city,let us call them 'the statists'-argue that selective state intervention allowed investment to be directed to key industries,which would not have occurred under normal market operations.The state,therefore,was responsible for the economic success of Japan,South Korea and Taiwan in the post-Second World War period.3 By contrast,neoclassical economists attribute Northeast Asian success to the free market.They acknowledge a positive role for public policies(and hence for govern- ment),but believe that the proper role of government is to facilitate market opera- tion rather than to alter it.?In explaining the 1997 financial crisis and subsequent economic decline in Northeast Asia,some neoclassical economists have been quick to link the region's problems with past state intervention.They argue that an excess- ively close relationship between state and business promoted special interests and thereby distorted the free market. I believe that the essential difference between these two perspectives lies in their different levels of analysis.The statists,like many other political scientists,take the state and its development interests as given,and perform their examination at the state-society level or above.Consequently,it is normal for them to treat the develop- mental state in Northeast Asia as a unitary actor dealing with social forces,which is thus able to manage state-business relations in the pursuit of economic develop- ment.Meanwhile,the neoclassical economists question the statist concept of the state and the state's interest.If the state is comprised of rational individuals,how can those individuals stick together,let alone pursue a common goal of industrialisation? Politicians and bureaucrats are self-interested human beings who seek to maximise their own economic wealth and political power.Rent-seeking theorists,in particular, argue that heavy state intervention necessarily leads to rent-seeking activity.0 Some statists,though,do acknowledge and even use the rent-seeking image of the state in their analyses.For example,Evans classified developing countries into three categories.At one extreme are developmental states,such as Japan,South Korea and Taiwan;at the other extreme are rent-seeking states like the former Zaire.In between,there are'intermediate'states like Brazil and India.However,these three types of state are the starting point of Evans'assessment of different levels of economic performance.The aim of his study is not to explain what gives rise to different types of state and how they may change.The developmental state as a unitary actor and its development interests continue to be assumed. It is not my intention to address the state-versus-market debate in detail,but the analytical differences concerning the state do raise a crucial question regarding theConflict, Security & Development 2:1 2002 8 city, let us call them ‘the statists’—argue that selective state intervention allowed investment to be directed to key industries, which would not have occurred under normal market operations. The state, therefore, was responsible for the economic success of Japan, South Korea and Taiwan in the post-Second World War period.5 By contrast, neoclassical economists attribute Northeast Asian success to the free market.6 They acknowledge a positive role for public policies (and hence for govern￾ment), but believe that the proper role of government is to facilitate market opera￾tion rather than to alter it.7 In explaining the 1997 financial crisis and subsequent economic decline in Northeast Asia, some neoclassical economists have been quick to link the region’s problems with past state intervention. They argue that an excess￾ively close relationship between state and business promoted special interests and thereby distorted the free market.8 I believe that the essential difference between these two perspectives lies in their different levels of analysis. The statists, like many other political scientists, take the state and its development interests as given, and perform their examination at the state–society level or above. Consequently, it is normal for them to treat the develop￾mental state in Northeast Asia as a unitary actor dealing with social forces, which is thus able to manage state–business relations in the pursuit of economic develop￾ment.9 Meanwhile, the neoclassical economists question the statist concept of the state and the state’s interest. If the state is comprised of rational individuals, how can those individuals stick together, let alone pursue a common goal of industrialisation? Politicians and bureaucrats are self-interested human beings who seek to maximise their own economic wealth and political power. Rent-seeking theorists, in particular, argue that heavy state intervention necessarily leads to rent-seeking activity.10 Some statists, though, do acknowledge and even use the rent-seeking image of the state in their analyses. For example, Evans classified developing countries into three categories.11 At one extreme are developmental states, such as Japan, South Korea and Taiwan; at the other extreme are rent-seeking states like the former Zaire. In between, there are ‘intermediate’ states like Brazil and India. However, these three types of state are the starting point of Evans’ assessment of different levels of economic performance. The aim of his study is not to explain what gives rise to different types of state and how they may change. The developmental state as a unitary actor and its development interests continue to be assumed. It is not my intention to address the state-versus-market debate in detail, but the analytical differences concerning the state do raise a crucial question regarding the