Brooks /CHANGING BENEFITS OF CONQUEST 649 ings by resistance fighters and a concomitant reduction in the extractable economic surplus in the conquered territory due to strikes,sabotage,and foot dragging by the defeated populace. Liberman (1996)directly confronts this conventional quagmire perspective and effectively shows that it is not nearly as compelling as many scholars currently assume. Liberman's first,essentially noncontroversial argument is that states with higher levels of economic development offer larger potential economic rewards to the conqueror; the richer a country,the more that can potentially be plundered from it.The key ques- tion then becomes whether the high levels of wealth in the most advanced countries can actually be extracted by the conqueror.If the quagmire view is correct,then these large levels of economic surplus in advanced states will not be available to the con- queror due to active resistance by the defeated populace.Liberman's key argument is that high levels of popular resistance are not a given in modern societies;rather,they vary inversely with the ruthlessness of the conqueror.He maintains that extremely ruthless conquerors,such as the Nazis in World War II,can effectively suppress popu- lar resistance and thereby are able to extract significant economic resources from the vanquished country.Liberman assesses his thesis through a detailed historical analysis of five occupations of industrial societies from the 20th century:Belgium and Luxem- bourg,1914-1918;Ruhr-Rhineland,1923-1924;the Japanese empire,1910-1945;the Nazi occupation of Western Europe,1940-1944;and the Soviet empire in Eastern Europe,1945-1989.He finds significant support for his argument that those invaders who are ruthless can suppress popular resistance and thereby make conquest pay.2 There are several reasons,however,to question Liberman's(1996)findings.Two issues in particular are worth highlighting.First,Liberman's study suffers from poten- tial selection bias in his cases.He limits his sample of cases to examples in which con- quest has occurred and"there is attempted extraction"(p.14).As a result,for a case to count in his sample,the conqueror not only has to be able to vanquish the opponent but also must establish sufficient territorial control that extraction can actually be attempted.Two key variables that affect whether a country can be successfully con- quered to this degree are(a)the extent to which the population is willing to make sacri- fices to repel the invader and (b)the level of resolve and effectiveness of the defender's military units on the battlefield.States with highly nationalistic populations willing to make significant sacrifices to reject an invader and,in turn,whose military forces ardently and effectively oppose the conqueror in battle (e.g.,Vietnam from 1946 to 1975,Britain in World War II,Afghanistan from 1979 to 1989)will be difficult to con- quer and even harder to establish sufficient territorial control for extraction to be attempted.The converse also will be true (e.g.,France in 1940). The key point is that some of the very factors that make societies easy to conquer (e.g.,lack of significant popular resistance,low levels of resolve and effectiveness on the battlefield)also will make them easier to subdue once they are occupied.Not sur- 2.To be clear,Liberman (1996)does not claim that conquest produced significant gains in all of the historical cases he examines.He maintains that conquest did not pay significant rewards in two of the cases he investigates:Belgium during World War I and the first phase of the Ruhr-Rhineland occupation.In cach of these cases,however,he argues that this ultimately can be traced to a lack ofruthlessness on the part of the conquering power.Brooks / CHANGING BENEFITS OF CONQUEST 649 ings by resistance fighters and a concomitant reduction in the extractable economic surplus in the conquered territory due to strikes, sabotage, and foot dragging by the defeated populace. Liberman (1996) directly confronts this conventional quagmire perspective and effectively shows that it is not nearly as compelling as many scholars currently assume. Liberman's first, essentially noncontroversial argument is that states with higher levels of economic development offer larger potential economic rewards to the conqueror; the richer a country, the more that can potentially be plundered from it. The key ques￾tion then becomes whether the high levels of wealth in the most advanced countries can actually be extracted by the conqueror. If the quagmire view is correct, then these large levels of economic surplus in advanced states will not be available to the con￾queror due to active resistance by the defeated populace. Liberman's key argument is that high levels of popular resistance are not a given in modern societies; rather, they vary inversely with the ruthlessness of the conqueror. He maintains that extremely ruthless conquerors, such as the Nazis in World War II, can effectively suppress popu￾lar resistance and thereby are able to extract significant economic resources from the vanquished country. Liberman assesses his thesis through a detailed historical analysis of five occupations of industrial societies from the 20th century: Belgium and Luxem￾bourg, 1914-1918; Ruhr-Rhineland, 1923-1924; the Japanese empire, 1910-1945; the Nazi occupation of Western Europe, 1940-1944; and the Soviet empire in Eastern Europe, 1945-1989. He finds significant support for his argument that those invaders who are ruthless can suppress popular resistance and thereby make conquest pay.2 There are several reasons, however, to question Liberman's (1996) findings. Two issues in particular are worth highlighting. First, Liberman's study suffers from poten￾tial selection bias in his cases. He limits his sample of cases to examples in which con￾quest has occurred and "there is attempted extraction" (p. 14). As a result, for a case to count in his sample, the conqueror not only has to be able to vanquish the opponent but also must establish sufficient territorial control that extraction can actually be attempted. Two key variables that affect whether a country can be successfully con￾quered to this degree are (a) the extent to which the population is willing to make sacri￾fices to repel the invader and (b) the level of resolve and effectiveness of the defender's military units on the battlefield. States with highly nationalistic populations willing to make significant sacrifices to reject an invader and, in turn, whose military forces ardently and effectively oppose the conqueror in battle (e.g., Vietnam from 1946 to 1975, Britain in World War II, Afghanistan from 1979 to 1989) will be difficult to con￾quer and even harder to establish sufficient territorial control for extraction to be attempted. The converse also will be true (e.g., France in 1940). The key point is that some of the very factors that make societies easy to conquer (e.g., lack of significant popular resistance, low levels of resolve and effectiveness on the battlefield) also will make them easier to subdue once they are occupied. Not sur- 2. To be clear, Liberman (1996) does not claim that conquest produced significant gains in all of the historical cases he examines. He maintains that conquest did not pay significant rewards in two of the cases he investigates: Belgium during World War I and the first phase of the Ruhr-Rhineland occupation. In each of these cases, however, he argues that this ultimately can be traced to a lack of ruthlessness on the part of the conquering power