Sunday, December 21, 2008

Finnermore and Floker: Constructivism in International Relations and Critics of Constructivism

Finnermoor: Talking Stock: Constructivism
Floker: Constructivism and Neoliberalism Compared

Part I: Summary

The main point of the Finnemore/Sikkink article was constructivism states that ideas are logically connected to the way the IR system works (p.394), and doesn’t provide anything more specific, so constructivism’s minimalist claims must be (and have been) supplemented in order to research the claim that ideas and identities matter (p.394). The constitution of society, the ideas and identities that make it up, are also causal in nature because the explanation of society and its parts are necessarily dependant on how those parts are made up (p.395). Constructivists then assess the evidence presented for their cause and compare it with alternative explanations of the same phenomena while searching for more evidence to further their claims and to disconfirm alternatives (p.395). Using this research design, constructivists focus on studying generalizations that are clustered around a several problems in the IR system including: the effect of global norms vs. local effects on the IR system (p.396), whether it is ideas or power that shape the IR system (p.398), and how identity and state action interact with each other (p.398). When looking at these social construction issues, constructivists also study how the social construction works by looking at ideas such as: the purposive efforts of individuals and groups to change the norms (p.400), the role of international organizations and law in changing social norms (p.401), what effect a group of experts with specialized knowledge can have on norms (p.402), the role of speech, argument and persuasion in changing social norms (p. 403), and what the structural effects are (p.403). Although this research may follow the lines of other, similar research in comparative politics, constructivists (and constructivist research) are not comparative because constructivism has a theoretical identity, whereas comparative studies work to solve problems without a theoretical identity (p. 404).

The main argument in the Sterling-Folker article is that constructivism is not anything new; it is a liberal paradigm and is related to neoliberal institutionalism (p. 98). The failure of constructivism to offer anything new is because its theoretical causal logic uses the same logic as is applied in almost all variants of liberal IR theory (p. 98). This is functional institutional theorizing, which is characterized by assuming pre-given interests, outside of social interaction, of the agents in a world where the function of institutions and social practices is to achieve particular goals for the agents involved based on their collective interests (p.105). Functionalist explanations are concerned with preference formation (p. 101): the particular policies and practices that an actor might adopt in pursuit of their interests, and how the institutions meet those preferences (p. 102). Constructivism follows this logic because they assume the institutions and social norms exists because they fill a collective need of the agents, and therefore the structures that exist must fill those needs or they will change (p.108). Additionally, constructivism uses functionalist causal logic because it focuses on what the institution does – which is a preference rather than an interest (p. 105). By using the functionalist causal arguments, one of the main tenets of constructivism – the importance of interests – is not present because the focus is instead on preferences (p.108). Neoliberal institutionalism and constructivism both state that changing perceptions and beliefs (in constructivist terms: identity and interests) will make cooperation more likely, and that in order to create these changes, the actors must interact so that various institutions can play a part in changing social norms (p.109). Therefore, because of their theoretical similarities and consistent underlying logic, constructivism is not a new paradigm, but rather a liberal/critical theory paradigm come to life again (p. 99).

Part II: Analysis

In the Finnemore article, there is a range of theories that can be called constructivism (p. 392). In order to establish a baseline for comparison, the author uses the features common to all the descriptions of constructivism (p. 392 et al). The author recognizes that constructivism is a framework for thinking and that other sources must be used to make constructivism work for analyzing content and predictive features (p. 393). With this understanding, the author compiles the articles together and does not separate out who uses what additional information, beyond constructivism, to make their claims. It may be that constructivism is compatible with all other theories in the sense that constructivism provides a framework for seeing the world, and the other theories provide the interpretation. Which means that there is no need to falsify all other predictive theories – which is what some of the research cited does. If the theories are all tools, rather than competing tools, then some of the argument goes away.

In the Sterling-Folker article, it’s interesting that she acknowledges differences in the theoretical scheme between constructivism and others (p. 100), but simply relegates them to being “small distinctions (p.100).” Acknowledging that they are different, means that they are different. Functionalist arguments focus on preference formation, rather than interest and identity formation (constructivism) (p. 101). There is even a listing of differences (p. 102). I think this is a case where simply because the results are similar, and because constructivism isn’t clearly defined, someone assumes the two theories have the same underlying theoretical basis, when it is clear that there are differences. Additionally, there are no instances of constructivism that falsifies this claim, and there is no claim that the author has taken into account all constructivist writings. So are the falsifying instances an issue?

No comments: