"Long before it's in the papers"
August 03, 2010

RETURN TO THE WORLD SCIENCE HOME PAGE

Violent conflicts fit into patterns, researchers find

Dec. 16, 2009
Courtesy Nature
and World Science staff

The size and tim­ing of vi­o­lent events with­in dif­fer­ent hu­man in­sur­gent con­flicts ex­hib­it many si­m­i­lar­i­ties, ac­cord­ing to a study pub­lished in the re­search jour­nal Na­ture this week.

The in­ves­ti­ga­t­ion puts for­ward a mod­el that claims to help quantify col­lec­tive vi­o­lent ac­ti­vity in hu­mans and make a con­nec­tion be­tween hu­man in­sur­gen­cy, glob­al ter­ror­ism and ecol­o­gy. The mod­el could be used to help pre­dict fu­ture vi­o­lent events faced by so­ci­e­ty and po­ten­tially pre­vent their rap­id es­cala­t­ion, ac­cord­ing to the au­thors, Neil John­son of the Un­ivers­ity of Mi­ami in Cor­al Gables, Fla., and col­leagues.

Un­til now col­lec­tive hu­man vi­o­lence has been a ne­glected top­ic, de­spite be­ing one of the most bas­ic forms of hu­man be­hav­iour, John­son and col­leagues said.

This may be be­cause po­lit­i­cal, ide­o­log­ical, cul­tur­al, his­tor­i­cal and geo­graph­i­cal in­flu­ences make con­flict one of the “messi­est” of all hu­man ac­ti­vi­ties to an­a­lyse, they pro­posed. Pre­vi­ous stud­ies have looked at size dis­tri­bu­tions of ca­su­al­ties in wars and ter­ror­ist at­tacks from the early 19th to the late 20th cen­turies, but un­iver­sal pat­terns across wars with re­gards to size or tim­ing with­in con­flict events have been lit­tle ex­plored.

In a study span­ning sev­er­al years, John­son and col­leagues looked at in­sur­gence events, such as those in Af­ghan­i­stan and pre­vi­ously Co­lom­bia, and found that they show re­mark­a­ble sta­tis­ti­cal si­m­i­lar­i­ties in rela­t­ion to tim­ing and “power laws.” Pow­er laws are math­e­mat­i­cal de­scrip­tions of events whose fre­quen­cy de­creases as their size in­creases. For in­stance, an earth­quake three times larg­er than an­oth­er is ex­pected to oc­cur one-ninth as of­ten.

The mod­el used to quantify the da­ta on wars chal­lenges tra­di­tion­al the­o­ries about in­sur­gen­cies that are based on rig­id hi­er­ar­chies and net­works, ac­cord­ing to John­son and col­leagues. It al­so, they added, hints at a pos­si­ble link be­tween col­lec­tive hu­man be­hav­iour in vi­o­lent and non-vi­o­lent set­tings.

* * *

The size and timing of violent events within different human insurgent conflicts exhibit many similarities, according to a study published in the research journal Nature this week. The investigation puts forward a model that claims to help to quantify collective violent activity in humans and make a connection between human insurgency, global terrorism and ecology. The model could be used to help predict future violent events faced by society and potentially prevent their rapid escalation, according to the authors, Neil Johnson of the University of Miami in Coral Gables, Fla., and colleagues. Until now collective human violence has been a neglected topic, despite being one of the most basic forms of human behaviour, Johnson and colleagues said. This may be because political, ideological, cultural, historical and geographical influences make conflict one of the ‘messiest’ of all human activities to analyse. Previous studies have looked at size distributions of casualties in wars and terrorist attacks from the early 19th to the late 20th centuries, but universal patterns across wars with regards to size or timing within conflict events have been little explored. In a study spanning several years, Neil Johnson and colleagues looked at insurgence events, such as those in Afghanistan and previously Colombia, and note that they show remarkable statistical similarities in relation to timing and “power laws.” Power laws are mathematical descriptions of events whose frequency decreases as their size increases. For instance, an earthquake three times larger than another is expected to occur one-ninth as often. The model used to quantify the data on wars challenges traditional theories about insurgencies that are based on rigid hierarchies and networks, according to Johnson and colleagues. It also, they added, hints at a possible link between collective human behaviour in violent and non-violent settings.