Norms, Behavioral Compliance, and Status Attribution in International Politics

Abstract

Extant work on status attribution has largely focused on major powers or state capabilities as key explanatory factors driving these social processes and suggests that status considerations increase conflicts between states. We argue for a more comprehensive approach to status attribution that considers international norms as another major factor that is weighed in the attribution process. We contend that states (policymakers) evaluate one another not only on the basis of economic and military capabilities but also on the extent to which there is behavioral conformance with normative expectations and reward one another dependent upon whether these expectations are met. However, this attribution of status is dependent upon the level of contestation pertaining to that norm. Using a data set that assesses consistency with six different norms (resource transference, multilateralism, economic liberalism, democratic governance, respect for human rights, and peaceful dispute resolution), we find that status attribution is associated with norm-consistent behavior but only when these norms are uncontested at the global level.

KEYWORDS:

• international norms
• norm contestation
• status attribution
• status seeking

Notes

¹ Motivations behind status concerns appear to range from material through ideational interests. The extant literature is in agreement about both the complex range of motives and the difficulties involved in disentangling them (for example, Dafoe, Renshon, and Huth 2014).

² For reviews of this literature, see Dafoe, Renshon, and Huth (2014); Paul, Larson, and Wohlforth (2014); and Volgy, Miller, Cramer, Hauser, and Bezerra (2013).

³ Objective approaches have been typical in sociology (Olzak and Tsutsui 1998), studies of voting behavior (Brodie 1996), and in some international relations literature (Galtung 1964; Maoz 2011).

⁴ Sylvan, Graff, and Pugliese (1998) argue that status considerations involve consequences including rights, responsibilities, and benefits, and none of those consequences is likely without public recognition of rankings.

⁵ The definition underscores a difference with SIT theory, which emphasizes perceived group membership and the consequences of those perceptions. We assume that issues about group membership (what constitutes a state) have been resolved, and what matters more is relative position in the status hierarchy within the group.

⁶ For a review of large-N published work on community based status attribution, see Rhamey and Early (2013) and Volgy et al. (2013).

⁷ For two exceptions, see Rhamey and Early (2013) and Bezerra et al. (2015).

⁸ However, as we note in the following, universal recognition of a norm simply means that it has reached past some threshold of acceptance; it does not mean that it is no longer contested or that there will not be substantial variation in behavioral consistency with the norm.

⁹ The “norm cascade” approach identifies key stages in the norm life cycle (Finnemore and Sikkink 1998). A second theory suggests that a norm is not formed until a shock to the system causes enough actors to create new practices that challenge the status quo (Goertz 2003). A third approach emphasizes a continuous cycle of conflict between existing normative structures and new prescriptions, with groups eventually embracing new norms (Sandholtz 2008).

¹⁰ For a similar argument regarding the need for a critical mass, see Marwell and Oliver (1993), Risse, Ropp, and Sikkink (1999), and Studlar and McAllister (2002).

¹¹ See also Checkel (2005), Ku and Diehl (2006), and Risse, Ropp, and Sikkink (2013) on institutionalization.

¹² Consistent with the literature, we are differentiating between norm contestation versus behavioral conformance to the norm (for example, Simmons 2009, chapter 3).

¹³ For an analysis of the variety of domestic costs to states, see Simmons (2009).

¹⁴ As elaborated in online Appendix 1, foundational principles are determined according to a general reading of the literature on each norm—for example, norms work on human rights focuses on physical integrity rights as the base requirement for individual or social rights. We derive these foundational principles from preexisting work, which investigates shared perceptions, primary norms documents, and existing empirical analyses. For a systematic analysis of the complexity of human rights norm contestation, regarding both foundational and procedural elements, see Simmons (2009: chapter 3).

¹⁵ This is due to a combination of factors, including the pace at which material capabilities change, institutionalized bureaucratic status ranking practices, and the time it takes for foreign policy bureaucracies to move from changes in social comparisons to public manifestation of status rankings.

¹⁶ An example of this kind of conflict can be found in Asia, where regional or local expectations pertaining to rights are communal in nature, while global expectations are more individually oriented (Sen 1996). Diverse regional conditions (high levels of integration in Europe; ongoing conflicts in the Middle East) are also likely to impact assessments of states located in those geopolitical spaces.

¹⁷ During the Cold War both superpowers actively worked to increase status for their allies. The PRC still engages in a successful, systematic campaign to minimize the status of Taiwan (responding swiftly to even small gestures, see Wright and Eishen 2014), even though Taiwan scores high on typical measures of economic and military capabilities.

¹⁸ We acknowledge the two sides of the foreign policy substitutability issue (for example, Most and Starr 1984): One policy can serve multiple objectives, while several policies may serve the same objective.

¹⁹ For an analysis of social creativity versus norm compliant behavior, see Bezerra et al. (2015).

²⁰ Some norms, such as the “responsibility to protect” or first use of nuclear weapons, have limited applicability either to all states or to the broad range of activities in which states engage.

²¹ For the complexity of norm contestation, see Clark (2005, chapter 10).

²² We asked numerous scholars, both students of international politics and norm-based scholars, if they would agree or disagree with our classification of the six norms. None of them disagreed with our classification, since there is broad agreement that the first three are relatively uncontested and the second three broadly contested, based on both structural conflicts and agent-based (major power) resistance to them.

²³ Online Appendix 3 delineates appropriate sources, references, and manipulations for all variables.

²⁴ According to Organization for Economic Cooperation and Development (OECD) statistics, the vast majority of nonmilitary aid occurs in the form of bilateral assistance (OECD, 2012).

²⁵ Data on aid donors are drawn from Tierney et al. (2011).

²⁶ We created this measure to reflect a variable that was restrictive enough to highlight states that were active traders without being so restrictive that only the wealthiest of states would be included. Data on trade are drawn from Barbieri and Keshk (2012).

²⁷ Data on IGO membership are drawn from Pevehouse, Nordstrom, and Warnke (2004).

²⁸ Data on Polity scores are drawn from Marshall and Jaggers (2012).

²⁹ Alternative measurement strategies were tested with similar outcomes, as discussed in the following.

³⁰ These include freedom from murder, freedom from torture, freedom from “being disappeared,” and freedom from political imprisonment. PTS data are drawn from Gibney, Cornett, Wood, and Haschke (2014).

³¹ See the following robustness discussion for alternative measurement choices.

³² Data on MIDs are drawn from Ghosn, Palmer, and Bremer (2004).

³³ We control for regional variation by creating a number of dummy variables corresponding to geopolitical regions (identified in Online Appendix 3). Regional assessments were modified from previous work on regional designations pertaining to status (Cline, Rhamey, Henshaw, Sedziaka, Tandon, and Volgy 2011).

³⁴ Data on embassies are drawn from Rhamey, Cline, Thorne, Cramer, and Volgy (2013).

³⁵ We attempted to create a number of additional status measures, including elections to positions in the UN General Assembly and its other organs, and in other global organizations. Unfortunately, these measures are tainted by both geographical and political quotas and “rotational” requirements, and the results do not reflect status hierarchical rankings well, even in five-year increments.

³⁶ We are also cognizant of the fact that some of the “noise” in the measure is created by some reciprocity between states in creating diplomatic relations. We have sought to minimize this contamination to the extent possible by excluding from the data set minimal reciprocity (diplomatic representation that is not in the state’s capital, and does not involve ambassadorial level staffing. Furthermore, since most of such reciprocity (a network-based measure of reciprocity shows that at least one-third of such linkages are not reciprocated) is at the regional level (Volgy et al. 2013), we use regional controls in our model.

³⁷ We exclude states with populations under 200,000 as of the year 2010.

³⁸ Online Appendix 2 compares the rankings of the top 50 states, and we also graph rankings for major powers and regional powers, comparing the 1980 rankings with those in 2010.

³⁹ Given the nature of our data, a single lag represents a 5-year time differential between the independent and the dependent variable, providing sufficient time for the potential effects of the variables in question. Lagging by more than one time period, we believe, creates too long a time frame, and may provide substantial errors, as other factors can arise across an entire decade. However, when we lag the independent variables by two periods (10 years), the primary relationships continue to hold, with one exception: Peaceful dispute resolution gains significance at the .05 level.

⁴⁰ Online Appendix 4 provides descriptive statistics for each measure.

⁴¹ Data for all the variables are available from 1975–2010, except for: IGOs (1975–2005); Human Rights (1980–2010).

⁴² We have also explored utilizing squared and cubed versions of the time counter in order to account for potential nonlinearity in the model. The inclusion of these additional time variables does not change our findings.

⁴³ Additional analyses using AIC/BIC statistics confirm this improvement over the base model.

⁴⁴ The lack of significant findings for uncontested norms does not appear to be due to multicollinearity problems. While there is an expected, significant relationship between democratic governance and human rights practices, the relationship between the other two variables averages correlations at .1. Individual VIF and tolerance estimates for the integrated model, furthermore, do not achieve scores higher than 3.46 or lower than 0.29 respectively, while the mean VIF is 1.88. By each measure, the model falls below broadly accepted standards used for identifying multicollinearity.

⁴⁵ Alternatively, we link the relationship between foreign policy activism and conferred diplomatic representation to assess a generic instrumental relationship that may be driving the sending of high-level diplomatic infrastructure. We conducted a preliminary test of this notion, using the Integrated Data for Event Analysis database (Bond et al. 2003) and lagging all events sent by states. The relationship for 2005–2010, as an illustration, indicates that 67% of the variation in diplomatic representation is unexplained by the extent of all forms of recorded activity engaged by a state.

⁴⁶ This was determined by analyzing dyadic trade (Barbieri and Keshk 2012) and diplomatic representation data for two periods, 1980 and 2000. The focal states are those that receive the value of one on the economic liberalism variable. The trade data was then used to determine a state’s trading partners (a state whose trade value met or exceeded the average trade amount conducted by the state in question). Change in diplomatic representation from 1980–1985 and 2000–2005 was compared against states that were and were not “trading partners.” In 1980, a net of two diplomatic representations was gained by economically liberal states from trading partners compared to a net of 22 from nontrading partners; in 2000, only a net of two diplomatic representations was gained by economically liberal states from trading partners, compared to a net of 77 from nontrading partners.

⁴⁷ For instance, President Obama publicly noted Russia’s ascension into the WTO (adherence to economic liberalism) as warranting recognition and respect (Obama 2014).

⁴⁸ Two primary exceptions are Larson and Shevchenko (2010) and Rhamey and Early (2013).