2. Analytical Framework
2.1 Constructivism
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
2. Analytical Framework
2.1 Constructivism
Over the course of the past sixty years, Cross-Strait relations have been the subject of a wide variety of often contradicting interpretations and definitions by the leaders on all sides as to the nature this relationship was supposed to be of. Like a recurring theme, each time when there was a change in identities this had an impact on the way in which the relationship between the two sides was being constructed by leaders, government officials and other agents across the Strait, in particular in Taiwan.
Therefore, when analyzing this contention over intersubjective meanings of “China,”
“Taiwan,” as well as their relation to one another, it seems to make sense to start with a framework that does not limit its focus on structures or organizations, but one that instead puts emphasis on the people that are primarily responsible for conducting this relationship. One of the basic concerns of constructivism is to analyze the way in which human beings, as social beings, interact with each other and how we, through our interactions, construct the world we live in. By implication, relationships between states in the international arena can be analyzed as a system of social constructions and arrangements. In these social worlds deeds, acts, and words that shape these relationships become the matter of analysis for “[t]hese are all that facts are” (ONUF 1989: 36).
Constructivism is now often regarded as a third approach to the field of International Relations6 and has either been granted an equal standing next to realism and liberalism (see KUBÁLKOVÁ 2001: 4, WEBER 2010: 62) or it has been described as a
“middle ground” between positivist and post-positivist epistemologies (ADLER 1997;
see also CHECKEL 1998: 327). Quite a few scholars have pointed out, however, that the definition of constructivism has become blurred over time and that under its many proponents there are actually quite different understandings as to what constructivism as a way of analyzing international relations is supposed to entail (KLOTZ / LYNCH
2007: 4; KUBÁLKOVÁ et al. 2001; MO 2002; ZEHFUSS 2002).
6 In this thesis I follow the convention of using the Upper Case for IR as an academic field of study and the lower case when referring to its subject.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Generally regarded as the mainstream of constructivism are the writings by Alexander Wendt (especially WENDT 1992, WENDT 1999).7 In his work, Wendt has questioned the prevalent realist concept of anarchy in the study of International Relations and advocated a new focus on state practices. He criticized neorealists and neoliberals for what he called their self-imposed rationalist constraints as well as their reification of structures, which these previous theories took as a natural given.
Instead, he introduced the concepts of state identity and state interests, which together he said are responsible for the way states interact with one another, and asserted that it was necessary to examine the process of how states construct their identities.
According to Wendt, the behavior of states in this process might be either conflictual, cooperative, or change from one to the other over time, resulting in different “cultures of anarchy,” which he elaborated on in his major work Social Theory of International Politics. Wendt argued that through interaction states do not only construct their respective identities and interests, but when these result in relatively stable structures, institutions are created, which he defined as “fundamentally cognitive entities that do not exist apart from actors' ideas about how the world works” (WENDT 1992: 399). In turn, the way in which states act within these institutions shape their identities and interests, therefore making both levels mutually constitutive. However, in order for his framework to function, Wendt, like his realist and liberal counterparts, had to hold on to the idea of state-centrism, that is, seeing states as the most important agents or decision-makers in international relations (therefore the title of his article “Anarchy is What States Make of It” [WENDT 1992; my emphasis]). He defended this view by saying that if the authors of practices, identities, interests etc., that is the states, were to be forgotten, then they could not be held accountable for their actions. However, it becomes clear that holding on to a state-centric view of international relations necessarily neglects domestic influences on foreign policy behavior (see SMITH 2001:
45). In fact, for Wendt the role of domestic politics in shaping state identity is external or precedes a state's international interactions which is the reason why he excludes them from his framework. For a case such as Taiwan, where the issue of identity politics on the domestic level directly relate to the problem of the state's standing in
7 See, for example, the discussion by PETTMAN (2000). For the fact that this still holds true today, see Weber's recently published introduction to IR studies, which in its discussion of constructivism limits itself to the Wendtian school and merely glosses over other constructivist scholars (WEBER 2010: 59-80).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
the international arena, that is, where the consistency of construction is often challenged domestically or essentially dependent on what party is in power, Wendt's take on constructivism seems to leave out a very vital part that is necessary for the way in which Taiwan interacts with other states, especially its big neighbor the People's Republic of China.8
Nicholas Onuf, who first came up with the term constructivism, and whose school of thought is going to be one of the main sources for the framework used in this thesis, has quite a different definition of constructivist ontology. For him, constructivism was not envisaged as a new paradigm to replace old ones but rather as a theoretical framework for the analysis of any world of social relations, including the one of International Relations, a field that Onuf saw in disarray and that he wanted to reconstruct (ONUF 1989: 1-31; ONUF 1998: 58; ONUF 2002: 120; 135-137). Voicing his opposition to the prevalent notion that international relations are defined by an environment of anarchy,9 he suggested to shift the focus on two more general properties of political society: rules and rule. Rules guide but do not determine human conduct by giving social meaning to political society. Rule, on the other hand, results when these rules cause an unequal distribution of advantages (ONUF 1989: 21-22).
Both properties are linked to each other by agents' use of language or their performance of social acts, called speech acts, that they perform in order to achieve their respective goals. The phenomena of both rule and rules can be seen as representations of the macro- and microlevels often found in social science scholarship. Their linkage is similar to the constructivist premise that people and society construct each other through recurrent practices in that neither of them has primacy over the other (ONUF 1989: 36-41; 58).10
This rule-based constructivism as initially conceptualized by Onuf and further developed by the Miami IR Group of scholars (cf. KUBÁLKOVÁ et al. 1998; KUBÁLKOVÁ
2001) is going to be the main framework for this thesis. The theory's emphasis on
8 Wendt's understanding of identity has been criticized before by ZEHFUSS (2002: 38-93) in a different thematic context. (See also ZEHFUSS 2001.)
9 Onuf's challenge to this concept has been conceived of as “a sophisticated attack” that “threatens the very foundations of the Neorealist framework” by some mainstream IR scholars (see BUZAN et al. 1993: 5).
10 The mutual constitution of these two levels is a pivotal ontological pillar of constructivism for Onuf as well as for Wendt (at least in his early writings), as both of them draw on Anthony Giddens' theory of structuration, which Onuf has described as a “constructivist social theory” (ONUF 1989:
58).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
agents' social construction of relationships via the use of speech acts seems to be more promising in gaining a deeper understanding of the Cross-Strait situation than would, say, the Wendtian constructivism, which has the notion of a state-as-actor at its center that fails to include domestic political influences on a state's foreign policy behavior.
After all, domestic politics seem to be an obvious factor for the changes in the way that different agents in Taiwan have defined their relationship to China over the past decades and as will be examined in later chapters. Suffice to say at this point that each time when identity or definitions of the respective “in-groups” and “out-groups” in this relationship shifted, this had an according impact on the choices that agents across the Strait could or wanted to make with regard to each other. In turn, these identities gradually became institutionalized, changing the environment in which the agents acted and continue to act.
Figure 1 is a simplified visual conceptualization of rule-based constructivist ontology.
The mutual constitution of agents and structures (here replaced by the term institutions) is the premise at the center of this framework. It indicates that neither of the two is taking precedence over the other. While institutions make people into
Figure 1: “Thinking Like a Constructivist” (adapted from KUBÁLKOVÁ 2001: 65)
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
agents, the way in which agents act within these institutions has defining effects on the latter. The course of actions of agents can be derived from the goals that they have (material ones may or may not be among them) and which they will try to achieve rationally while they are limited by certain constraints such as the (non)availability of information and resources as well as by the actions of other agents who may pursue diametrically opposed goals. Central for constructivism are rules which govern the relationship between agents and institutions. These rules come into existence by the use of language, or more precisely, speech acts, that are used by agents to affect existing rules or create new ones. The repetition of certain speech acts influences rules in the way that they either support them or are aimed at breaking them. At the same time, rules thus being established, supported or broken in turn reveal who the active participants in society, that is, its agents are. The relationship between rules and agents is therefore also one of mutual constitution. Finally, the frequent reaction to rules by agents leads to practices that have intended as well as unintended consequences. Together they in turn have an impact on the institutions that all agents act in, thus concluding the circle of mutual constitution.
It is important to note at this point that Onuf's framework does not function without logocentrism. Although language and its use to construct social reality are given a prominent role within his framework, in opposition to poststructuralist and in particular deconstructivist schools of thought, constructivism does not question that there exists a natural or material reality to individuals as biological beings. In other words, constructivism takes the separateness of “words” and the “real world” as a given. This epistemological premise is reflected in the assumption that “[s]ociety constructs human beings out of the raw materials of nature, whether inner nature or […] the outer nature of their material circumstances” (ONUF 1989: 46). In fact, for constructivists the relation of rules to resources entails “at least some control over material conditions” (ONUF 1989: 60). This tendency to logocentrism has, of course, been criticized by more poststructuralist-oriented scholars. For example, ZEHFUSS
(2002: 195) concluded her thorough critique of Onuf's approach by stating:
It seems awkward […] to present material reality as the explanation for the limitations of our constructions. This sits uneasily with the notion that there is something behind them, so to speak.
Even if there was, it could never matter to us other than within our constructions. Even if material reality imposed a limit, what is significant is how we conceptualise this limit. […] In remaining abstract and aloof from particular constructions of reality, Onuf seems to exclude the problematic of the political character of constructions.”
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Zehfuss followed up on this deconstruction with a reading of the writings by Jacques Derrida with a special emphasis on what she calls the “politics of reality” in his writings (see ZEHFUSS 2002: 236-249). In his critical analysis, Derrida has questioned the logocentrism of Western thought in general and asserted that a reality as a separate
“world” of its own, even if it did indeed exist, would never be accessible to us apart from our representations of that reality. For whatever is conceptualized as “real” is necessarily also an effect of representation or, in other words, there is no reality outside of a textual level.
Onuf was well aware of deconstructivist objections against his framework. In World of Our Making he stated that “[w]hat cannot be reconciled with constructivism is deconstruction, at least when that practice is carried very far” due to the fact that deconstructivism is also logocentric “in its own narrow way” (ONUF 1989: 42; see also ONUF 2002: 126). However, in pointing out that Onuf's conceptualization of reality is in itself also a construction, criticisms such as the one by Zehfuss that are inspired by poststructuralist insights, provide valuable contributions in that they call on observers to heighten their own sense of self-awareness and involvement in the process of construction while discussing a certain subject matter. On the other hand it also becomes clear that there are in effect two different world views or epistemologies colliding with each other, and which to some degree will always remain incompatible.
Furthermore, ZEHFUSS (2002: 260) herself has pointed out that Derridean thought may not provide “security” or what passes as scientific knowledge in the field of IR. She admitted that deconstruction, by subverting texts, can not solve the metaphysical problems it addresses and may only offer a possibility to analyze these problems on its own terms (ZEHFUSS 2002: 204-205).11 In any case, problems related to epistemology will not be solved here, so that for the moment I can but acknowledge their existence.12 Instead, I am going to follow Onuf's “ontological turn” (ONUF 2002:
138) in that I will elaborate on three important properties of the constructivist ontology in the next sections and for the main part of this thesis: rules, speech acts, and agents.
11 However, using her case study she was able to show convincingly that material or “real” constraints do not necessarily have to be a crucial limit to the construction of reality by agents.
12 For a more recent discussion of the epistemological divide that continues to plague the field see KLOTZ / PRAKASH (2008: 1-7).
‧
國立 政 治 大 學