Foreign Policy Change from an Advocacy Coalition Framework Perspective

Why does a state change its foreign policy objectives and who is respon- sible for instigating such change? According to Hermann, four primary change agents are central to this process: leaders, bureaucracies, changes in domestic constituencies, and external shocks. This paper argues that the Advocacy Coalition Framework (ACF) is a complementary policy pro- cess framework that can explain foreign policy change (FPC) and that accounts for all four of these primary change agents. Additionally, it is a broader framework of the policy process that facilitates a more comprehensive understanding of the factors inﬂuencing FPC than traditional FPC research. The ACF has the potential to broaden our understanding of FPC by emphasizing the intersection of the international system with domestic politics and focusing on a myriad of policy actors coordinating their advocacy efforts to inﬂuence FPC. To support this argument, the paper dis- cusses how FPC can beneﬁt from the ACF and reviews past applications. It proposes a research agenda using the ACF to study FPC and draws conclu- sions about future challenges and directions. politique étrangère et tire des conclusions concernant les orientations et déﬁs futurs.

investigación usando el ACF para estudiar el FPC y llega a conclusiones sobre las orientaciones y los desafíos futuros.
Pourquoi un État change-t-il ses objectifs de politique étrangère et qui est responsable de l'instigation d'un tel changement? Selon Hermann (1990), quatre principaux agents de changement sont au coeur de ce processus: les dirigeants, les bureaucraties, les changements de groupes d'influence nationaux et les chocs extérieurs. Cet article soutient que le Cadre de coalition de plaidoyer (Advocacy Coalition Framework, ACF) (Sabatier et Jenkins-Smith, 1993) est un cadre complémentaire des processus politiques qui peut expliquer le changement de politique étrangère et tenir compte de l'ensemble de ces quatre principaux agents de changement. De plus, il s'agit d'un cadre plus large des processus politiques qui facilite une compréhension plus complète des facteurs influençant le changement de politique étrangère qu'avec une recherche traditionnelle sur le changement de politique étrangère. Le cadre de coalition de plaidoyer a le potentiel d'élargir notre compréhension du changement de politique étrangère en mettant l'accent sur l'intersection entre système international et politique intérieure et en se concentrant sur une myriade d'acteurs politiques coordonnant leurs efforts de plaidoyer pour influencer le changement de politique étrangère. Pour soutenir cet argument, cet article discute de la manière dont le changement de politique étrangère peut bénéficier du cadre de coalition de plaidoyer et évalue ses applications passées. Il propose un programme de recherche utilisant le cadre de coalition de plaidoyer pour étudier le changement de politique étrangère et tire des conclusions concernant les orientations et défis futurs.

Introduction
In response to episodes of global upheaval and transformation, international relations (IR) scholars sought to better understand foreign policy change (FPC). For example, at the end of the Cold War, when the East-West divisions of the global system stopped functioning as a stabilizing framework, many states in the world made substantial changes in their foreign policies. Again, after the United States' unipolar moment, with the emergence of a multipolar world, some states found that they had a new freedom to shape their own foreign policies. At such times, foreign policy scholars began to look for and develop approaches to understand FPC. However, the development of theories of policy change from IR has been characterized as a slow and sporadic process (Hudson 2012).
Two promising theories of policy change from IR shifted their focus to the role of domestic political factors. Charles Hermann (1990) classified four primary change agents: leaders, bureaucracies, changes in domestic constituencies, and external shocks. He argued that these agents were central to understanding the importance of international and domestic structural conditions, the political agency of individuals, and the decision-making process within the machinery of government that can lead to a major FPC. Writing nearly a decade after Hermann, Jakob Gustavsson (1999) developed his own model of FPC. He drew inspiration from public choice theory, which also focuses on domestic political factors.
This research builds on Hermann's (1990) and Gustavsson's (1999) efforts to learn from the insights of domestic public policy theories and brings them to the realm of IR and FPC. Although this research is not the first to place greater emphasis on domestic processes to understand FPC, it does move the research agenda forward. Other scholars combined theories such as a leader's operational code (Holsti 1977) or trait analysis (Yang 2010) as well as loss-aversion decision-making (Welch 2005), within a narrow scope. Focusing only on a leader or a decision-making process does have its benefits, but policymaking includes a wide assortment of actors who seek to influence the policy process as well as other domestic and international political vectors that provide opportunities as well as constraints for foreign policymaking.
Alternatively, the Advocacy Coalition Framework (ACF) developed by Paul Sabatier and Hank Jenkins-Smith (Sabatier 1988;Sabatier and Jenkins-Smith 1993) is a broader framework of the policy process that facilitates a more comprehensive understanding of the factors influencing FPC. This is partly because Sabatier created the ACF framework in 1988 precisely to comprehend why policy change takes place or, conversely, why it does not. The purpose of this paper is to explain why and how the ACF can be successfully utilized to study FPC as a complementary framework that can integrate existing FPC-specific theories, such as Hermann (1990) and Gustavssson (1999). The primary benefit of relying on the ACF as a broader policy process explanatory framework is that it provides a more robust and generalizable understanding of FPC.
The ACF is a well-defined and empirically grounded theory of the policy process (Jenkins- Smith et al. 2014). Sabatier and Jenkins-Smith (1993) developed the ACF to explain the public policy that involves contentious processes and policy actors who hold divergent beliefs. Such actors may include members of government and nongovernment agencies and may reside in interest groups, in academia and/or in the media. These policy actors coordinate their behavior as advocacy coalitions and compete with other advocacy coalitions that seek to translate their policy-related beliefs into public policy. Additionally, the ACF has three central theories about advocacy coalitions (e.g., their structure, their stability, and their behavior), about policy learning (e.g., the conditions and attributes of learning among and between coalitions), and about policy change (e.g., the identification of pathways that trigger opportunities for coalitions to seek policy change).
To date, scholars applied the ACF hundreds of times to explain change and stasis in the public policy realm (Weible et al. 2009. During the past decade, a growing number of studies also apply the ACF to better understand public policy in the foreign and defense domain (Pierce and Hicks 2017). As a framework that was developed to take into account the mechanisms for and the principal causal factors of change in the domestic policy process (Sabatier 1988), it certainly holds promise to do the same in the foreign policy process (Pierce and Hicks 2017).
This research will first consider the literature that has focused on FPC and identify its inherent weaknesses as well as advance why frameworks from the public policy process hold potential to rectify these shortcomings. The next section of this article expounds on the means by which the ACF bridges the gap between public policy and foreign policy, in particular, to explicate policy change. The subsequent section clarifies how the ACF bridges this gap by accounting for all four of Hermann's primary change agents within a single theory of policy change, while, at the same time, it considers the mechanisms for change within the policy process. To bolster the theoretical argument, two recent applications of the ACF to explain FPC are presented: the 2009 FPC of new statutory regulation regarding Swedish intelligence policy (Nohrstedt 2011) (Haar 2010). The paper concludes with a proposed research agenda for utilizing the ACF to better understand FPC. Hudson (2012) argues that foreign policy analysis is at a methodological impasse as it grapples with the question of what constitutes a proper examination of foreign policy in the American, British, and European schools. Thus far, no broad explanation of foreign policy has emerged, with critics even arguing that it is a sterile field, devoid of innovation since the 1980s (White 1999). At the turn of the twenty-first century, scholars were calling for a "new phase of study" (Hill 2003, xviii). After a "first generation," the members of which focused on comparative foreign policy, a "second generation," which emerged in the late 1990s, was more circumspect in scope, preferring a "more modest approach" rather than develop a general theory of foreign policy (Neack, Hey, and Haney 1995, 30). This is because past theories falter in explaining why policymakers choose one path among a range of possible responses to any given set of international or domestic stimuli. This is not a new problem and one inherent to IR, which must deal with at one level states and their relationships, and another level, the impact of individuals, as well as the impact of bureaucratic-domestic policy on policymakers. For example, Bryan Jones (2017, 64) tackles the issue of integrating the cognitive science of human individuals with understanding the interactions of people in political institutions, finding it "a difficult problem." To overcome such difficulties in foreign policy, scholars developed middle-range theories that focus on particular factors to compensate for the lack of a general theory.

Past Attempts at Explaining Foreign Policy Change
Along with the inability to explain why policymakers choose one path over others via a general theory is the equally thorny question of why does policy change. Understanding the conditions for change is one of the most difficult theoretical problems for policymakers and foreign policy scholars (Hermann 1990). IR scholars began to ask this question as early as the 1950s. For example, Snyder, Bruck, and Sapin (1954) explained the change in foreign policy by way of the decisionmaking process, either in small groups or in organizations. Their decision-making methodology was an early critic of structuralist approaches to foreign policy, or that foreign policy is essentially determined by the international system. Critics argued that a consideration of the international system variable on its own left a good deal of foreign policy decision-making unexplained. Instead, Snyder, Bruck, and Sapin (1954) argued that structural conditions were influential only when perceived and reacted upon by the decision-makers who possess agency. Thus, foreign policy, and FPC, was not only more than the macro analyses of structuralism, but also the result of the decision-makers' subjective understanding of the situation.
Theoretical development for understanding FPC increased during the 1980s and 1990s, as scholars began to question the stabilizing effects of interdependence and the Cold War (Holsti et al. 1982;Goldmann 1988;Hermann 1990;Carlsnaes 1992;Carlsnaes 1993;Rosati et al. 1994). For example, Kalevi Holsti et al. (1982) argued that some states restructured, or dramatically altered their foreign policy, because the interconnectedness of the postwar era was breaking down due to nationalist impulses. The burst of scholarship that came in the 1990s was also increasingly dissatisfied with the prevailing structural theories and tools available to explain FPC.
This group of scholars also began to differentiate between types of change. For instance, Holsti identified two types of change: first, a "normal" version, which is slow, incremental, and linked to specific sectors, and second, a "restructuring" type, which is non-incremental, linked to multiple sectors, and incurs considerable costs to the states that pursue it (Holsti et al. 1982, 2). Following on Holsti's two categories of change, Charles Hermann further delineated four graduated levels of FPC: (1) adjustment changes, referring to the level of effort applied to a current policy; (2) program changes, denoting a shift in the methods to achieve the goals of a policy; (3) problem/goal changes, referring to the replacement or forfeiture of a policy; and (4) international orientation changes, involving the redirection of an actor's entire orientation toward world affairs. For Hermann, the last three forms (means, ends, and overall orientation) constitute a "major foreign policy redirection," while the last change goes beyond one issue or policy and encompasses an actor's international role, its alignment with other nations, and its activities on the world stage (Hermann 1990, 6). Rosati et al. (1994), writing a few years later, brought the types of change back down to three: refinement/minor, reform/moderate, and restructuring/major change. Overall, this group of scholars also identified restraints to FPC, which included bureaucratic, regime, resource, global, and regional restrictions.
Although the stabilizing effects of the Cold War and interdependence were rapidly eroding by the mid-1990s, these scholars writing about change still viewed it as rare, which is the same view that earlier realist international theorists held. For the scholars of the 1990s, bureaucratic politics and domestic political interests tended to ensure foreign policy continuity (Rosati et al. 1994). Moreover, there was a tendency to conclude that regime change was the only way to bring about major foreign policy redirection (Hermann 1990). All cases of substantial change were also multicausal (which included global events and structures, domestic political realignment, and possible alterations in the policymaking process). For instance, regime change would not bring about an adjustment to foreign policy if the ideological orientation stayed the same or if the domestic environment was constraining.
While these scholars writing around the end of the Cold War did make advances in explaining FPC, they also recognized their work was unfinished. They did not complete a framework that would explain future foreign policy restructuring. Instead, they acknowledged that "comprehensive explanations of foreign policy change require a synthesis and the development of integrative theory" (Rosati et al. 1994, 271). Hermann (1990, 4) agreed with his fellow scholars that FPC is multicausal, but he disagreed that regime change is "virtually the only way to achieve profound shifts in a nation's foreign policy." As an alternative idea, Hermann (1990, 4) argued that research reveals "cases in which the same government that initiated a course in foreign policy recognizes that significant change must be undertaken." Given that such cases exist, Hermann's goal was to understand the circumstances in which governments self-correct foreign policy. He found his answer in four areas of scholarship, located in different academic disciplines, which further led to his classifying four primary change agents that he believed were central to understanding why a state changes its foreign policy. The four primary change agents are leaders, bureaucracies, changes in domestic constituencies, and external shocks. Hermann (1990) also argued that the manner in which these sources of policy adjustment are processed inside the machinery of government-in other words, the decision-making process-further affects policy outcomes. 1 Gustavsson (1999, 74) picked up Hermann's thread of argument by putting forward a model that focused on "changes in fundamental structural conditions, strategic political leadership, and the presence of a crisis of some kind." As Gustavsson rather dryly points out, in the intervening years after Hermann and others advanced ideas to explain why governments sometimes choose to make changes to their foreign policy, few empirical studies applied the concepts to cases of FPC. 2 Gustavsson was part of yet another uptick in research in FPC, reflecting a world with a perceived declining American hegemony that was moving toward a multipolar balance of power. However, similar to the previous burst of scholarship on FPC, this upsurge did not provide ready frameworks-in other words, no models that might analyze the increasing freedom to shape their own foreign policy that many states were experiencing at the time.
In order to rectify the lack of prototypical applications to cases of FPC, Gustavsson applies his own model to an instance of FPC, that being the Swedish decision to abandon its policy of non-European Community (EC) membership. In formulating his model, Gustavsson (1999) clearly states that he drew inspiration from public choice theory, which focuses on the domestic political factors needed to uphold a certain policy. Gustavsson emphasizes electoral results, opinion polls, and the coalitions formed between major political actors in the domestic level. At the end of his article, Gustavsson (1999, 92) calls for more theory development to explain FPCapproaches that combine the "importance of international and domestic structural conditions, political agency, and the decision-making process."

Attributes of the ACF
It is clear that since the 1950s researchers who focus on FPC are on a journey that continually looks beyond the theoretical constraints of the field of IR to find effective ways to analyze adjustments in foreign policy. 3 This investigation builds on Hermann's (1990) and Gustavsson's (1999) encouragement to learn from insights gained in public policy and argues that the ACF, as developed by Sabatier and Jenkins-Smith (1993), explains FPC. In fact, the ACF was devised by Sabatier in 1988 specifically to understand why policy change does or does not take place.
Pierce, Peterson, and Hicks (2020)  The scholars who employed the ACF to foreign policy found it surprising that they were pioneers; they wondered why public policy scholars neglect foreign, security, and defense policy and why foreign policy scholars shy away from adopting models that have demonstrable efficacy in explaining policy change in a wide array of case studies (Hirschi and Widmer 2010;Archuleta 2016;Pierce and Hicks 2017). For example, Litfin (2000) hoped her research contributed to "bridge-building between 'the two solitudes'" of IR and domestic policy studies.
This article similarly endeavors to engage in bridge building, arguing that the ACF bridges the gap between public policy and foreign policy in explaining pol- icy change (Pierce and Hicks 2017). The ACF makes connections between the two solitudes by, first, consolidating existing assumptions and approaches about how coalitions of actors, who advocate specific policy solutions, make foreign policy. At present, there is a wealth of foreign policy literature that focuses on how various groups seek to influence foreign policy (Holsti 1989). Such studies go beyond politicians and bureaucrats and focus on such elements as epistemological communities (Haas 1992), ethnic groups (Smith 2000;Ambrosio 2002;Rubenzer 2008), business and labor organizations (Lipset 1986), or religious affiliations (Mearsheimer and Walt 2007), as the rationale for the formation of coalitions that seek to implement policy. As an alternative, the ACF proposes that the beliefs of actors are a rationale for their formation into coalitions. This turn to beliefs is consistent with other FPA theories such as the operational code of leaders (George 1969;Holsti 1970Holsti , 1977. However, the Operational Code Analysis (OCA) does offer an understanding (an operationalization) of beliefs as related to individual decisionsmakers, while the ACF conceptualizes individuals as a group (who share certain beliefs). Friman (1993) also argues that a policy process theory that emphasizes beliefs as the element that brings policy actors together is relevant to the foreign policy process. The ACF, with its assumption that shared beliefs bind policy actors together (Sabatier and Jenkins-Smith 1999), offers a pathway to incorporate the role of ideas and beliefs in understanding foreign policy processes. It is also able to explain foreign policy when there is serious disagreement within a government about the best course of action.
Pierce and Hicks (2017), second, point out that foreign policy analysis tends to remain in the domain of traditional IR approaches, which often mitigate the role of ideas or beliefs as they relate to groups as well as individuals (Goldstein 1988;Litfin 2000). While institutional constraints are important to the policymaking process, an actor-centric approach to foreign policy analysis, like the ACF, represents a promising alternative. This is because an actor-centric approach investigates the sources of change while at the same time unpacking the decision-making process (Hudson and Vore 1995). The ACF understands the policy process as a competition between the coalitions in a policy subsystem.
Third, the ACF's actor-centric approach could also answer the questions that Hudson identified in her 1995 overview of the development of foreign policy analysis. Hudson identifies a third generation of foreign policy analysts who had begun to unpack the cognitive tasks that individuals in groups engage in while making decisions. Hudson's (Hudson and Vore 1995, 224-25) questions included: "How does a group come to share an interpretation of a situation?" "How are situations "framed" by a group?" "How does a group change an established interpretation?" "How does a group innovate and learn?" and "How are group structure and process a function of society?" In this perspective, "framed" relates to how a policy is understood in the existing context of beliefs and values.
Fourth, and most importantly, in the language of scholars focusing on FPC, all four of Hermann's primary change agents (or Gustavsson's three) are accounted for within the ACF theory of policy change. Hermann's multilevel approach is already unlike most FPA research, which tends to focus on one level and, thus, cannot capture the political interplay between government, society, and the international environment that impacts foreign policy. Uni-level approaches are incomplete (Risse-Kappen 1994;Rosati, Hagan, and Sampson 1994). The ACF overcomes this shortfall by incorporating external and domestic agents of change.
It not only integrates multiple levels to explain major foreign policy redirection, it further incorporates the decision-making process that exists between the inputs (the four agents) and the outputs (the four types of change). This makes the ACF a coherent framework for examining cases of FPC. The next section clarifies how the ACF accommodates Hermann's four primary change agents.

Hermann's First Concept: Leaders
Starting with the first agent, leaders, Hermann assumes that leaders occupy a central role in the decision-making process, perceiving and interpreting information using their own beliefs, which may affect policy choices significantly. The ACF also views individual actors, including leaders, as central; any individual trying to influence the relevant policy subsystem is a policy actor. The subsystem includes all the actors who are engaged with a policy issue, in this case foreign policy or even a subarea within foreign policy, such as policy toward the Middle East or even toward Iraq (e.g., Haar 2010; Hirschi and Widmer 2010; Lantis 2019). It should be noted that the ACF does not explicate a theoretical role for leadership that is found in leaderoriented approaches such as Hermann's Leadership Trait Analysis (2005) or Holsti's Operational Code (1970), a possible limiting factor to the ACF's applicability.
The ACF adds to FPA a broader scope of actors whose influence is considered and aggregates them into coalitions based on their common beliefs, and their ability to advocate for policy change or stasis (Sabatier and Jenkins-Smith 1999). In contrast, more traditional FPA scholarship often limits the examination to top decision-makers; investigations that single out a few centralized actors potentially exclude coalition members whose inputs might be crucial to change. According to the ACF, coalition members could include politicians, civil servants, members of interest groups, researchers, academics, and journalists who focus on the relevant area. Both Hermann (1990) and the ACF also predict that coalitions that include policymaking elites/leaders are more successful. This means groups will attempt to either place members in executive positions or sway leaders to become allies. Moreover, coalitions may persist for long periods, attempting to influence several governments or administrations. Hirschi and Widmer (2010, 540) point out that the ACF proved its value as a framework in analyzing policy processes "with a time perspective of a decade or more." Above all, the ACF buttresses foreign policy analysis and explanations of change because of the way it handles beliefs held by strategic political leadership (Jenkins- Smith et al. 2014). For instance, Hook (2008) emphasizes the importance of understanding beliefs to explain FPC in US foreign aid policy. More broadly, beliefs are an equally important concept for foreign policy analysis (Snyder, Bruck, and Sapin 1954;George 1969;Axelrod 1976;Holsti 1977;Goldstein and Keohane 1993;Kaarbo 2003;Hudson 2005;Carter and Scott 2009). Beliefs incorporate cognitivepsychological approaches as they relate to individuals; such approaches focus on how policy preferences are shaped by key policymakers (Rosati 2000). Rather than view the individuals in a coalition as rational actors, the ACF assumes that individuals have a limited ability to process information and that they have bounded rationality (Simon 1985). This means that decisions are made based on satisficing. Furthermore, consistent with prospect theory, individuals tend to weigh losses more than gains (Quattrone and Tversky 1988) and engage in cognitive dissonance whereby information is filtered by pre-existing belief systems (Lord, Ross, and Lepper 1979). Such a model of the individual is not rivalrous or exclusionary to other theories and assumptions about leaders more commonly used by FPA scholars.
This further means that the ACF disallows the realists' state (the one voice), which Gustavsson (1999, 84) rejects as "the reification involved when abstract collectives are ascribed human qualities." Instead, the ACF allows individual decision-makers to trigger alterations based on their beliefs, which they share with other members of a coalition. The ACF is able to investigate relevant individuals in detail, particularly those who have the greatest impact on a distinct policy area or in a specific process within established institutional structures. Indeed, the individual coalition members may have inside knowledge that helps them navigate established institutional structures and build alliances with leaders.
An advocacy coalition's beliefs are about both the international and domestic situation as related to the policy subsystem, in this case the foreign policy environment. In the ACF, beliefs operate on three different levels: deep core beliefs (fundamental moral or philosophical principles), near-core beliefs (related to policy preferences), and secondary beliefs (views on policy implementation). The core beliefs are highly resistant to change, while near-core and secondary beliefs are more negotiable. Like the previous models of FPC, the ACF scales change from minor to major, with significant policy change resulting from a revision in policy core beliefs and minor policy change stemming from an adjustment in secondary beliefs. Given the challenges associated with altering the status quo, understanding individual decision-makers' deep commitment to a set of beliefs might be necessary for altering the existing state of affairs. Alongside this insight, the ACF assumes that advocacy coalition members may learn within and/or across coalitions, depending on the experienced stimuli (Jenkins- Smith et al. 2014). For example, Schröer (2014) in his analysis of German security policy delineates the core beliefs held by relevant policymakers in the German security policy subsystem and considers whether they learned lessons from Germany's failed policy in Afghanistan.
Another important point is that the ACF is able to aggregate individuals into coalitions even at the state level-coalitions can cross institutions and states, thus drawing together internal and external processes. This means it is possible to construct a coalition of actors that includes multiple countries. In fact, explaining change at this level is one of the ACF's major contributions to foreign policy analysis. For example, Litfin (2000) and Pierce (2011) apply the ACF to explain FPC at the state level. However, the ACF does encounter challenges at the international level because it focuses on hierarchical policy change within the subsystem, which is not geographically bound. Moreover, at the international level, the territorial scope is not always identifiable and the only hierarchical jurisdiction is the Security Council at the United Nations.
In the ACF, the unit of analysis is the policy subsystem, which is characterized by three components: (1) the policy problem or issue, (2) the scope of actors seeking to influence a functional or substantive domain; and (3) a territorial domain or some authority or potential authority for policymaking (Jenkins-Smith et al. 2014). The third element, territorial domain, delineates the scope of inquiry, which subsequently raises obstacles to foreign policy applications because a clear territorial scope is not always identifiable (e.g., Richardson 1996;Farquharson 2003).
It is important to note that the territorial scope is a legacy of the ACF's origins in the environmental and energy sectors (Sabatier and Jenkins-Smith 1993). An alternative method for delineating the scope of inquiry is to focus on a government body that has authority or potential authority to make and implement foreign policy, such as a Foreign Ministry or an executive office (e.g., Pierce 2011). This method is a viable alternative for limiting the scope of inquiry given that the ACF prescribes that subsystems will include some authority or potential authority (Jenkins- Smith et al. 2014).

The Second Concept: Bureaucracies
While the placing of key individuals in strategic policymaking positions is an important factor in whether or not change occurs, fundamental structural conditionsthe bureaucracies of the decision-making process-are equally significant. A consideration of the structure of the decision-making process completes the picture of how policy is made, especially if it is assumed that each decision is not an ad hoc episode (Kuperman 2006;Lee 2012). The ACF also includes the concept of the policy venue, an institutional setting where policy advocates express alternative policies. Members of coalitions will strive to move a decision-making authority to decision settings (venues) that bolster their chances of success. Thus, effective coalitions will strive to overcome structural constraints found in administrative venues that explain inertia, especially in democratic structures in which coalition members are constrained by bureaucratic webs (Kleistra and Mayer 2001;Breuning 2013). For instance, in the case of Sweden's military intelligence policy, Nohrstedt (2011) postulated that policy change was linked to one coalition's success in opening new policy venues (from one ministry to another) that ultimately pushed intelligence reform forward.
An attempt to understand the political and structural constraints of the policymaking process has been a part of foreign policy analysis since Allison's (1969) classic on bureaucratic politics. Allison's goal was to understand what bureaucrats do, why they work in their particular way, and is there an organizational culture that influences the decision-making process. Hermann (1990) also finds that the nature and structure of bureaucratic agencies strongly affect foreign policy outcomes, with the general rule being the greater the change the stronger the bureaucratic resistance. Resisters to change include the standard operating procedures (SOP) present in every bureaucracy, the procedural scripts that structure a decision process, the cultural rationales that define bureaucratic interactions, and the bureaucratic roles that influence how a participant sees him/herself in the decision process.
The ACF accommodates these resisters, making clear that major policy change needs a source, which can emerge in a bottom-up or top-down fashion (Pierce and Hicks 2017). The impetus of dynamic socioeconomic factors that come from outside the policy subsystem may shift well-established governmental policy. Thus, the ACF framework takes into account the bottom-up sources that build up pressure for change. In top-down processes, bureaucracies can explain why decisions can take the form of results distinct from what leadership intended but "according to the power and performance of proponents and opponents of the action in question" (Allison and Zelikow 1999, 256). Meaning, the ACF takes into account any brakes on policy that a bureaucracy may construct and any bureaucratic infighting that takes place in a policy subsystem between coalitions, which might lead to contrary results from a leadership's perspective.
In Hermann's (1990, 11) understanding, the manner in which advocacy coalitions interact is labeled "bureaucratic advocacy," when a group within government advocates the redirection of policy. A bureaucratic advocacy group may be located in one agency or scattered among several agencies. Echoing strategies that Pierce (2011) delineates, Hermann indicates that effective bureaucratic advocacy groups have their members "sufficiently well placed to have some access to top officials" (Hermann 1990, 12). However, the ACF assumes that the members of a coalition that reside in an agency are more moderate in their views than members in other positions, such as those in think tanks (Jenkins-Smith et al. 2014).
The ACF identifies both domestic structural parameters (the politico-institutional setting) and relevant international constraints. As Blavoukos and Bourantonis (2014) argue in their piece on FPC, fundamental structural conditions exist at both levels. At the international level, structural parameters refer to both "systemic changes that may bring about foreign policy realignment" and "the country's role in the international system and its interactions with other countries that may activate foreign policy changes" (Blavoukos and Dimitris Bourantonis 2014, 483). Brazys, Kaarbo, and Panke (2017) add to the research on international structural parameters through their analysis of the effects of international norms on FPC. The norms associated with the current international system involve shared formalized and codified standards of behavior that influence state conduct.

The Third Concept: Changes in Domestic Constituencies
The framework further links policy change to the structural social and governmental aspects of the domestic policymaking process, thereby accommodating Hermann's third concept, changes in domestic constituencies. Herman (1990, 7) argues that those who create foreign policy "depend for their continuance on the support of certain constituencies," which sustain the regime. Comparable to the ACF's advocacy coalitions, Hermann's constituencies emerge from a variety of places in society: political parties, factions within parties, the military, interest groups, or leaders within society. The winning coalition for Hermann relates to the dominant alignment of the constituencies in a political system.
Applied to the ACF, members of advocacy coalitions compete to translate their policy core beliefs into public policy (Sabatier and Jenkins-Smith 1999). The members of these advocacy coalitions combine their resources to coordinate strategic actions seeking to influence the policy process. The ultimate goal is to influence government authorities to make decisions favorable to the policy beliefs of the advocacy coalition. Various strategies seek to frame their policy goals in such a way that they resonate domestically, including the "devil shift," a phenomenon where political leaders "perceive opponents to be stronger and more 'evil' than they actually are," in essence labeling them devils (Sabatier, Hunter, and McLaughlin 1987, 450). The successful use of such strategies by a winning coalition in its pursuit of translating beliefs into policy is clearly identified in Lee's (2015) analysis of the two competing views of American perceptions regarding the nature of the Soviet Union during the early Cold War years. The goal of these strategies and framing techniques is to realign the views of the many to create a dominant domestic constituency that can support a shift in policy (Risse-Kappen 1994). The coalition that is able to construct a message that successfully contests previously held views will win the policy battle. The degree of domestic support or the levels of indifference or even the levels of opposition to a policy will affect the chances of realignment.
Thus, the ACF framework is able to analyze the dynamic interaction between policy actors and social structures. The broader sociocultural-political context contributes to the policy subsystem and therefore may be linked to FPC. For example, Rynhold (2007, 419) finds that the cultural shift toward post-materialism in Israeli society led to the rise of a new more liberal generation that was responsible for "conceiving, initiating and enacting the Oslo Accords"-and thus representing a major shift in Israeli foreign policy. Battles over moral and practical concerns regularly take place in society, with elections in democratic societies affecting the domestic environment and the main attitudes held by relevant segments of society. A clear example of this in the United States is the breakdown of the domestic Cold War consensus in the 1960s and 1970s, as foreign policy views shifted in response to the Vietnam War and the rise of the insurgent George McGovern (Haar 2017).
Foreign policy issues may also become a focus in the struggle for political power. For example, the renegotiation of the 2015 international agreement between Iran and the P5 + 1 became a cornerstone for Donald J. Trump's winning coalition in 2016. As Herman points out, insurgent politicians and their supporters sometimes use a foreign policy position to set them apart from political rivals in order to gain power, and "if those out of power succeed, then the foreign policy changes" (Herman 1990, 7). Meaning change results from domestic factors that coalitions can exploit, but the effects are in the foreign policies the nation pursues. In a role reversal, Doeser's (2011) research on Danish foreign policy found that shifts in the domestic environment created opportunities for the Danish government to use FPC as a strategy to increase its political power domestically.

The Fourth Concept: External Shocks
Hermann's fourth concept, external shocks, is also unmistakably part of the ACF framework. In fact, the ACF expects that the potential to orchestrate change greatly depends on whether a coalition can take advantage of external shocks (which may be international in origin) to the policy subsystem. The notion of "policy windows," or opportunities that are useful for introducing change, is an idea introduced in 1984 by Kingdon's concept of "space windows" (Kingdon 1984, 174) and agendasetting in public policy. The notion of the policy entrepreneur, individuals who have expert knowledge and are committed to reform, is essential for Kingdon's account. Policy entrepreneurs are both inside and outside government bureaucracies; in Kingdon's explanation, the key to their activity is that they seize the opportunity created during a policy window to put their policy proposal on the agenda and then use various tactics to persuade others that their policy is the ideal solution to the problem at hand. If the window is a crisis, the sense of fear and urgency will unlock institutional structures, according to Keeler (1993), who refined Kingdon's thinking on policy windows. Keeler argued that a crisis situation creates greater prospects for a resolute policy entrepreneur to overcome lodged interests and institutional inertia. Gustavsson (1999) also argues that ideas from Kingdon and Keeler can easily fit in a model of FPC (which is also accounted for in the ACF). For example, in Hirschi and Widmer's (2010) examination of why Switzerland for the first time in history imposed economic sanctions against another state (toward Iraq during the 1990/91 Gulf War), external shocks were essential in allowing a younger generation of proponents to convince Swiss federal agencies, the Swiss government, and the Swiss parliament to adopt their established position on international sanctions. Gustavsson argues that building coalitions is necessary for reform, but he imagines that coalitions form after a crisis (or a policy window), while the ACF assumes that coalitions already exist of like-minded individuals who may strategically seek out allies in their pursuit to bring about their desired outcome.
Moreover, in the ACF, one clear pathway to policy change is a significant external event. As Sabatier and Weible (2007, 199) observe, "[e]xternal shocks can shift agendas, focus public attention, and attract the attention of key decisionmaking sovereigns." External shocks may not directly cause change but rather provide the window or opportunity for a coalition to affect policy. For instance, an external shock may lead to large-scale domestic discontent with the existing governmental policy that a coalition might contest and use a variety of strategies in order to bring about policy change. The domestic environment may then realign to meet the demands of a new powerful coalition. Sabatier and Jenkins-Smith (1999) also viewed the necessity of external shocks to a policy subsystem as a precursor for change. Sabatier and Weible (2007) subsequently included internal events to a subsystem or negotiated agreements between competing coalitions as further pathways to new policy formation.
External or internal shocks also redistribute resources within the policy subsystem. Often an external shock results in the replacement of the previously dominant coalition by a minority coalition. The policy that emerges after the shock will depend on which coalition is able to make good use of the circumstances and its own resources and which coalition successfully implements its previously devised strategies.

Decision-Making Process
Hermann's 1990 model not only delineates the three graduated levels of FPC (program change, goal/problem change, and international orientation change), and the four primary change agents (leaders, bureaucracies, changes in domestic constituencies, and external shocks), but it also outlines seven suggested stages that occur when a decision-making process results in significant change. Here Hermann unpacks the manner in which the sources of policy adjustment are processed inside the machinery of government and, similar to ACF expectations, the process is dynamic and competitive.
Unlike the rational choice model of decision-making, Hermann did not view a decision process that results in change as linear; instead, it involves pauses, cycles, and stages. Hermann's suggested seven stages of a decision-making process that results in change are as follows: (1) identifying the initial policy expectations, (2) recognizing environmental stimuli, (3) recognizing that the initial formulation of the problem is discrepant with the environmental stimuli, (4) postulating that there is a connection between the problem and current policy, (5) developing changes to the policy or shifting the definition of the problem, (6) building authoritative consensus for the preferred choice, and (7) implementing new policy consistent with committed policymakers' preferences. Hermann's stages supplement and open up the ACF's decision-making process within a policy subsystem, the ACF's focus of analysis.
In fact, the ACF expects a stage 5 shift in the definition of the problem. An advocacy coalition's potential to orchestrate change greatly depends on whether it can take advantage of external shocks-the opportunities that are useful for introducing change. Indeed, the ACF notion of a "policy window" is exactly the stage that Hermann (1990, 18) describes when a "major reconceptualization of the policy" takes place after internal actors, who do not have vested interests in the current policy, seek to apply their existing options directly, "or in a readily modified way, to the present problem." The ACF predicts that a rising coalition uses a shock to reinforce its position in the subsystem "by demonstrating that its belief system is best equipped to interpret and solve the policy problem" (Cairney 2015, 488).
Stage 6, the "choice stage" of the decision process, is equally important for the ACF, as here the various strategies and resources that actors in the coalitions bring to bear on the competition will determine choice. In stage 7, a policy change necessitates a committed advocacy coalition in order that policy is implemented in a manner consistent with a coalition's expectations and beliefs. This is because change must overcome bureaucratic inertia as well as strong advocates of policy stasis or powerful advocates of another option, who seek to implement policy in line with their own beliefs and preferences. Again, resources and strategies, such as whether the coalition can retain the attention of top leadership, determine whether the new policy is implemented fully or not. See figure 2, which illustrates Hermann's model in full.

Examples of ACF Employed to Explain Foreign Policy Change
In order to illustrate the ACF's transferability and inclusivity of Hermann's ideas to explain FPC, the following section provides examples of the ACF explaining FPC in two cases; namely, Nohrstedt (2011) andHaar (2010). Indeed, an important stated goal of Nohrstedt's research is to clarify how the ACF explicates policy change and what propositions might be derived from the ACF regarding the causal process that leads to major policy change. Haar's case analysis explicitly aims to explain how an advocacy coalition facilitated a dramatic change in government authorities and how it managed to prevail over other views and coalitions. The two cases were also selected for practical reasons: of the eight investigations that currently employ the ACF to explain FPC, Nohrstedt and Haar provided the most generalizable examples of the various strategies employed and the variety of resources exploited by members of the winning coalition to convert policy to reflect their beliefs and relevant solutions. Both cases also illustrate that all four primary change agents were present making fundamental policy change in the contested subsystem possible.
Nohrstedt tests his propositions empirically through a case study, the reform of Sweden's intelligence policy from 1999 to 2009. It was in 2009 that Sweden adopted an intelligence act that significantly expanded the monitoring mandate of the Swedish signals intelligence agency, the National Defence Radio Establishment (NDRE), meaning that the shift in policy was akin to Hermann's "International Source: Hermann (1990, 13) Orientation Change." The events on 9/11 in the United States also affected discussions in Sweden, which had been ongoing since the 1990s, about how to reform the intelligence "in order to enable more effective use of military intelligence to combat terrorism and to support Swedish foreign, defense, and security policy" (Nohrstedt 2011, 467). The winning coalition's assessment of 9/11 was that the nature of terrorism had changed significantly, which in turn justified a redefinition or broadening of the objects for military intelligence to monitor to include nonmilitary threats.
Within the Swedish intelligence subsystem, Nohrstedt finds two advocacy coalitions (in four different time frames from 1999 to 2009) that had a clear set of shared policy stances in mind. He further found that these shared policy views were underpinned by the coalition members' core beliefs on the role and the objectives of defense intelligence within society. The two coalitions held beliefs that were either pro-expansion or anti-expansion of Sweden's signals intelligence monitoring policy. Coalition members included agency officials, elected officials, interest group leaders, and researchers from forty-one different Swedish organizations.
Two parameters of the Swedish intelligence subsystem stalled the process toward reform in the 1990s. First, the two coalitions remained relatively stable over the time frame in question. Second, bureaucratic struggle and protectionism exhibited by the branches involved in intelligence policy prevented modification, despite a Swedish 9/11 inquiry that concluded that reform was warranted. Thus, in order for change to occur, the winning coalition, the pro-expansion coalition, necessarily utilized a number of strategies and exploited a variety of resources to convert policy to reflect their beliefs and relevant solutions. At the leader-driven level, these strategies encompassed placing coalition members in formal legal authoritative positions in the Swedish government, in particular the Minister of Defense, Mikael Odenberg, and the Director of the NDRE, IngvarÅkesson. The members of the pro-expansion coalition also skillfully used their parliamentary majority over their opponents, in part via the entrepreneurial talents of their comembers in formal legal authority. For instance, coalition member Minister of Defence Sten Tolgfors amended and resubmitted proposals to appease the opposition and Prime Minister Fredrik Reinfeldt publicly confronted the anti-expansionist coalition's criticism of the reform law.
At the bureaucratic advocacy level, strategies included crafting a convincing narrative to attract new members, switching to more accommodating venues (i.e., from the Ministry of Justice to the Ministry of Defence and to the "inner cabinet" of the government), amending policy to meet personal integrity concerns and democratic oversight issues, legitimizing policy through judicial review (via the Council of Legislation), claiming that anti-expansionist coalition members were uninformed about defense intelligence, increasing the number of written comments on proposals for legislation, and increasing the average number of appearances of coalition members in a variety of venues. Of these strategies, "venue shopping to realize policyoriented beliefs" was the most effective (Nohrstedt 2011, 465). This dovetails with the ACF, which expects that in their quest to overcome their subordination to dominant coalitions, minority coalitions constantly try to move decision-making authority to different institutional settings.
At the domestic restructuring level, strategies involved amending policy to offset fairly low levels of public support, restoring public confidence in the government through firm political leadership, increasing the number of actors involved in the public debate (i.e., academics and former bureaucrats from the NDRE), and mobilizing "sub-coalitions" of policy advocates in various venues. Given the classified nature of the topic, the winning coalition was not able to readily exploit information to win political battles against opponents in the public sphere, as the ACF predicts. As a result, the anti-expansion coalition won the argument at the domestic level. Nevertheless, as table 1 indicates, all four primary change agents were present and were fundamental to actual policy change in the contested subsystem of Swedish intelligence policy.
In the second example, Haar (2010) examines US FPC under the George W. Bush administration in its decision to go to war with Iraq in 2003, a historic decision that represented a significant shift in Bush's worldview. For example, in the 2000 campaign, Bush said he wanted to break with Bill Clinton's "Wilsonian interventionism" (Haar 2010, 978). Haar's study aggregates policy entrepreneurs, in this case the individual members of the neoconservative coalition, and examines their behavior after 9/11. According to the ACF conditions, the shift in foreign policy in this case resulted in a major policy change (Sabatier and Weible 2007) or as articulated within Hermann's (1990, 5) graduated levels of FPC, the case meets the Crafting a convincing narrative to attract new members; switching to more accommodating venues (from the Ministry of Justice to the Ministry of Defense and introducing proposals to the "inner cabinet"); amending policy to meet personal integrity and democratic oversight; legitimizing policy through judicial review (via the Council of Legislation); claiming that anti-expansionist coalition members were uninformed; increasing the number of written comments on proposals for legislation; and increasing the average number of appearances of coalition members for all venues.

Domestic restructuring
Amending policy to offset fairly low levels of public support; restoring public confidence in the government through firm political leadership; increasing the number of actors involved in the public debate (i.e., academics, and former bureaucrats); mobilizing "sub-coalitions" of policy advocates in various venues External shock Terrorist attacks on 9/11; 2006 regime shift in Sweden criteria of an "International Orientation Change," involving the redirection of an actor's entire orientation toward world affairs. Within the policy subsystem, the neoconservative advocacy coalition had a clear set of shared beliefs that were underpinned by the core belief "that the moral purpose of U.S. power is to change the world for the better" (Haar 2010, 970). Other consolidating principles that impacted the coalition's policy choices were a desire to remake the Middle East through the removal of Saddam Hussein and that America should use its prodigious military power to bring about a free Iraq. These beliefs were stable over time (at least a decade), having emerged strongly after the Gulf War in 1991 when George H. W. Bush chose not to remove Saddam Hussein from power.
The actors in the subsystem came from a variety of public organizations, including government agencies and departments, the media, think tanks, and academia. Bush appointed many members of the neoconservative coalition to administration positions. Already in January 2001, members of the coalition began working on a military blueprint to implement their desired policies in the Middle East and specifically in Iraq. However, their ideas did not find any traction until after the terrorist attacks on September 11, 2001. The neoconservative coalition utilized a number of strategies and exploited a variety of resources to change existing policies to reflect their preconceived solutions. At the leader-driven level these encompassed: acquiring allies and access to the highest levels of policymaking, including the President, the Vice President, the Secretary of Defense, and the US Ambassador to the United Nations.
At the bureaucratic advocacy level, strategies included: crafting well-defined and coherent policy goals over the 1990s, managing or using information selectively to convince decision-makers, securing funding for research to support coalition goals, disseminating research that affirmed coalition goals, placing coalition members in formal and legal positions (i.e., the Deputy Secretary of Defense, the Undersecretary of Defense, the Chairman of the Defense Policy Advisory Group, and several Crafting well-defined and coherent policy goals; managing or using information selectively to convince decision-makers; securing funding for research to support coalition goals; disseminating research that affirmed coalition goals; placing coalition members in formal and legal positions; switching to more accommodating venues when their advocated policy was challenged; and pressuring intelligence agencies to validate their claims Domestic restructuring Securing allies in the media to help mobilize the attentive public; harnessing public opinion through media venues; utilizing key external supporters in think tanks and academia; securing funding from sympathetic foundations; and founding new think tanks to promote coalition principles and policy goals External shock Terrorist attacks on 9/11 members of the National Security Council staff), switching to more accommodating venues when policy was challenged, and pressuring intelligence agencies to validate their claims. At the domestic restructuring level, strategies involved: securing allies in the media who then became key in mobilizing the attentive public; harnessing public opinion through talk radio, cable news, popular magazines, and newspapers; utilizing key external supporters in think tanks and academia to host conferences and write policy papers, editorials, and opinion pieces that propelled their agenda forward; securing funding from sympathetic foundations; and founding new think tanks to promote coalition principles and policy goals. As table 2 indicates, all four primary change agents were present and essential to bring about FPC on US policy toward Iraq.

Conclusion
The ACF is one of the most widely tested theoretical frameworks in public policy research today, with hundreds of applications spanning over fifty different countries (Weible et al. 2009. Thus, the ACF addresses another general criticism of foreign policy analysis-that it is too North America-centric (Brummer and Hudson 2017). In fact, today the ACF is more extensively applied by European academics to understand various policy processes than by US scholars . Furthermore, the ACF is already widely used to explain policy processes including policy change and learning in a myriad of non-Western contexts such as China and South Korea (Henry et al. 2014;Jang, Weible, and Park 2016;Li and Weible 2019).
The ACF also has the potential to further augment the rich tapestry of foreign policy analysis. Moreover, it has distinct advantages to better understand FPC. First, it focuses on the nexus between the domestic and international systems. It explicitly connects domestic policy subsystems with broader international ones such as the United Nations (e.g., Elliott and Schlaepfer 2001) and the European Union (e.g., Nedergaard 2008). Second, the ACF shifts the focus from being decision-makingcentric to advocacy-centric. This has two major benefits in terms of broadening the scope of analysis in future research. First, the ACF incorporates a wide-ranging spectrum of actors including political leaders, members of foreign policy-making agencies, and actors outside of government. Such an expansive spectrum of actors is clearly shown in Haar's (2010) analysis of US foreign policy toward Iraq. Second, an advocacy-centric rather than a decision-making-centric focus has the potential to result in more comprehensive studies. By broadening the scope of research to include nongovernment policy actors as well as expanding the policy process to years or even decades, a researcher has the capacity to investigate why an FPC occurs over a longer time span.
Advantageously, the research methods used by ACF scholars to understand policy change are similar to existing foreign policy analysis methods (Brummer et al. 2019). ACF scholars use case studies that often cover multiple years and that rely on a myriad of sources, such as media and government reports, executive and foreign policy agency memorandums, legislative hearings and debates, executive speeches, and interviews with policy actors within government and those also outside government (see  for a review of ACF research methods).
Thus, the ACF offers foreign policy analysis a well-established framework that broadens the boundaries and scope of research. By looking at the nexus of domestic and international politics, by focusing on the beliefs of government and nongovernmental actors, and by extending the time frame to years or even decades, the ACF pushes the boundaries and limits of foreign policy analysis to new places. Additionally, this endeavor to expand the capabilities of research builds on the previous work done by Hermann (1990) and Gustavsson (1999). The ACF provides a framework that has recognizable value for the foreign policy community in its understanding of how and why foreign policy changes over time.