­tems In­sti­tute in Cam­bridge, Mass., and Bran­deis Un­ivers­ity in Wal­tham, Mass., ap­pears in the Sept. 14 is­sue of the re­search jour­nal Sci­ence.

The au­thors de­vised a mod­el based on the as­sump­tion that vi­o­lence does­n’t arise in highly mixed re­gions, since no groups con­sid­er the space en­tirely their own. Vi­o­lence is al­so un­likely in re­gions where groups are sep­a­rate, be­cause they don’t im­pose on each oth­er and the bound­aries are clear. In­stead, par­tial separa­t­ion with un­clear bound­aries fos­ters con­flict, the re­search­ers said.

The mod­el ac­cu­rately pre­dicted the loca­t­ions of re­ported con­flict in the form­er Yu­go­sla­via and in In­dia, the sci­en­tists re­ported. In es­sence, they said, the situa­t­ion is much as de­scribed by the po­et Rob­ert Frost in a well-known po­em, “good fences make good neigh­bors.”

“Vi­o­lence takes place when an eth­nic group is large enough to im­pose cul­tur­al norms on pub­lic spaces, but not large enough to pre­vent those norms from be­ing bro­ken,” said Bran­deis re­searcher May Lim. “Usually this oc­curs in places where bound­aries be­tween groups are un­clear.”

Re­flect­ing an emerg­ing di­rec­tion in sci­ence ap­plied to so­cial pol­i­cy, the study ap­plied sci­en­tif­ic prin­ci­ples of pat­tern forma­t­ion—of­ten used to de­scribe, for ex­am­ple, how chem­i­cals sep­a­rate by type or by state—to a thorny so­cial prob­lem. The re­search­ers found that eth­nic vi­o­lence oc­curs in pre­dictable pat­terns, just as do oth­er col­lec­tive be­hav­iors in oth­er phys­i­cal and bi­o­log­i­cal sys­tems. 

“The con­cept of pat­tern forma­t­ion, while it may have been orig­i­nally de­vel­oped to un­der­stand chem­i­cal sys­tems, is really a sci­en­tif­ic mod­el of col­lec­tive be­hav­iors, in which you look at those as­pects that con­trol over­all be­hav­ior,” said co-author and Com­plex Sys­tems In­sti­tute pres­ident Ya­neer Bar-Yam.

* * *

Send us a comment on this story, or send it to a friend


Sign up for

On Home Page         


  • St­ar found to have lit­tle plan­ets over twice as old as our own

  • “Kind­ness curricu­lum” may bo­ost suc­cess in pre­schoolers


  • Smart­er mice with a “hum­anized” gene?

  • Was black­mail essen­tial for marr­iage to evolve?

  • Plu­to has even cold­er “twin” of sim­ilar size, studies find

  • Could simple an­ger have taught people to coop­erate?


  • F­rog said to de­scribe its home through song

  • Even r­ats will lend a help­ing paw: study

  • D­rug may undo aging-assoc­iated brain changes in ani­mals

Can the cold reasonableness of math head off the flaming irrationality of violent conflict? Maybe, say researchers who this week reported that a mathematical model can predict where ethnic battle will erupt. Such studies could help policymakers devise solutions before problems get out of hand, the scientists said. More than 100 million people have died in violent conflict in the past century, they added, often because of clashes between ethnically or culturally distinct groups. The investigators found that ethnically mixed areas with poorly defined boundaries were prone to conflict. The study, by scientists at the New England Complex Systems Institute in Cambridge, Mass., and Brandeis University in Waltham, Mass., appears in the Sept. 14 issue of the research journal Science. The authors devised a model based on the assumption that violence doesn’t arise in highly mixed regions, since no groups consider the space entirely their own. Violence is also unlikely in regions where groups are separate, because they don’t impose on each other and the boundaries are clear. Instead, partial separation with unclear boundaries fosters conflict, the researchers said. The model accurately predicted the locations of reported conflict in the former Yugoslavia and in India, the scientists reported. In essence, they said, the situation is much as described by the poet Robert Frost in a well-known poem, “good fences make good neighbors.” Well-defined borders help prevent ethnic tension. “Violence takes place when an ethnic group is large enough to impose cultural norms on public spaces, but not large enough to prevent those norms from being broken,” said Brandeis researcher May Lim. “Usually this occurs in places where boundaries between groups are unclear.” Reflecting an emerging direction in science applied to social policy, the study applied scientific principles of pattern formation—often used to describe, for example, how chemicals separate by type or by state—to a thorny social problem. The researchers found that ethnic violence occurs in predictable patterns, just as do other collective behaviors in other physical and biological systems. “The concept of pattern formation, while it may have been originally developed to understand chemical systems, is really a scientific model of collective behaviors, in which you look at those aspects that control overall behavior,” said co-author and Complex Systems Institute president Yaneer Bar-Yam.