Responding to the call: Human security INGOs and countries with a history of civil war

Abstract

When are human security international non-governmental organizations (INGOs) involved in countries with a history of civil war? Though the literature mentions the necessity of these actors in war-torn countries, not much is known about the factors which make INGO involvement more or less likely in countries with a history of civil war. We develop a theory that highlights the role of INGO safety and country salience in encouraging INGO involvement. Looking at the involvement of both development and human rights INGOs in countries with a history of civil war, we find support for our overall argument.

Keywords

INGOs civil society post conflict civil war human security

Introduction

During or in the aftermath of civil war, why do some states attract far more human security international non-governmental organizations (INGOs) than others? Take, for example, the state of Tajikistan. For the entire decade after the start in 1992of its civil war, the state never had more than four dozen INGOs active within its borders. In 2002, the British-based INGO Merlin even remarked that the area was historically ‘forgotten’ by organizations and that the arrival of new INGOs would mean ‘good news’ for human security (IRIN, 2002). Unlike Tajikistan, other countries with civil wars around the same time period had far greater numbers of human security INGOs providing necessary human rights promotion and development services within their borders. For example, after Georgia (a country less-populated than Tajikistan) experienced civil war in 1993, the country spent much of the next decade with double the number of human security organizations as in Tajikistan and had more than 100 organizations active within it in 2002.

Human security INGOs can form a critical backbone for service provision and human rights promotion after civil war (Stahn, 2001).¹ Their work has been praised as essential for human security, even leading to many Nobel Peace prizes for members of INGOs in war-affected areas. For example, after the civil war in Guatemala, human rights INGOs helped to craft legislation and reintegrate soldiers into society (Ratner and Abrams, 2001; Sikkink, 1993). Likewise, in Nepal, many organizations were active in providing a plethora of support functions in health and developmental services (Shah, 2008). Even Colin Powell, then US Secretary of State, reported, ‘NGOs are such a force multiplier for us, such an important part of our combat team’ (Powell, 2001).

Although the role of these organizations has been heralded, not much is known about the factors which lead human security-related INGOs to become involved in countries with a history of civil war. Where exactly do these organizations go? What factors, if any, attract human security INGOs to war-torn areas? Why were so many organizations involved in Georgia, for example, and so few organizations involved in Tajikistan? This question is a necessary first step in developing any theoretical understanding of the potential impact of INGOs in civil war reconstruction. Without understanding the factors that make INGOs become involved in a war-affected state, it is difficult to disentangle the potential effects of these organizations as human security providers from the factors which led to their involvement in the country in the first place (Cingranelli and Richards, 2001). Because INGOs can be such critical ‘force multipliers’ in war-affected states, answering this question is also important for INGOs themselves and other post-conflict interveners who might have an interest in the goods and services INGOs provide, such as donor foundations, military organizations, and intergovernmental agencies. Answering this question could ultimately matter for post civil war reconstruction efforts.

In this article, we build a theory of human security INGO involvement in war-torn states that is based on two concepts: (a) organizational safety and (b) international issue-salience. We argue that human security INGOs, like all organizations, are interested in maximizing the achievement of their organizational goals while minimizing the costs to themselves and their staff. As such, we argue that INGOs, even if motivated solely by principles to help a war-affected population, will be more active in states which are relatively safe for their workers and salient in the minds of their likely volunteers and donors. This safety and salience is necessary for sustained INGO operations and speaks to the space INGOs fill within a larger body of governmental and nongovernmental agents.

INGOs will be more likely to be involved in a state where renewed conflict is at a minimum; this can ensure that their operations are not interrupted by continued bloodshed, as happened recently in Sri Lanka and Afghanistan, for example (Al Jazeera, 2008; Kumarasinghe and Wijayapala, 2006). We also see INGOs as organizations that can benefit from the safety provided by others (Doppler, 1996). Because peacekeeping forces (e.g. UN Peacekeepers, peacekeepers of the African Union, etc.) can be essential in keeping the peace in countries with a history of civil war, INGOs will be more likely to be involved in countries with a history of the presence of peacekeepers. We suspect that INGOs, as actors involved in dangerous situations but typically without any military equipment themselves, tap into resources provided by more powerful governmental and intergovernmental actors, even if those resources are largely an externality to the basic mission of the more powerful actors. In arguing this, we acknowledge that peacekeeping forces are not on the ground primarily to aid in the mission of INGOs; we simply see the security they provide as critical to the safety of INGO workers, whether intentional or not (Bellamy et al., 2004; Stahn, 2001).

INGO operations can be expensive and responding to areas where there is a lot of international media attention allows organizations to respond to those states most salient in the minds of possible donors and volunteers, and those individuals to which these organizations are often beholden. INGOs may also be responding to certain countries because of attractive aid possibilities after civil wars. This would be consistent with arguments by Cooley and Ron (2002), for example, who see all NGOs as actors participating in a ‘scramble’ for donor funds. INGOs may be ‘scrambling’ to get bilateral donor funds in war-affected states; however, we do not see bilateral funds as the only motivation of INGOs. In fact, by ‘scrambling’ to countries with more aid, INGOs may be showing their motivations to utilize all resources available to help a domestic population. The importance of salience for INGOs, however, does reinforce ideas about how INGO agenda-setting can limit improvement on certain, less salient, human security issues (Bob, 2005; Carpenter, 2007). By acknowledging how international media attention and aid influence INGO operations, hopefully future policy and funding decisions can be crafted to attract INGO operations to areas that have previously been forgotten in the international press.

We test these arguments using a dataset of human security INGO involvement in areas that have experienced a civil war in the prior 10 years. Our findings largely support our basic theoretical argument of INGO behavior: safety matters to INGOs and INGOs are more likely to be in areas where security is likely to be provided. Further, INGOs are also going to salient countries with more international attention and larger donor packages. These empirical findings offer many insights to those interested in expanding the role of INGOs for human security outcomes during conflict reconstruction.

This paper proceeds as follows. First, we outline our theoretical argument for INGO involvement in post-conflict countries and situate this argument within the existing literature. We then present the empirical hypotheses that can be derived from this argument. After outlining our research design, we discuss our empirical findings and conclude with the policy and theoretical insights this project provides.

Theoretical development

For much of the existing literature, INGOs are seen as organizations that serve to connect citizens to the large global society (Boli and Thomas, 1999; Keck and Sikkink, 1998). As such, we should both see a proliferation of INGOs in areas of the world that are heavily globalized and connected to the larger world polity (Smith and Wiest, 2005) and we should see INGOs in areas of the world where they have been called to help a domestic population (Keck and Sikkink, 1998; Risse and Ropp, 1999). Unfortunately, most studies of INGOs tend to lump these two very different sets of states together, focusing on the factors that make all states more likely to see INGOs. It is no wonder that existing empirical studies have found that INGOs are more likely to flourish in areas with higher incomes, more democratic institutions, and in countries with greater trade flows (Smith and Wiest, 2005; Tsutsui and Wotipka, 2004). This would be areas where INGOs would develop organically, as part of a globalized civil society within a state.

We, however, take a much more different approach in this article; we seek to explain where INGOs are likely to go, as Keck and Sikkink (1998) would contend, in response to real or assumed calls to help a domestic population. As such, we restrict both our theoretical and our empirical focus to a set of countries where calls for help in line with human security concerns can be universally assumed to exist: countries with a history of civil war.²

Civil conflict in general is on the rise in the world; it is, by far, the most common type of political conflict. The effects of civil war on human security are profound. Because civil war makes it difficult to tell potential rebels from rebel sympathizers, civil war causes physical integrity human rights to be abused, even leading to mass killings (Krain, 1997; Poe and Tate, 1994). These devastating effects can be difficult to control even after the end of organized conflict. Civil conflict is also extremely harmful to a country’s economy, leading some to refer to civil war as ‘development in reverse’ (Collier and Hoeffler, 2004). The negative effects of civil war on a country’s economic output can be felt for some time, as has been the case in the struggling economies of Congo and Chad, for example.

With these human security problems of civil wars in mind, we see restricting our theoretical focus to only countries with a history of civil war as especially pertinent for our understanding of how INGOs respond to calls for help in line with their mission statements. Though this is theoretically a big part of the extant literature, a systematic account for the factors that lead to INGO involvement in these countries is largely missing. Empirical examination of INGO involvement in conceptually distinct sub-segments of the world is also, to our knowledge, something missing in the existing scholarship.

Our understanding of INGO behavior, however, draws heavily on key insights from the extant literature, together with insights gained from the civil war literature. Below, we first review the cross-disciplinary literature on INGOs and then discuss the existing literature on civil war and post-conflict peacekeeping. We build on these points in developing our theoretical framework for INGO involvement in post-conflict countries.

Existing studies on INGOs: safety and salience

Within the canonical literature, INGOs are viewed as actors somewhat unique from others in that they are motivated primarily by principles to help a domestic population in what that domestic population wants and not by material concerns (Keck and Sikkink, 1998: 8–9). In short, INGOs go where called by domestic populations. Then, when present within a country, they work to both empower the local population and to get other actors to support a population from abroad. This is the key dynamic in Keck and Sikkink (1998)’s ‘boomerang model.’ When INGOs respond to a local demand for help, they become part of a ‘boomerang’ from the state to the outside world and back again. In this manner, INGOs, especially advocacy INGOs, are said be ‘wildcards’ in leading to improvements in human rights and other human security outcomes (DeMars, 2005).

Other scholarship is much less optimistic about the motivations of INGOs. Some of this work focuses predominately on service provision INGOs, like development INGOs (Cooley and Ron, 2002; Edwards and Hulme, 1996; Petras, 1999). Other work tends to focus on advocacy organizations (Bob, 2005). In both literatures, however, the basic idea is the same: INGOs have personalistic motivations that lead them to respond not necessarily to where they are needed the most but to areas that are most advantageous to their personalistic motivations, whether for private monetary gains or some sort of ‘missionary’ agenda (Manji and O’Coill, 2002).

Although these two strains of literature on INGO motivations are very different, one idea underlying both general arguments is that INGOs have to be concerned with meeting the expectations of their stakeholders. As such, we should expect INGOs to desire to keep their stakeholders safe in their work; even if INGOs are all principled actors desiring only to aid a domestic population, their aid to the population depends on keeping their volunteers and professional staff safe. If, on the other hand, INGOs are motivated mainly by some sort of personalistic gain, it still can be assumed that volunteer safety is essential to getting any such gain. This very basic point has been largely missing in the extant scholarship.

Another idea reverberating through the larger INGO scholarship is the need for international attention. For the boomerang model of advocacy INGOs, on one hand, international attention is necessary for INGOs to maintain the support of the international community. This support, of course, is needed to get concessions and changes in behavior from governmental actors (Risse and Ropp, 1999). Bob (2005) and Carpenter (2007) both contend that the need for international issue salience plays a large role in the decision of INGOs to pick up certain issues for advocacy and avoid other issues that would not resonate as much with existing international norms. For development or other service INGOs, on the other hand, international attention and donor funds are also necessary to fund the projects of INGOs and provide much-needed services on the ground. Even if we were to take a view that all INGOs are private rent-seeking organizations, international attention to INGOs is still necessary for INGOs to get the donations and aid that the organizations are ultimately after (Bob, 2005; Cooley and Ron, 2002).

In short, although the INGO literature offers conflicting understandings of the motivations of INGOs, the collective literature does make some general predictions about the necessity of safety for volunteers and international attention for the working of INGOs. We think these ideas are extremely important for understanding INGO involvement in relatively unsafe post-conflict states.

States with a history of civil war: safety concerns

The rich literature on civil wars and post-conflict states offers many insights that correspond to our ideas of safety for INGOs. Most of this literature concerns variation in war-affected states which could ultimately be related to security concerns for INGOs themselves. First, countries with a history of civil war vary in the intensity of continued conflict (Fearon and Laitin, 2003; Lacina, 2006; Lacina and Gleditsch, 2005). Some wars quickly lose intensity while others continue on with extremely high battle death counts. We assume that states with a history of civil war where fighting has dissipated will tend to be more safe for INGOs. To the extent that continued conflict intensity could serve as a way to gauge underlying danger within a state with a history of civil war, it follows that INGOs, even those completely motivated to help in human security provision, would tend to avoid countries where fighting was still high intensity, with more battle deaths. In other words, because of an interest in making sure their volunteers are safe, INGOs may tend to feel the risks to the safety of volunteers outweigh any rewards from aiding in human security in these post conflict states. INGOs may fear that problems with safety could limit their impact in states where continued or renewed conflict levels are heightened, leading INGOs to select those countries where current conflict intensity is diminished.

Another key insight related to the idea of INGO safety in states with a history of civil war concerns the role of peacekeepers. According to much of the literature on civil wars, peacekeepers, especially those orchestrated by the UN, (a) are sent to the most egregious states but (b) do help provide peace and security, significantly diminishing the likelihood of war re-emergence (Fortna, 2004; Fortna, 2008). This key insight also applies to INGOs. Many UN Peacekeeping mandates include the necessity of providing assistance to humanitarian and human security organizations; this was a critical component of the UN Peacekeeping mission to Bosnia, for example. As such, the presence of peacekeepers, in addition to keeping the peace between former belligerents, is also useful for the safety of INGOs in war-torn countries. Thus, INGOs should be more likely to be involved in states that have a presence or a history of peacekeeping. Regardless of the motivations of the organizations, INGOs can utilize the security peacekeepers provide to aid in their safety and allow them to work in accordance with their underlying motivations. The logistical support peacekeepers provide to a state could also work to protect the interests of INGOs.

Empirical implications: safety and salience

The existing INGO literature, as discussed above, highlights our two central ideas about INGO safety and issue-salience in influencing INGO operations. Further, as just discussed, the civil war literature stresses how variation in states with a history of civil war could influence organizational safety. Combined, this literature leads to a number of empirical implications about where INGO involvement will be heightened in states with a history of civil war. First, as to organizational safety, we could expect a number of safety concerns to influence human security INGO participation, and we suggest that:

Hypothesis 1: Within states with a history of civil war, human security INGOs will be more likely to be involved in states where continued conflict intensity is at a minimum.

When the state is not still experiencing high-intensity war fighting, INGOs could more freely provide advocacy and service provision on the ground without putting their staff and volunteers in undue danger. Further, even when conflict intensity is high, INGOs can utilize the security provided by peacekeepers to aid in their ability to function throughout the state without undue safety concerns. This logic would suggest that:

Hypothesis 2: Within states with a history of civil war, human security INGOs will be more likely to be involved in states if peacekeepers are currently present or have been within the state previously.

For INGOs, safety concerns can also be the result of overall government policies designed to hinder the ability of INGOs to function unencumbered within the state. Although human security INGOs may be called to a state because of violations of human rights (Keck and Sikkink, 1998), human security organizations do face many dangers with being involved in a state where the number of violations of human rights, in particular abuses of physical integrity rights, is high. For human rights INGOs that operate through advocacy tactics, this might lead an organization to adopt a ‘naming and shaming’ tactic from abroad as a way to avoid putting organizational representatives at risk for governmental actions (Murdie and Bhasin, 2011). For human security organizations that are more focused on service provision, such as providing development assistance and food relief, the danger from government abuses of physical integrity rights may not be as great because these organizations are not primarily trying to change governmental practices on physical integrity rights.

For both service and advocacy human security organizations, however, all organizations would be more likely to operate in a country where there are no restrictions on the right of association. When a government makes it more difficult for citizens to associate, it encumbers the ability of INGOs to operate with local populations and increases the safety concerns of organizational members, both domestic and international. It puts operations in jeopardy and can dissuade organizations from participating in a state, as has occurred in Sudan in the last decade (Pantuliano et al., 2009). As such, safety concerns should be heightened in states where there are restrictions on freedom of association:

Hypothesis 3: Within states with a history of civil war, human security INGOs will be more likely to be involved in states where freedom of association is respected.

In addition to these implications concerning INGO safety, the existing INGO literature is clear that INGOs have multiple organizational motivations that would lead them to states that are most salient in the minds of the international community. Countries that are receiving more international attention would be likely to be ones to which volunteers and stakeholders would want to see their INGO respond. They may be countries where INGOs could attract the most private donations and volunteers and may be countries where INGOs interested in advocacy could generate the media attention necessary for government concessions (Risse and Ropp, 1999). As such, we propose:

Hypothesis 4: Within states with a history of civil war, human security INGOs will be more likely to be involved in states with more international media attention.

Finally, what about issues of donor funds, especially governmental aid to INGOs? Consistent with overall arguments of the necessity of international attention, we remain sympathetic to the idea that some INGOs are choosing states where they expect donations and governmental aid. The discussion by Cooley and Ron (2002) of NGO ‘scrambles’ for donor funds does appear to be playing out in many areas of the world. Sometimes, of course, this aid could be used for rent-seeking or other non-principled motivations (Cooley and Ron, 2002; Petras 1999). This behavior would be occurring, however, even if human security INGOs were all principled actors, only interested in aiding a domestic population with their own goals: they would still need the donor funds aid agencies are providing. And, if aid agencies were providing funds to those areas with some human security needs, we would expect even predominately principled INGOs to respond to this aid. As such, we could expect that:

Hypothesis 5: Within states with a history of civil war, human security INGOs will be more likely to be involved in states with more international aid.

Research design

To test directly these implications of our argument of how concerns with safety of INGO workers and organizational needs for international attention contribute to the involvement of INGOs in war-torn states, we use a dataset of the activities of human security INGOs in countries with a history of civil war in the period 1982 to 2003. We begin by discussing our data sources and empirical modeling decisions before turning to our results.

Dependent variables and sample

Because we are concerned predominantly with the behavior of INGOs that could directly contribute to human security in post-conflict states, for this project our dependent variable must capture the number of human security INGOs active within a country.³ For this, we use two separate dependent variables, one capturing the number of INGOs with a focus on human rights and one capturing the number of INGOs with a focus on sustainable development. Both human rights and development are central to the overall concept of human security (Paris, 2001). Because they are slightly different types of INGOs (human rights as an advocacy INGO and development as a type of service INGO) that have been mostly discussed in the extant scholarship independently, we first look at them in different empirical models and then combine the two measures into one count and examine this count as a variable of total number of human security (human rights + development) INGOs. Importantly, our expectations are consistent for both of these subcategories of human security INGOs.

To measure both human rights and development INGOs within a country, we rely on data coded by Smith and Wiest (2005) from the Yearbook of International Organizations (UIA 2008/2009). These data capture the number of organizations that self-report having a member or volunteer within a country in a given year and are predominately classified as human rights or development focused, respectively.

Our expectations, again, concern only INGOs within countries with a history of civil war. Therefore, following Murdie and Davis (2010), we restrict our sample to countries that have experienced a civil war in the past 10 years. When we combine data on civil wars with the data on INGOs used in this study, this leaves us with a sample of 479 observations for the years 1982 to 2003.

Key independent variables

Our hypotheses concern concepts of state characteristics that reflect our underlying argument concerning INGO safety and international issue salience. For Hypothesis 1, concerning the role of continued conflict intensity on the safety of INGO volunteers, we rely on the yearly measure of maximum conflict intensity without a country-year from the UCDP/PRIO Armed Conflict Dataset (Harbom and Wallensteen, 2010). We use their ordinal variable of civil conflict intensity, where ‘0’ indicates no conflict with more than 25 battle deaths within that country in that year, ‘1’ indicates a conflict with greater than 25 but less than 1000 battle deaths, and ‘2’ indicates a full-blown civil war in that country in that year, with more than 1000 battle deaths.⁴

For Hypothesis 2, concerning the role of peacekeepers in providing security necessary for INGO safety, we rely on two indicators adapted from the Mullenbach and Dixon (2010) dataset on third party intervention. First, for our general argument, we include a dichotomous indicator of the presence or absence of third party (UN, other intergovernmental organizations, states) peacekeepers. Second, reflecting the idea that peacekeepers could have lasting impacts on security even after they leave, as would be consistent with our argument, we include an indicator of the years since peacekeeping.

Our third hypothesis concerned the role of freedom of association as indicative of a safe environment for INGOs. We argue that abuses of association rights will serve to dissuade INGOs from operating within a country’s borders: these abuses make it difficult for the organization to function within the state and work with domestic groups and individuals. In addition, freedom of association might be a proxy for the degree to which host governments are open to non-governmental organizations (or other international actors) entering the host country.⁵ Therefore to test this hypothesis we include the measure of freedom of association proposed by Cingranelli and Richards (CIRI) (2010) as an independent variable. The scale varies from ‘0’, indicating that a state has severely restricted freedom of association, to ‘2’, indicating that a state has no restrictions on freedom of association. In our empirical models, we also include a control for physical integrity rights from the Human Rights Database developed by Cingranelli and Richards (2010). As mentioned, although abuses of physical integrity rights violations may call organizations to a country, there is also a countervailing influence from the need to keep INGO volunteers safe, especially INGOs with a human rights advocacy focus, which may also be able to advocate outside of the state’s borders.⁶

Hypothesis 4 concerns the influence of international media attention as a driver for human security INGOs. We argue that this media attention is an indicator for underlying issue salience, which has long been argued to be linked to INGO decision-making (Bob, 2005; Carpenter, 2007). To capture this concept, we use an indicator of the natural log of the number of events reported in the Reuters Global News Service for a given country in a given year. This variable comes from the Integrated Data for Event Analysis (IDEA) project (Bond et al., 2003). Because Reuters has a truly global presence it is a good indicator of the salience of a particular country in the international media (Murdie and Bhasin, 2011). Unfortunately, this variable is only available starting in 1990 and because of this we run two models for each of our dependent variables: one with the inclusion of this indicator and one without its inclusion.⁷

Finally, our fifth hypothesis argues that foreign aid will contribute to INGO behavior in our sample. To test for this, we include a measure of the natural log of all aid per capita, as reported in the World Development Indicators (WDI, 2008).

Control variables and model specification

Due to the count nature of our dependent variable, we use negative binomial regression models. Because we have a cross-sectional time series design, we utilize a generalized estimating equation approach with an AR (1) correlation structure and robust standard errors (Zorn, 2001). To account for any confounding effects, we control for GDP per capita (ln) and population size (ln), using variables from the World Development Indicators (WDI, 2008). In line with past literature, we would expect larger and richer populations to have a larger INGO sector, even within war-affected countries (Smith and Wiest, 2005). In addition, we include controls for democracy using a dichotomous indicator of whether a country was above a value of 6 or above on the Polity IV −10 to +10 measure (Marshall and Jaggers, 2007).⁸ In all states, INGOs are typically located in democracies more often than in authoritarian regimes (Boli and Thomas, 1999). Also, as mentioned above, we include Cingranelli and Richards’ physical integrity rights measure (Cingranelli and Richards, 2010), given previous research on all countries which found that INGOs tend to be located in countries that respect human rights (Tsutsui and Wotipka, 2004). All independent variables are lagged one year in the analyses.

Findings

Our findings generally support our hypotheses, indicating the need for INGO safety and issue salience. These results are provided in Table 1. The table first outlines the determinants of human rights and development INGOs separately and then looks at the determinants of an indicator for the total number of human rights and development INGOs, both of which are key components of the overall concept of human security as freedom from ‘fear’ and ‘want.’ These three dependent variables are first examined for the entire 1982 to 2003 time period (Models 1–3) and then, with the addition of the variable for Reuters news coverage (ln), for the 1991 to 2003 time period (Models 4–6).

Table 1.

Determinants of human security INGOs in countries with a history of civil war.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	Human Rights INGOs	Sustainable Development INGOs	Human Security (Human Rights + Sustainable Development) INGOs	Human Rights INGOs	Sustainable Development INGOs	Human Security (Human Rights + Sustainable Development) INGOs
GDP per capita (ln)	−0.10480	−0.04890	−0.06361	0.02396	0.03840	0.03171
GDP per capita (ln)	(0.08999)	(0.09056)	(0.09034)	(0.06442)	(0.05853)	(0.06017)
Population (ln)	0.38677***	0.31802***	0.33529***	0.27397***	0.27409***	0.28138***
	(0.07021)	(0.06037)	(0.06355)	(0.06690)	(0.06213)	(0.06345)
Democracy	0.04421	0.05483*	0.05233*	0.02303	0.00805	0.01083
	(0.03377)	(0.03007)	(0.02958)	(0.03810)	(0.01916)	(0.02263)
Reuters News Coverage (ln)				0.04285**	0.03639**	0.03734**
Reuters News Coverage (ln)				(0.02023)	(0.01621)	(0.01683)
CIRI Association Rights	0.03187***	0.02066***	0.02273***	0.02591***	0.01772**	0.01921**
CIRI Association Rights	(0.00977)	(0.00713)	(0.00708)	(0.00975)	(0.00731)	(0.00759)
CIRI Physical Integrity Rights	−0.00525	−0.00684**	−0.00642**	−0.00392	−0.00165	−0.00209
CIRI Physical Integrity Rights	(0.00363)	(0.00312)	(0.00308)	(0.00376)	(0.00258)	(0.00274)
Conflict Intensity	−0.03413***	−0.02168**	−0.02419**	−0.05206***	−0.03551***	−0.03893***
Conflict Intensity	(0.01197)	(0.00974)	(0.00977)	(0.01268)	(0.00792)	(0.00855)
Peacekeepers present	0.08183*	0.08524***	0.08361**	0.09441**	0.07294**	0.07751**
Peacekeepers present	(0.04394)	(0.03058)	(0.03272)	(0.04487)	(0.03157)	(0.03421)
Years since peacekeepers	0.01472***	0.01689***	0.01633***	0.01766***	0.01582***	0.01619***
Years since peacekeepers	(0.00518)	(0.00570)	(0.00533)	(0.00623)	(0.00553)	(0.00556)
Aid per capita (ln)	0.05068**	0.03115*	0.03497**	0.03670*	0.02207	0.02525
Aid per capita (ln)	(0.02271)	(0.01631)	(0.01744)	(0.01937)	(0.01465)	(0.01563)
Constant	−2.70546**	−0.51047	−0.48261	−1.80230*	−0.44172	−0.30768
	(1.21856)	(1.09212)	(1.12823)	(1.06707)	(0.99247)	(1.01031)
Observations	479	479	479	331	331	331

Generalized Estimating Equation with an AR(1) Correlation Structure, Robust Standard Errors in Parentheses. Models 1–3 cover the years 1982 to 2003. Models 4–6 cover the years 1991 to 2003. All independent variables lagged one year. *** p < 0.01, ** p < 0.05, * p < 0.1.

As shown in Columns 1-6 of Table 1, human security INGOs of all stripes are less likely to be involved in states with intense ongoing conflict, as anticipated by Hypothesis 1 and supported by the statistically significant and negative coefficient on the independent variable ‘conflict intensity.’ In all, this would support our notion of the need for INGO safety in determining their involvement in states with a history of civil war.

Hypothesis 2, the role of peacekeepers, is also supported, at least at the 10% level (two-tailed), in all of the models shown in Table 1. As shown, human security INGOs are more likely to be involved in countries with ongoing peacekeeping missions and even more likely to be in countries with a past peacekeeping mission.

Hypothesis 3 stated that INGOs are more likely to be involved in countries with greater respect for freedom of association. Across the models shown in Table 1, we find consistently positive and statistically significant results for the CIRI indicator of freedom of association. Especially noteworthy, as shown in Column 2, is the fact that we find that sustainable development INGOs are more likely to be found in states with less respect for physical integrity rights. This finding continues in Column 3, where the dependent variable captures the total count of human rights and sustainable development INGOs. When the dependent variable captures only human rights INGOs, as shown in Column 1, however, we find no statistically significant impact of physical integrity rights performance on the number of human rights INGOs within states with a history of civil war. We take this to imply, perhaps, that even though human rights INGOs may be called to states with worse respect for physical integrity rights, safety concerns are a counterweight to their involvement within the state, perhaps leading human rights-specific organizations to adopt alternative strategies for rights-promotion, like ‘naming and shaming’ states from abroad (Murdie and Bhasin, 2011; Risse and Ropp, 1999). We find no evidence of a statistically significant effect of physical integrity rights in the more temporally-restricted sample shown in Columns 4–6. Overall, we take these results on association rights and physical integrity rights to indicate that INGOs are concerned with their safety and the safety of the domestic populations that could associate with them, leading them to avoid states where restrictions of association are the most pronounced.

We find strong support for Hypothesis 4, concerning the role of international attention as motivating INGO involvement in states with a history of civil war. As shown in Columns 4–6 of Table 1, across the three different dependent variables our indicator for coverage in Reuters Global News Service (ln) is consistently statistically significant and in the positive direction. We take this result to imply that INGOs respond to those countries that have received the most attention, perhaps as a way to attract additional advocacy network actors and funds to a war-affected area (Risse and Ropp, 1999).

Finally, Hypothesis 5’s argument that in war-torn countries INGOs are responding to concerns about aid is supported by the positive and significant coefficients on the aid variables in each of the models for the full sample (Columns 1–3) and in the more temporally-restricted model, where the dependent variable is human rights INGOs. Again, we think this does not automatically suggest anything negative about the INGO community or motivations; INGOs are merely responding to the same countries where the international community is providing attention and funds. Our control variables, when significant, are also largely consistent with the literature, indicating that there are more INGOs involved in countries with larger populations and that, at least for development INGOs, there may be more INGOs in democracies.⁹

What do these results mean substantively?¹⁰ In our sample of war-torn states, the mean state has about 26 human rights INGOs active within its borders (range from 0 to 65). If a country was to go from a situation where it had no low-scale conflict (conflict intensity is ‘0’) to a situation where there was low-scale violence (conflict intensity is ‘1’), with all other variables at their means or medians if dichotomous in the dataset, the number of human rights INGOs active within the state’s borders is expected to drop by almost 2 (95% confidence interval of a drop of 0.0365 to 3.472 human rights INGOs). For sustainable development INGOs, this impact is also substantial. In the sample, we find a mean number of sustainable development INGOs of around 9 (range from 2 to 274). As a state moves from a no-conflict situation to one of low-level conflict, the expected number of sustainable development INGOs will drop by over 8 (95% confidence interval of a drop of 2.142 to 14.032 organizations). Again, this is a quite substantial decrease due, we would argue, to concerns about volunteer safety in areas with even low-intensity conflict.¹¹

We find that peacekeepers can help INGOs overcome safety concerns in post conflict areas. For sustainable development INGOs, for example, the presence of a peacekeeping mission increases the expected number of INGOs by around 8 (95% confidence interval from 2.280 to 13.380).¹² The peacekeeping variable is only statistically significant at the 10% level in the human rights INGO model but does indicate that the presence of a peacekeeping mission will increase the number of INGOs within the state by almost 2 (90% confidence interval increase of 0.212 to 3.681 human rights INGOs). Similarly, the safety effects of peacekeepers remain even after the active peacekeeping mission has been withdrawn. As our indicator for years since a peacekeeping mission goes from its mean value in the dataset to one standard deviation above the mean, the expected number of human rights INGOs increases by a little over 2 (95% confidence interval of 1.732 to 3.308). For sustainable development INGOs, a similar change in the number of years since a peacekeeping mission is expected to have an increase of a little over 8 (95% confidence interval from 5.394 to 11.785).

Association rights are also quite important. A change from the worst association rights (0 on the 0/1/2 scale) to middling respect for association rights is expected to lead to an increase of 12 sustainable development INGOs (95% confidence interval from 6.820 to 18.009) or 3 human rights INGOs (95% confidence interval from 1.664 to 4.461). Collectively, then, we take these results to indicate that safety is important for human rights and development INGOs in states with a history of civil war.

Issue-salience is also important. As the number of stories about a country in Reuters increases from the mean to one standard deviation above the mean in the sample, there is expected to be an increase of almost 18 sustainable development INGOs (95% confidence interval of 6.458 to 30.233 organizations) or over 3 human rights INGOs (95% confidence interval from 0.948 to 6.490 organizations).

Finally, foreign aid does seem to make a difference in the behavior of INGOs. Again, for sustainable development INGOs, this effect is rather substantial. As foreign aid to the state increases from the mean in the sample to one standard deviation above the mean, the count of INGOs in-country is expected to increase by around 13 (95% confidence interval from 7.443 to 18.606 organizations). This rather large effect may be heightened for sustainable development INGOs, which rely heavily on aid for service provision. However, also in line with our argument, the result is still rather large for human rights INGOs; the expected increase is around 3 organizations (95% confidence interval from 1.945 to 5.322 organizations). Again, we think these overall findings illustrate both the statistical robustness and the substantive importance of our general argument.

We summarize our substantive findings for the combined human rights and development indicator, as shown in Columns 3 and 6 of Table 1 and in Figure 1. This figure provides the predicted marginal effect of each substantively important variable, evaluated at its median with all other variables at their medians, on the expected count of human security INGOs. As can be seen, many indicators for safety and security influence INGO operations in states with a history of civil war.

Figure 1.

Marginal effects on predicted number of human security (human rights + development) INGOs.

Conclusion

What does this research imply for the civil war reconstruction and human security communities? First, we argue and find that INGO involvement in states with a history of civil war is driven both by concerns about safety to volunteers and staff as well as by concerns about international attention. These issues, although relatively basic, have been missing in the growing cross-national literature on the topic. INGOs, even if highly motivated to provide aid in the human security of peoples affected by civil war, still need to be concerned with the safety of their staff. As such, they tend not to go to conflicts where fighting is intense. However, they do tend to go to states with a presence or history of a presence of peacekeepers. Because peacekeepers tend to go to violent conflicts (Fortna, 2008), INGOs, through following the ‘blue helmets’ of the UN or other peacekeeping forces, may be able to find ways to utilize the stability that peacekeepers provide in ways that keep their organizations safe.

Further, we find that organizations, although not shying away from states with egregious physical integrity rights abuses, are more likely to be involved in states where freedom of association is better respected. This could be out of concern both for their own safety and for the safety of the individuals who must associate with the organization in order to receive many of the services INGOs can provide. This result would support moves by the international community to pressure states to keep their borders open for civil society actors during and immediately after civil war.

We do find that INGOs are responding to international attention, going to areas with more donor funds and international media attention. Although this finding might appear intuitive, it does reiterate how concerns with international salience can keep some states and human security issues off the international agenda, which could attenuate the role INGOs play in ‘forgotten’ civil war states. In fact, the data concerning our two examples given in the introduction –Tajikistan and Georgia– fit well with our overall results. Tajikistan had far more restrictions on association rights, more intense ongoing violence, and much less international media attention than Georgia, leading, perhaps, to less involvement by human security INGOs. Because human security INGOs can be such important actors in civil war reconstruction efforts, this lack of INGO involvement may explain the somewhat diminished human security situation in Tajikistan even a decade since the fighting stopped.

We see these findings as important to the work of INGOs on-the-ground in three regards. First, INGOs, in responding to the presence of peacekeeping forces, are illustrating the need for coordination from the INGO community to intergovernmental organizations (Shiras, 1996; Stahn, 2001). Coordination between civilian and military actors could help ensure the safety of INGO personnel in post-conflict states (Roberts, 2012).

Second, we find that INGOs are responding to the aid community. Instead of the often negative connotations this brings, we argue that this merely illustrates that INGOs writ large are responding to similar situations as the aid community and want to receive the international attention and funds that are being provided. International attention has long been argued to be necessary for INGO success (Risse and Ropp, 1999). In this regard our findings suggest that human security INGOs may not necessarily be ‘scrambling’ for donor funds as much as they are ‘scrambling’ to help areas towards which the international community is turning its attention.

Finally, these findings illustrate some patterns of the behavior of INGOs in states with a history of civil war. Future work could look at how INGOs behave in other human security disaster situations – for instance, after tsunamis or droughts. Given the neeed for coordination between INGOs for efficiency on the ground, these findings would indicate that such coordination could take place, because INGOs tend to operate in predictable patterns across human security fields.

Footnotes

Acknowledgements

We are grateful for assistance from four anonymous reviewers and Mark Kesselman. A previous version of this paper was presented at the 2011 conference of the InterUniversity Seminar on Armed Forces and Society; we thank all panel participants for their suggestions.

Funding

We acknowledge and express appreciation financial assistance received from a 2011 American Soldier Grant at Kansas State University.

Notes

Author biographies

Amanda Murdie is an associate professor of political science at the University of Missouri. She also holds a joint appointment in the Truman School of Public affairs. Her interests center on the strategies and effects of international non-governmental organizations in human security.

Sean Webeck is a Public Policy PhD student at Indiana University. His research interests include civil–military relations, collaboration, and management.

References

Jazeera

(2008) Aid workers killed in Afghanistan. Available at: http://www.aljazeera.com/news/asia/2008/08/20088139251345245.html.

Bellamy

Alex J

Williams

Paul

Griffin

Stuart

(2004) Understanding Peacekeeping. Maxwell, MA: Blackwell Publishing.

Bob

Clifford

(2005) The Marketing of Rebellion: Insurgents, Media, and International Activism. Cambridge, MA: Cambridge University Press.

Bond

Doug

Bond

Joe

Churl Oh

Jenkins

Craig

Taylor

Charles Lewis

(2003) Integrated data for events analysis (IDEA): An event typology for automated events data development. Journal of Peace Research 40(6): 733–745.

Boli

John

Thomas

George M

(1999) INGOs and the organization of world culture. In Boli

John

Thomas

George M

(eds) Constructing World Culture: International Non-governmental Organizations Since 1875. Palo Alto, CA: Stanford University Press, 13– 49.

Carpenter

R Charli

(2007) Studying issue(non)-adoption in transnational advocacy networks. International Organization 61(3): 643–667.

Cingranelli

David L

Richards

David L

(2010) CIRI Variables List and Short Description. Available at: http://www.humanrightsdata.org/documentation/ciri_variables_short_descriptions.pdf.

Cingranelli

David L

Richards

David L

(2001) Measuring the impact of human rights organizations. In Welch

Claude E

(ed.) NGOs and Human Rights: Promise and Performance. University of Pennsylvania Press, 225–238.

Collier

Paul

Hoeffler

Anke

(2004) Greed and grievance in civil war. Oxford Economic Papers 56(4): 563–595.

10.

Cooley

Alexander

Ron

James

(2002) The NGO scramble: organizational insecurity and the political economy of transnational action. International Security 27(1): 3–39.

11.

DeMars

William E.

(2005) NGOs and Transnational Networks: Wild Cards in World Politics. London, UK: Pluto Press.

12.

Doppler

Bruno

(1996) The ICRC in complex emergencies. In Whitman

Jim

Pocock

David

(eds) After Rwanda: the coordination of United Nations humanitarian assistance. New York: St. Martin’s Press, 128–135.

13.

Edwards

Michael

Hulme

David

(1996) Too close for comfort? The impact of official aid on nongovernmental organizations. World Development 24(6): 961–973.

14.

Fearon

James D

Laitin

David D

(2003) Ethnicity, insurgency, and civil war. American Political Science Review 97(1): 75–90.

15.

Fortna

Virginia Page

(2004) Does peacekeeping keep peace? International intervention and the duration of peace after civil war. International Studies Quarterly 48(2): 269–292.

16.

Fortna

Virginia Page

(2008) Does Peacekeeping Work?: Shaping Belligerents’ Choices After Civil War. Princeton, NJ: Princeton University Press.

17.

Harbom

Lotta

Wallensteen

Peter

(2010) Armed conflicts, 1946–2009. Journal of Peace Research 47(4): 501–509.

18.

Howard-Hassmann

Rhoda E

(2012) Human security: undermining human rights? Human Rights Quarterly 34(1): 88–112.

19.

IRIN (2002) TAJIKISTAN: Coordination needed for new NGOs. Available at: www.http://www.irinnews.org/Report/18114/TAJIKISTAN-Coordination-needed-for-new-NGOs.

20.

Krain

Matthew

(1997) State-sponsored mass murder: the onset and severity of genocides and politicides. Journal of Conflict Resolution 41(3): 331–360.

21.

Keck

Margaret E

Kathryn

Sikkink

(1998) Activists Beyond Borders: Advocacy Networks In International Politics. Ithaca, NY: Cornell University Press.

22.

Kumarasinghe

Uditha

Wijayapala

Ranil

(2006) No restrictions on NGOs in N-E, The Daily News, 1 September. Available at: http://www.dailynews.lk/2006/09/01/news26.asp.

23.

Lacina

Bethany

(2006) Explaining the severity of civil wars. Journal of Conflict Resolution 50(2): 276–289.

24.

Lacina

Bethany

Gleditsch

Nils Petter

(2005) Monitoring trends in global combat: A new dataset of battle deaths. European Journal of Population 21(2): 145–166.

25.

Manji

Firoze

O’Coill

Carl

(2002) The missionary position: NGOs and development in Africa. International Affairs 78(3): 567–583.

26.

Marshall

Monty

Jaggers

Keith

(2007) Polity IV Project: Center for Systemic Peace. Available at: http://www.systemicpeace.org/polity/polity4.htm.

27.

Mullenbach

Mark J

Dixon

William J

(2010) Third Party Interventions in Intrastate Disputes Project Available at: http://uca.edu/politicalscience/dadm-project/.

28.

Murdie

Amanda

Bhasin

Tavishi

(2011) Aiding and abetting: human rights INGOs and domestic protest. Journal of Conflict Resolution 55(2):163–191.

29.

Murdie

Amanda

Davis

David R

(2010) Problematic potential: the human rights consequences of peacekeeping interventions in civil wars. Human Rights Quarterly 32(1): 49–72.

30.

Pantuliano

Sara

Jaspars

Susanne

Ray

Deepayan Basu

(2009) Where to now? Agency expulsions in Sudan: consequences and next steps. Available at: http://www.odi.org.uk/publications/3198-agency-expulsion-sudan.

31.

Paris

Roland

(2001) Human security: Paradigm shift or hot air? International Security 26(2): 87–102.

32.

Petras

James

(1999) NGOs: in the service of imperialism. Journal of Contemporary Asia 29(4): 429–440.

33.

Poe

Steven

Tate

C Neal

(1994) Repression of human rights to personal integrity in the 1980s: a global analysis. American Political Science Review 88(4): 853–872.

34.

Powell

Colin

(2001) Remarks to the National Foreign Policy Conference for Leaders of Non-Governmental Organizations. Available at: http://2001-2009.state.gov/secretary/former/powell/remarks/2001/5762.htm.

35.

Ratner

Steven R

Abrams

Jason S

(2001) Accountability for Human Rights Atrocities in International Law: Beyond the Nuremberg Legacy. Oxford, UK: Oxford University Press.

36.

Risse

Thomas

Ropp

Stephen C

(1999) International human rights norms and domestic change. In Risse

Thomas

Ropp

Stephen C

Sikkink

Kathryn

(eds) The Power of Human Rights: International Norms and Domestic Change. Cambridge, MA: Cambridge University Press, 234–278.

37.

Roberts

Nancy C

(2010) Spanning ‘bleeding’ boundaries: humanitarianism, NGOs, and the civilian-military nexus in the post–cold war era. Public Administration Review 70(2): 212–222.

38.

Shah

Saubhagya

(2008) Civil Society in Uncivil Places: Soft State and Regime Change in Nepal. Honolulu, HI: East-West Center.

39.

Shiras

Peter

(1996) The role of NGOs. In Whitman

Jim

Pocock

David

(eds) After Rwanda: The Coordination of United Nations Humanitarian Assistance. New York, NY: St. Martin’s Press, 106–117.

40.

Sikkink

Kathryn

(1993) Human rights, principled issue-networks, and sovereignty in Latin America. International Organization 47(3): 411–441.

41.

Smith

Jackie

Wiest

Dawn

(2005) The uneven geography of global civil society: national and global influences on transnational association. Social Forces 84(2): 621–652.

42.

Stahn

Carsten

(2001) NGOs and international peacekeeping issues, prospects and lessons learned. Heidelberg Journal of International Law 61(2–3): 379–402.

43.

Tomz

Michael

Wittenberg

Jason

King

Gary

(2003) CLARIFY: Software for interpreting and presenting statistical results. Journal of Statistical Software 8(1): 1–30.

44.

Tsutsui

Kiyoteru

Wotipka

Christine M

(2004) Global civil society and the international human rights movement: citizen participation in human rights international nongovernmental organizations. Social Forces 83(2): 587–620.

45.

UIA (2008/2009) Yearbook of International Organizations. Volume 1–4, 45th edition. Brussels, Belgium: Union of International Associations.

46.

UN (1948) Universal Declaration of Human Rights. Available at: http://www.un.org/en/documents/udhr/.

47.

WDI (2008) World Development Indicators – 2008. Washington DC: World Bank. Available at: http://data.worldbank.org/products/data-books/WDI-2008.

48.

Zorn

Christopher JW

(2001) Generalized estimating equation models for correlated data: a review with applications. American Journal of Political Science 45(2): 470–490.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.01 MB