A Time to Jettison the Ambiguity : Parsimony, Inter-subjectivity and Selection Bias
HAN DAEHEE(I33011)
2014-03-31
I. Introduction
Empting our intellectual mind to the complexity of the world is difficult. Nevertheless, IR realists have recently sought to become mid-range theorists, relaxing their own core assumptions and trying to explain the complex world with complex independent variables. Clearly there remain plenty of methodological and theoretical issues. At least these kinds of ambiguous attitudes are not suitable in order for us to familiarize with the perspective from the positivism. That’s because positivism which observes and finds out causal effects in a social scientific manner originally tries to jettison the ambiguity of terminologies together and explain the complexity with the simplest methodologies under an assumption of observable implication. In this respect, one main task of this essay is to examine MA thesis example 1 and perform a methodological critique from the positivism. Simultaneously, another important duty of this essay also tries to discuss issues such as debates of parsimony, inter-subjectivity problems as valid independent variables, and selection bias.
To discuss these three methodological issues, this methodological critique will be done by the following three steps: restating the author’s major research question and theories (or hypotheses) in “if, then” type of generalization, assessing the definitions the author had adopted for his independent and dependent variables and whether they are adequately used, and evaluating the procedures used in measuring variables and the tests devised to show causal relationship between the author’s IVs and DV. Yet, critique is not only for the criticism in the void vacuum. Based on basic outcomes of the critique, this essay also tries to demonstrate the research design of this example thesis might have been enhanced in a much more scientific and better way for seeking out roles of International Relations.
II. Debates of Parsimony
Here is a hypothesis for MA Thesis Example 1.
Hypothesis> If influences of hegemonic states, spill-over effects of beliefs and norms by negotiation and interactions, and individual’s role to prevent nuclear proliferation increase, the possibility of nuclear rollback(or reversal) will be likely to increase.
That is to say, nuclear rollback equation can be numerically expressed as F(Possibility of NR)=a*(influences of hegemonic states)+b*(effects of beliefs and norms)+c*(individual’s role).
That is to say, the underlying hypothesis of this thesis highly wants to test if nuclear rollback may be decided by the three factors mentioned above. Yet, the problem is whether independent variables to explain causal effects with dependent variable at least seem extremely problematic in theoretical perspectives as well as methodological perspectives.
According to the positivism, that equation and hypothesis especially violates ‘parsimony condition’ among the four conditions to become a good theory or hypothesis: testability, generalization, observable implications, and parsimony. What’s most problematic is that by seeking a mid-range or ambiguous theoretical framework between neo-realism and constructivism[1], an author of this thesis did combine objective effects of material capabilities represented by ‘hegemonic states’ with subjective effects of beliefs and norms and individual’s role, even though author admitted that influences and roles of hegemonic states which can be linked to ‘the relative distribution of power’ in the international system might be the most decisive factors in analyzing the state behavior to prevent nuclear proliferation. Rather, Although a, the magnitude of influences of hegemonic states, is regarded as higher than b (the magnitude of the effects of beliefs and norms) and c (the magnitude of the individual’s role), the author of this thesis asserts that being affected by hegemonic countries, effects of beliefs and norms and individual’s role can play a significant role in leading to the nuclear rollback.
Yet, a theory should be parsimonious, which means that a theory cannot explain everything about lots of paths of phenomena of international relations. A theory must effectively, powerfully, and economically explain the complex phenomena as simple as possible with reduced amount of independent variables. Then, extravagant dissipation of independent variables combining objective factors with subjective factors makes the theoretical model of IR look hard to understand. Andrew Moravcsik had harshly criticized degeneration of ambiguous methodological frameworks to seek mid-range theories or minimal realism even in ‘realist’ camp. Even he harshly considered Jack Snyder, Joseph Grieco, Fareed Zakaria, Randall Schweller, Stephen Van Evera, and Charles Glaser as minimal realists who want to look ‘realist’, not focusing the ‘material capability’, the most important and decisive independent variable to explain the real world but combining independent variables related to ‘material capability’ with other independent variables related to ‘institutional’ and ‘epistemic’ dimensions. In short, at least according to Andrew Moravcsik’s perspective, realist-looking and minimal realists mentioned above tried to smuggle other competing theoretical frameworks of IR. It is quite natural that he raised a question whether anybody is still a realist. In fact, what he really pinpointed was not lied in criticizing ‘diluting realism’ but logically requiring a clarification against ‘complex and inefficient dissipation of independent variables in building up the research framework’.[2]
Logically, if influences of hegemonic states should be considered as the most decisive, the rest of other independent variables are at best complementary or supplementary in explaining state behavior of nuclear rollback. It is quite dubious whether the author of this thesis overestimates the value of b(the magnitude of beliefs and norms) and c(the magnitude of individual’s role). The problem is that the value of b and c cannot be measured nor operational in terms of positivist approach. That is, clarification of definitions of IVs should be clearly set.
III. Inter-subjectivity in the Scientific Research?
As mentioned earlier in KKV, science is not a panacea, which means that activities including human interactions and emotions are difficult to answer with scientific methods. At the same time, KKV gave students of scientific methodologies an additional advice that culture, intentions, motivations, norms are hard to be operationalized. The most serious flaw of the procedures in measuring variables in this thesis is that in constructing causal relationships between IVs and DV, the author of this thesis had no choice but to depend on ‘descriptive’ manner to look directly into ‘what happened historically’ sometimes and ‘partial interpretative’ manner to depend on interpretation or partial speculation of his or her own in analyzing why and how norms dominated in the individual’s role and why and how perception shift happened in terms of nuclear rollback.
For example, his procedures of performing a case analysis is a continuation of explaining historical events related to nuclear rollback and complementary comments regarding the interpretation of ‘norms’, ‘beliefs’, or ‘perception shift’. When the author interprets about the roles of them in terms of nuclear rollback in each case analysis, he or she always avoids measuring the variation of them in a scientific and accurate way like these statements:
l I think the Western Society’s beliefs about the importance of the negotiation and their decision making process based on the principle of not militarily intervening in South Africa played a large role in its nuclear rollback. (p. 43, 3-2. South Africa’s Nuclear Rollback: Imposing sanctions controlled by the hegemonic power)
l I think that at the beginning of developing nuclear weapon, they wanted to use if leverage in negotiation with the West, and blackmail the West. But after a series of imposed sanctions and the effort of diplomacy by the West made South African political leaders speculate on gain and loss of possessing nuclear weapons. That is South Africa realized the norm in the international community. (p. 47)
Why did this happen? This answer is very simple because first, fundamentally the author didn’t clarify basic characteristics of IVs in scientific research. Since the norms and beliefs as IVs were directly intervened in the equation mentioned in the previous chapter of this essay, the author seems to have chosen ‘descriptive’ way as the second best policy to continue his or her research.
In addition, that’s because it is not quite clear that some definitions for measuring IVs and DV are adequately used. Especially for the equation mentioned in the previous chapter, the definition regarding the ‘Hegemony’ should be set whether this concept can be only decided by the hierarchical international system and relative distribution of material capabilities or components of soft power to increase bargaining power in terms of negotiation and interaction should be included.
At the same time, the author seemingly confused between the definition of the ‘norm’ and that of ‘perception’. According to the regime theory suggested by Stephen Krasner, ‘norm’ can be defined as a ‘standard’ of behavior defined in terms of rights and obligations. Yet, the author’s confusion can be exemplified like this;
l This is a kind of belief and norms in the international society where small and weak nations like North Korea need nuclear weapons to acquire anything they want and to be against the US.
In this statement, ‘belief and norms’ should be at best altered into a sort of ‘perception’ of North Korea viewed from the international society. That’s because statements that small and weak nations like North Korea need nuclear weapons to acquire anything they want and to be against the US is not a ‘standard’ of behavior that Stephen Krasner defined.[3]
Because of ‘inter-subjectivity’ that includes components of dynamic variations and mutual empathy, there remain hot debates over ‘norms’ and ‘belief’ can be valid independent variables to explain some phenomena of IR in a better way. Even though Stephen Krasner ignited debates of regime theory by recognizing components of the regime as ‘intervening variables’ to be conceived of affecting related behavior and outcomes, Traditional structuralists like Susan Strange, and even Kenneth Waltz deny because regimes do not represent direct connection between changes in basic causal factors and changes in behavior and outcomes. Anyhow, as Susan Strange mentioned, IVs that have inter-subjective components make researchers have difficulty in finding out ‘clear’ paths such as ‘observable implications’ for searching for a desirable research design.
IV. Selection Bias?
In order to explain the most decisive IVs of nuclear rollback, his or her research procedures are laid under the qualitative Small-N case analysis including Ukraine, Libya, and South Africa. In KKV, Since Porter was interested in the sources of what he called “competitive advantage” for contemporary industries and firms, He selected successful cases such as ones that already hold competitive advantage successfully in a range of such industries, or in the case of Korea and Singapore, show signs of an improving ability to do so. To explore the puzzle that interested him, Poter intentionally selected on his dependent variable, making his observed dependent variable nearly constant. As a result, any attempts by Porter, or anyone else using these data will have no choice but to produce seriously biased causal effects. At the same time, KKV also points out that in designing a research to find out causal effects between American investment and internal violence, even though there are other observations that illustrate the other combinations but they are “conveniently” left out.
Thus, in doing a small-n case analysis research, lots of cautions should be paid because researchers can encounter a problem of “many variables, small number of cases”. Yet among small number of cases, there is something uncontrollable due to the limitation to explain the complexity of the world. As the author said, another task of this research was to combine components of neo-realism and neo-liberalism with constructivism. The problem is that only three ‘successful’ cases chosen for this research were totally and intentionally chosen to be matched up with the hypothesis of the author. In three cases, even though motivations for holding nuclear weapons are somewhat different for each country[4], ‘interaction’, ‘roles of individuals to change his perception and beliefs’, and ‘negotiation’ through domestic and non-state actors ignited by the roles and influences of hegemonic states made Ukraine, Libya, and South Africa reverse nuclear weapons. Since cases were chosen to consider ‘dependent variable’, some failure cases related to the roles of the hegemonic states were estimated. Even though intervention of the US to North Korea still high, this case cannot be a successful case of nuclear rollback and North Korea is still assumed to hold and develop nuclear weapons. A hypothesis of this thesis does not explain North Korean issue of nuclear rollback. Therefore, three cases in this thesis are based on ‘selection bias.’ Eventually, in this kind of case analysis, as concluding remarks of KKV show, the dependent variables should have no choice but to be varied further.[5]
At the same time, in each case, fundamentally it remains a question whether individual’s role such as Gaddafi’s perception shift is autonomously reflected by the secretive negotiation and interaction to understand each other between the U.S and its alliance, the U.K and Libya. Isn’t it possible to assert that although negligible mutual understanding by coercive interaction exist, Libya also succumbed a enormous political and military influence of an hegemonic power, the U.S which dominates hierarchical distribution of relative power difference under the international system? If the path to reason is obscure and ambiguous, there remains a question whether this kind of case should be a case to support a hypothesis that If influences of hegemonic states, spill-over effects of beliefs and norms by negotiation and interactions, and individual’s role to prevent nuclear proliferation increase, the possibility of nuclear rollback (or reversal) will be likely to increase. Then, although Libya’s case is a successful case of nuclear rollback, Libya’s case is not a suitable one to analyze within mid-range hypothesis.
V. Alternatives to a Positivist Theory?
Stick to ‘objective’ and ‘material’ factor as independent variables for inevitable parsimony. Once again, a positivist theory is not a panacea. A theory cannot explain everything, but it can explain some phenomena effectively with scientific methodologies. Even though author of this thesis tries to be ambitious by combining neo-realism perspectives with constructivism, that kind of model only explains the complex world complicatedly. Why? As mentioned before, intervention of inter-subjective factors such as norms, beliefs, and perceptions make the measurement of variations of those factors ambiguous. As KKV guides, those kinds of factors are not easy to operate within the theoretical and scientific methodological framework. Therefore, this thesis might be better if an author enjoyed a intellectual courage to jettison analysis of inter-subjective factors, which means that the author should have demarcated boundaries of independent variables between ‘objective’ or ‘material’ and ‘inter-subjective’ ones. In fact, that kind of demarcation is a trade-off between effectiveness of parsimony and incompleteness to explain the international system. Finding the simplicity out of the complexity seems more valuable than searching for the complex model out of the complex world.