Millions of Ghandhis: the power of nonviolent revolution.

"Spreading freedom" has become the White House's all-purpose justification for war, but wars aren't necessarily the best way to plant the seeds of liberty.

A recent study from the D.C.-based group Freedom House suggests that the most effective movements against oppressive rule are those rooted in civic action from below rather than intervention from outside or above.

How Freedom Is Won, released in June, examines the most significant transfers of political power since the organization began categorizing countries as "free," "partly free," and "not free" in 1972. (Palace coups were excluded, as were changes in very small nations.) In each country, the report notes the strength of nonviolent civic coalitions, the social forces driving the transition, the level of violence, and whether that violence came from the state, the opposition, or both.

"The force of civic resistance was a key factor in driving 50 of 67 transitions," the study concludes, "or over 70 percent of countries where transitions began as dictatorial systems fell and/or new states arose from the disintegration of multinational states." Of those countries, the number categorized as free has grown from zero to 32, the number categorized as partly free has dropped from 25 to 14, and the number categorized as not free has fallen from 25 to four. Top-down movements for freedom were much less successful.

Furthermore, the prospects for freedom are "significantly enhanced when the opposition does not itself use violence." Indeed, whether the rebels were violent tunas out to have more of an impact than whether the state uses violence to maintain power.

One can quibble with the criteria for the categories, of course--one man's freedom is another man's "partly free"--and sometimes with the account of events. (The Iranian revolution of 1979 was indeed marked by "high violence," as the paper contends, but you can distinguish the largely nonviolent overthrow of the Shah from the fundamentalists' subsequent consolidation of power.) It's hard to deny the general thrust, though, especially when you couple Freedom House's findings with some statistics assembled by the Rutgers sociologist Kurt Schock in the 2005 book Unarmed Insurrections. By Schock's count, there were 31 predominantly nonviolent rebellions in the Second and Third Worlds from 1978 to 2001. In 23 of those revolts, the result was regime change.

Turns out that nonviolence isn't just better for liberty and self-government. Often it's good strategy as well.