Introduction
The introduction of ideational variables into positivist frameworks of inquiry has routinely posed considerable theoretical and methodological challenges with regards to compatibility. To be sure, ‘strict’ positivists do not consider ideation sufficient or even necessary for the development of scientific accounts of social life, since beliefs and ideas do not lend themselves to public observation.Footnote 1 Those ‘softer’ positivists who do consider ideational variables causally relevant inevitably encounter difficulties of operationalisation, measurability, and causal circularity when using ideational or cognitive concepts.Footnote 2 Imprecise specifications of different types of ideation and their respective role in the causal process have led to considerable confusion as to the role of ideas in policy formulation within such ‘soft’ positivist accounts.Footnote 3 Correspondingly, these problems are encountered in most if not all paradigms that grant ideas a causal role within a loosely positivist framework, including thin constructivism, (neo)liberal institutionalism, and even some variants of realism.Footnote 4
In this article, I use one such realist variant, neoclassical realism (NCR), as a case study to investigate the integration of ideation in a (soft) positivist approach to state behaviour. More specifically, I discuss if and how NCR can embrace concepts frequently associated with constructivist (and institutionalist) frameworks in a theoretically sound way.Footnote 5 I argue it must uphold the definitional distinction between interests and ideas by ontological necessity. A (re)conceptualisation of ideas as beliefs externalised into political deliberation through language allows neoclassical realists to avoid the equation of different ideational concepts and focus on the role of competing ideas in foreign policy processes.
Notably, this conceptualisation of ideas as variables distinct from yet intricately linked with beliefs and language can similarly be applied in other positivist-oriented frameworks. In fact, NCR serves here as a least likely case of integration with regards to ideational variables because most neoclassical realist analyses are also committed to some form of materialism as well as a loose rationality assumption. This expands on J. Samuel Barkin's discussion of the three charges against the compatibility of realism with constructivist concepts,Footnote 6 and paves the way towards further reflection on the theoretical assumptions, differences, and overlap between IR paradigms, not least in the empirical study of foreign policy.Footnote 7
To be theoretically successful, negotiating the integration of ideational factors into a neoclassical realist framework therefore requires grappling with the approach's ontological and epistemological commitments. One way to do so for neoclassical realists has been to embrace open, reflexive, and even interpretivist variants of NCR aligned more closely with classical realist authors.Footnote 8 After all, the use of ideation can be traced in (classical) realist authors, such as Niebuhr, Carr, and Morgenthau, who dwelled on the extent to which moral principles and ideas about appropriate behaviour would lend decision-makers the ability to affect courses of action beyond the mechanistic determinism of the international order.Footnote 9
Another, pursued in this article, is to explicitly examine the incorporation of ideas and beliefs into a framework oriented in its core commitments towards neorealism, that is, as a positivist, rationalist, and materialist theory of state behaviour. If the use of ideational factors can work for a thusly defined NCR, it can work elsewhere in principle.Footnote 10 Indeed, providing a theoretical way forward for materialist, rationalist, and positivist NCR-variants may also indicate similar venues for other approaches, for example, more positivist minded variants of (neo)liberalism and constructivism.Footnote 11 Such shared applicability highlights points of convergence and conceptual overlap. It is different, of course, to genuinely bridging paradigmatic approaches, as in contrast to neoclassical realists, thin constructivists view ideas and norms as fundamentally defining the intersubjective, social reality within which actors exist as well as the interests that those actors hold.Footnote 12 Similarly, institutionalists may lend ideas and ideational discourses a more pronounced role in shaping the institutional landscape in which states operate.Footnote 13
At the same time, this article also speaks to the recurrent debate within and outside NCR as to its paradigmatic boundaries. To its proponents, NCR's use of intervening variables promises a way in which the unit-level specifics of the foreign policy process can be made causally important for state behaviour. The initial insight concerns the limitations placed on the state by domestic factors in responding to systemic incentives and constraints, which neoclassical realists introduce as intervening variables. However, the proliferation of these variables has led some authors (including from within the approach) to charge NCR with incoherent, ‘degenerative’ and regressive theorising that falls outside the boundaries of realism.Footnote 14 It risks producing an ‘identity dilemma’ between either clearly (neo)realist but indeterminate or precise but indistinctive explanations.Footnote 15 Neoclassical realists can address this critique by clarifying how their contributions increase precision yet remain distinct, that is, committed to assumptions of positivism, rationality, and materialism.
I suggest one such clarification regarding the use of ideas in NCR. Firstly, I argue the root of ideas lies in individual beliefs. Individual beliefs help decision-makers appraise the international environment by filling gaps of knowledge about the material world. However, to matter in foreign policy deliberations, individual beliefs must have interpersonal relevance. Secondly, therefore, ideas function deliberatively as externalised beliefs. Decision-makers feed their individually held diagnoses of a respective decisional scenario into foreign policy deliberations, where they argue with and try to persuade other decision-makers. Given a specific context, decision-makers employ ideas to deliberate the implications of a systemic environment for state ehavior and make decisions about the costs and benefits of alternative policy sets. I demonstrate that this is consistent with the retention of realist commitments to the causal primacy of materially given interests, to (bounded) rationality, and to ‘soft’ positivism.
Neoclassical realism and intervening variables
NCR differs from other realisms primarily in the degree to which the translation of systemic conditions into state behaviour is explicitly analysed rather than either simply assumed or treated as a conceptually distinct sphere of inquiry.Footnote 16 This is possible because NCR has a less strict understanding (compared to neorealism) of the constraints systemic conditions impose on state behaviour. In neorealist approaches, decision-makers are assumed to perceive systemic conditions, deduce the national interest, and pursue a foreign policy accordingly. Foreign policy choice exists in only two situations: either there are two options that have the same costs and consequences, which makes choice redundant.Footnote 17 Alternatively, decision-makers fail to interpret systemic conditions accurately.
Indeed, neoclassical realists argue, this is the case more often than not. Systemic conditions ‘merely [set] parameters’ for state behaviour.Footnote 18 They are neither specific enough to guide decision-makers to only one possible course of action, nor obvious even to the most adroit observer. The decision-maker's task is rendered difficult for several reasons: firstly, decision-makers ‘may not necessarily track objective material power trends closely or continuously’.Footnote 19 Also, they may be uncertain which state behaviour is appropriate in response to a given scenario.Footnote 20 Finally, decision-makers are limited in their choice by domestic political conditions.
NCR's core contribution has thus been its attempt ‘to explain variations of foreign policy over time and space’ by supplementing neorealist assumptions on structure and material capabilities with intervening variables mediating the impact of systemic stimuli.Footnote 21 All the while, most neoclassical realists seek to uphold neorealist commitments to materialism (giving pride of place to materially given interests derived from systemic stimuli), positivism and an at least implicit assumption of rationality.Footnote 22 Not least because of this broad argumentative thrust, some authors have argued NCR has developed into the ‘new orthodoxy’ for scholars aligned with realist thought.Footnote 23 However, this breadth has also made some (including from within the approach) criticise NCR because of an eclectic range of intervening variables, degenerative theorising and unclear commitments to the realist paradigm.Footnote 24 NCR is argued to risk an ‘identity dilemma’ between either clearly realist but indeterminate or precise but indistinctive explanations.Footnote 25
Some neoclassical realists have (reluctantly) engaged this general critique,Footnote 26 or argued that better explanations and intellectual diversity may well be more important than paradigmatic purity.Footnote 27 Others have attempted to systematise and even unify the approach.Footnote 28 However, this may obscure the theoretical differences between various authors and their analyses, and unnecessarily complicates the framework where it suggests that intervening variables operate cumulatively. Alternatively, yet others have attempted to categorise rather than integrate different approaches. For example, neoclassical realist approaches may be ordered according to the degree to which they deviate from neorealism.Footnote 29 This clarifies that neoclassical realists face different degrees of problems associated with the possible violation of paradigmatic assumptions depending on their objects of study and the variables they employ.
Firstly, then, some neoclassical realists treat intervening variables as incrementally adding explanatory value to neorealism: when systemic variables cannot sufficiently explain state behaviour, intervening variables account for the error. This version of NCR becomes a ‘subordinate and subsidiary component’ of neorealism.Footnote 30 Secondly, for other neoclassical realists, the intervening variables work permanently rather than in exceptional cases. Decision-makers always consider systemic conditions in the light of domestic politics and increasingly more so when systemic conditions favour it, for example, in security-abundant environments.Footnote 31 Along these lines, numerous authors have focused on factors such as institutional constraints, public opinion and media reporting, mobilisation capacity, and the extraction of resources.Footnote 32 Most of these variables are (directly or indirectly) measurable and usually grounded in some material fact. They do not, therefore, constitute a major problem for commitments to (soft) positivism or materialism. However, neoclassical realists also investigate the role of ideation in the translation of systemic stimuli into state behaviour.Footnote 33 This is wrought with conceptual difficulties: particularly, it remains unclear whether the inclusion of ideational variables upholds the same paradigmatic commitments.
Perceptions, beliefs, and ideas in neoclassical realism
In NCR, three different types of variables connected to ideation are employed frequently: perceptions, beliefs, and ideas. The authors usually share a common starting point: that decision-makers have ‘predispositions that lead [them] to notice certain things and to neglect others’.Footnote 34 In Zakaria, for example, states only expand their interests abroad once ‘decision-makers perceive a relative increase in state power’.Footnote 35
Focusing on the role of perceptions and/or beliefs introduces a version of ideation inspired by and derived from cognitive approaches in foreign policy analysis and political psychology. It supplements expectations of rational behaviour (at least) implicit in neorealism with a more realistic view of information gathering and processing in decision-makers.Footnote 36 In so doing, however, these neoclassical realists do not usually follow up on the origins of perceptions, or on the use of perceptions in decision-making. Where do perceptions originate? How do decision-makers feed their perceptions into foreign policy processes? Instead, aspiring to ‘parsimony’ and generalisation, these authors frequently focus on change in perceptions across different decision-makers or cases.Footnote 37
In turn, beliefs may serve as an individually anchored cognitive basis for perceptions.Footnote 38 Beliefs are associated with specific individuals, such as prime ministers or presidents, and inform how they see the world, process information, and steer foreign policy.Footnote 39 As detailed below, this has conceptual advantages, especially regarding consistency with (bounded) rationality, but also suffers from a major shortcoming: to matter in political processes involving multiple people (for example, presidents, advisers, experts, generals), beliefs must have some interpersonal effect.
By contrast, yet other neoclassical realists focus on social ideas, ideology, culture, and national identity.Footnote 40 Caverley, Dueck, Schweller, and K. Smith interrogate paradigms, programmes, or worldviews as ideational sets that influence foreign policy and grand strategy.Footnote 41 N. R. Smith suggests the ‘normative power role identity’ of the European Union intervenes in its foreign policy.Footnote 42 Dueck, Kitchen, and Onea suggest strategic cultural paradigms can help explain grand strategy, arguing that ‘ideas have a role in the making of grand strategy because they help specify national interests amidst conditions of uncertainty’.Footnote 43 Ideation thus fills the gap between indeterminate systemic stimuli and concrete policies by narrowing down the range of policy options.Footnote 44
These approaches are faced with difficult conceptual questions. For one, employing ideation as a causally influential variable may fundamentally challenge assumptions of materialism because it is not easily anchored in any specific material fact. Two, as internal mental events, can ideas be operationalised in a (soft) positivist framework? Finally, how does the use of ideation fit with a principally rationalist approach to state behaviour?
In the following, I suggest amendments that aim at making ideation work in NCR while explicitly upholding paradigmatic commitments to materialism, positivism, and rationality. Firstly, I clarify the difference between materially given interests and ideation, and their interrelation as independent and intervening variables, respectively. Secondly, I suggest that the distinction between beliefs and ideas is conceptually necessary, but that the two concepts are also interlinked through a process of externalisation. Thirdly, such a reconceptualisation allows ideas to have interpersonal relevance in ideational competition during policy deliberations. It also renders them more easily operationalisable in positivist research. This may suggest a way forward for other (soft) positivist approaches to international politics that seek to integrate causally relevant ideation into their frameworks.
Material reality, interests, and ideation
In neorealism, a state's interests are primarily derived from material reality, that is, the distribution of capabilities and relative power, in an anarchical system. Opportunities and threats emanating from this environment are grounded in the distribution of material capabilities among the system's units.Footnote 45 The interplay between units within an anarchic structure provides at the very least a baseline interest in survival, and likely interests in security, power, welfare, prestige, and so forth, that is, ‘the things a state must secure if it is to maintain its place in the balance-of-power’.Footnote 46 These interests can in principle be known as fact.Footnote 47 Any difference in culture, ideology, or other ideation matters only at the margins, and can ultimately be reduced to ‘particular expressions’ of material capabilities.Footnote 48 Note that this form of materialism, which most neoclassical realists share,Footnote 49 is not quite the simplistic materialism at times alleged by its (constructivist) critics, that is, a reduction purely to natural phenomena and technology.Footnote 50 Instead, it acknowledges, even necessitates, social interaction between units in the system.Footnote 51 The decision-maker is tasked with deciphering these interactions, translating them into threats and opportunities for foreign policy, and pursuing the state's interests accordingly. What makes NCR realist is the shared assumption that ultimately, state behaviour is a product of ‘environmental compulsion’.Footnote 52
The accurate appraisal of systemic incentives and constraints, and their translation into state behaviour, however, is complex, time consuming, and fraught with difficulty, especially so under conditions of uncertainty, lack of or ambiguous information, and time constraints. It is here that neoclassical realists have argued neorealist explanations are insufficient, and that ideation helps to guide decision-makers towards alternative policy options. Ideation matters because it handles practical problems related to the perception, interpretation, and translation of ambiguous external incentives and constraints.Footnote 53
To function as an intervening variable (rather than as independently causal) in NCR, such ideation must follow interests in causal sequence. Neoclassical realists seek to evaluate ‘the causal impact of specific hypothesized independent variables (IVs) and intervening variables (IIVs) on the dependent variables (DVs)’.Footnote 54 The commitments to discernible variables or events (‘both contingent and contiguous’),Footnote 55 and to the independence and temporal priority of cause over effect, rather than constitutive or iterative effects between different variables, means that NCR employs a Humean concept of causality.Footnote 56
At the same time, beliefs or ideas are not simply ‘free-floating’.Footnote 57 Instead, specific individuals and specific beliefs gravitate towards one another. Some groups of individuals find it in their interest to hold some ideational content rather than another that they then ‘elect’ because of affinity, that is, ‘a point of coincidence or convergence’.Footnote 58 For Max Weber, such ‘elective affinity’ may describe the relationship between Protestant ideas and capitalism. It may similarly help to understand, for example, the prevalence of beliefs surrounding military technology (for example, the presumably risk-free nature of drones) among specific US or UK decision-makers.
This relationship between interests and ideational factors is not strictly determining. Rather, it implies a higher likelihood (but not certainty) that people hold beliefs that mirror their interests, or because it is in their interest to do so.Footnote 59 Ideation that corresponds to people's respective interests is likely to influence their behaviour. Since ideation is conceptually different from interests, however, and no direct causality or determinacy is implied, it may still contradict interests or have unintended consequences. In neoclassical realist terms, then, ideation may be erroneous given material facts, even though those beliefs or ideas that run counter to national interests are less likely to gain and sustain support among decision-makers.Footnote 60 Indeed, if decision-makers come to hold ideas that deviate too much from objective reality, that is, if the state is co-opted by parochial ideation, they risk strategic failure because the system's material reality persists and punishes continued digression.Footnote 61
Beliefs and/or ideas
Beliefs guide decision-makers in their interpretation of the international system. Beliefs are ‘views or opinions held by political actors that are relatively limited in scope or relate to relatively circumscribed areas of politics’.Footnote 62 Beliefs function as cognitive devices or mental shortcuts that help to order and simplify complex situations, and ‘specify national interests amidst conditions of uncertainty’.Footnote 63 By this understanding, beliefs are cognitive products of the human mind, inherent and internal to the individual. They help individuals fill gaps of knowledge about the material drivers of international politics, so that decision-makers start from ‘predispositions that lead actors to notice certain things and to neglect others’.Footnote 64 They contain information as to the relevance of some material facts over others in specific contexts. They do not alter the primary motives of state behaviour but rather make their pursuit less efficient.
Different authors have developed a frequently three-pronged system for classifying different beliefs according to their generality, for example: (1) specific policy or causal ideas/beliefs;Footnote 65 (2) beliefs that underpin the policy ideas, also ‘paradigms’Footnote 66 or ‘programmatic beliefs’;Footnote 67 and (3) philosophical ideas, for example, ‘public philosophies’,Footnote 68 ‘deep cores’,Footnote 69 or worldviews. Causal beliefs have probably been employed most frequently in empirical policy analysis and infer a connection of causality between events, peoples, and things.Footnote 70 Such causality can be narrowly understood, as when beliefs relate to, for example, cause-effect chains between two sequential events. It can also be broadly applied, by establishing a general relationship of some sort in an individual's mind between, for example, a specific foreign policy stance and societal trends or vice versa: that the post-Cold War order influences German foreign policy positioning is not directly causal but assumes a linkage between order and state behaviour. The role of causal beliefs in decision-making is then understood (narrowly) as providing strategic guidance for human action. Causal beliefs ‘provide guides for individuals on how to achieve their objectives’ and ‘imply strategies for the attainment of goals’.Footnote 71
Focusing on individual beliefs has major advantages: for neoclassical realists, it upholds commitments to rationality, and to prioritising material over ideational factors in explanation.Footnote 72 By extension, focusing on beliefs corresponds with (methodological) individualism. The rationality assumption and individualism may be understood as analytically connected because the former tends to take the individual as the basic unit of analysis for specific instances of rational choice. Social phenomena are then understood as based in the actions, decisions, and motivations of individuals. A focus on individual beliefs distinguishes these approaches from others treating ideation as socially produced.
The core assumptions of any type of rationalist approach lie in the context-dependence of rationality, the material nature of context, and the rational character of human conduct.Footnote 73 Substantive rationality implies a rational process has led to a decision that can be shown to be optimal.Footnote 74 This can, in principle, be attained if interests can be known as material facts. However, both ambiguous international environments as well as the empirical reality of decision-making put unrealistic conditions on the processing and scope of available relevant information. Predictions derived from substantive rational actor models are frequently deficient for specific empirical outcomes precisely because ideas do not matter in them. Instead, rationality may refer less to the attainment of optimal outcomes and more to the process of decision-making.Footnote 75 Decision-makers do make cost-benefit calculations about options and their likely outcomes. Under conditions of uncertainty over systemic stimuli and even more so under constraints of time and information, however, decision-makers are bound by their human nature, which leads them to favour simplicity and employ ideational heuristics to simplify the world. Decision-makers’ beliefs function as ‘inputs’ into rational decision-making.Footnote 76
However, a focus on individual beliefs as influential in the policy process has at least two major shortcomings, both of which transcend NCR: they relate to positivism and to the social impact of ideas. One, as outlined, NCR shares with neorealism (as well as other mainstream approaches to IR) a commitment to (soft) positivism, that is, positivism paired with an awareness of the inherent limitations of social scientific research.Footnote 77 Positivists generally maintain that it is possible to objectively view observable phenomena in a world that exists independently of the researcher. This observable reality creates the conditions for social phenomena, such as the foreign policy process. In principle, this allows for the development of theoretical models, for example, that an independent variable causes a dependent variable to vary, which can be tested against empirical evidence.Footnote 78 In practice, the study of human behaviour introduces limitations related to subjectivity, interpretation, unpredictability, and research ethics, which makes it difficult to define, measure, and evaluate the object of research accurately. This holds especially for essentially unobservable, cognitive events such as individual beliefs. Some positivists have suggested that beliefs can be observed indirectly, that is, through their effect on the dependent variable. This, however, creates a causal circularity that is theoretically irrefutable and empirically unfalsifiable.Footnote 79
Two, the definitional exclusion of social effects and dimensions makes a focus strictly on individual beliefs conceptually and methodologically awkward. Beliefs cannot be treated both as internal mental events and as publicly employed tools in policy deliberations external to specific individuals, or as coordination points under conditions and uncertainty.Footnote 80 Such a conflation is evident in authors using simultaneously, at times literally in parallel, ‘ideas’ and ‘beliefs’.Footnote 81 For example, in contradiction to their own definition Goldstein and Keohane recognise that ‘ideas can have an impact even when no one genuinely believes in them’.Footnote 82 And yet, politics is an interpersonal process of deliberation and persuasion. Even in highly centralised systems where single individuals carry more decisional weight than others, multiple individuals will be involved in decision-making. Bar a conceptual mechanism by which multiple individuals are theorised to exchange their internally held beliefs, it remains underspecified how beliefs influence the transmission belt from systemic stimuli to state behaviour.
Conceptualising ideas as externalised beliefs can circumvent these problems. Individual beliefs have interpersonal relevance and matter in policy deliberations when they are introduced by individual participants, for example, decision-makers. Rather than achieve relevance purely by their existence as mental states, beliefs must be ‘championed’ by carriers, who are ‘individuals or groups capable of persuading others to reconsider the ways they think and act’ vis-à-vis given systemic conditions.Footnote 83 This fits with an intuitive understanding of the social impact of ideas, and has been picked up frequently in relevant literature: ideas are beliefs ‘introduced’Footnote 84 or ‘inserted’ into political deliberations by ‘carriers’,Footnote 85 or ‘political entrepreneurs’Footnote 86 who are ‘advocates for proposals or for the prominence of an idea’.Footnote 87 While discursively relevant, then, ideas remain anchored in individual action and use (‘individuated’).Footnote 88 Externalised into deliberations, ideas provide ‘an edge in the battles of the day or guidance through uncertain periods’.Footnote 89 Decision-makers seek ‘to make policy decisions reflect their preferred interpretation’Footnote 90 and utilise ideas to compel their opponents ‘to endorse a stance they would otherwise reject’.Footnote 91 They do so through language, here understood as a conduit of ideation, which enables political debate, which in turn consists of the exchange of (sometimes incommensurable) ideas.Footnote 92 Ideas are thus deliberative tools, wielded instrumentally and strategically in an interactive process to communicate, exchange, and convince by their use in language and text, from which they can be extracted.Footnote 93
Such a semi-discursive conception of ideas is ontologically ‘adequate’: ideas can be created by particular people (based on beliefs), distributed (in communication), embedded in causal sequences (as variables), and changed (for example, they can be forgotten).Footnote 94 Ideas as externalised beliefs are conceptually distinct from more collective, coherent, and durable phenomena, for example, ideology or culture. They remain anchored to individuals as their carriers. Per their use as intervening variables, they do not have independent causal force. At the same time, the importance of ideas as points of political convergence increases the more uncertain actors are about the scenario and the exact consequences of different alternatives they choose from.Footnote 95 Their effect is therefore likely more easily traceable in short-term decisional processes (see below).
Notably, such a reconceptualisation is not necessarily exclusive to NCR either in its theoretical roots or in its potential application. Explicitly defining ideas as beliefs externalised into communication may help overcome some of the definitional awkwardness identified by Mark Laffey and Jutta Weldes, all the while keeping in place conceptual distinctions between interests and ideas, on the one hand, and commitments to Humean causality, on the other. This awkwardness relates to the equation of beliefs, ideas, and a set of other ‘idea-like entities’ in (thin) constructivist and institutionalist research.Footnote 96 It follows up on the intuition that beliefs as individual mental events and ideas as the externalisation of such beliefs are by definitional necessity different conceptual entities, but also intricately linked. As suggested above, in contrast to constructivists who can make use of this reconceptualisation to clarify their conceptual understanding of ideas, neoclassical realists cannot grant ideas thusly defined a constitutive role for interests, or an independently causal effect on policy outcomes.
This twist also helps to address the operationalisation problem outlined above because the translation from individual beliefs to the use of ideas in (social) deliberation is explicitly conceptualised. To know which beliefs a person had at any one point in time is difficult (if not impossible). This is because the nature of decision-making deliberations may create pressures for individuals to strategically rather than genuinely communicate. They may have incentives to misrepresent their beliefs, exaggerate the relevance of ‘good’ motives,Footnote 97 or conform with majority opinion and/or with those who hold power.Footnote 98 While ideas are properly understood to originate in individual beliefs, they have explanatory power when employed in deliberation. Highlighting the difference between beliefs as internal products and ideas as deliberative tools allows the extraction of ideas from various ‘conduits’ (for example, texts, speeches).Footnote 99 This directly operationalises ideas and links them to a measurable, empirically traceable phenomenon that is more evidently either present or not. As Alan M. Jacobs highlights, an explanation concerning the impact of ideas on decision-making must ‘find significant verbal references to the ideational constructs hypothesized to have been influential’.Footnote 100 As outlined above, it helps avoid a potential causal circularity.Footnote 101
This understanding of ideas, while consciously embracing non-realist conceptual roots, may still retain commitments to materialism, rationality, and (soft) positivism. Arguably, many neoclassical realists seem to already implicitly share it, for example, where they analyse memos, meeting minutes, speeches, and so forth. It also points towards the role of conflicting ideas in foreign policy deliberation, rather than emphasising ideas as ‘shared’ entities that evoke consensus among elites.Footnote 102
Ideational competition and the foreign policy process
Given a systemic stimulus (for example, a scenario that requires a foreign policy response), members of the foreign policy elite propose actions they deem appropriate in answer to the perceived situation, and base these proposals on their individually held beliefs, that is, externalise them into deliberation as ideas. If, as argued, systemic incentives are often not specific enough as to determine a course of action, they may lead decision-makers towards vastly different alternative foreign policy stances that cannot be predicted as the result purely of either systemic conditions or ideological alignment. Even within an administration or a close circle of advisers, then, people have different, indeed incommensurable beliefs, which they carry into the foreign policy process as ideas. They are likely to reach consensus with those individuals that employ similar ideas and run into competition with others. For example, such a conceptualisation may well help to explain the uncertainty and internal divisions apparent in the Chinese foreign policy elite on the implications of China's rise for its own foreign policy.Footnote 103
The competition between divergent ideas in the foreign policy executive concerns the interactive process during which decision-makers engage in communication.Footnote 104 This process is not characterised by internal or ideological consistency, but instead by contradiction and confusion over how to correctly deduce systemic stimuli. Ambiguity opens space for politics as decision-makers seek to utilise their ideas to persuade their opponents to endorse a policy they would otherwise reject. As per a longstanding tradition in foreign policy analysis, foreign policy deliberations can then be interpreted as a process of ideational competition: the ideas that decision-makers carry ‘vie with one another for dominance and autonomy’.Footnote 105 Decision-making is then characterised by ‘pulling and hauling among individuals with different perceptions at stake’.Footnote 106 Embedding this process as intervening in the transmission belt from causal systemic variables to state behaviour renders the framework neoclassical realist rather than constructivist or psychology-focused.
Different such ideas are frequently combined into distinct ‘theories of grand strategy’Footnote 107 or even ‘national ideologies’Footnote 108 in which an interpretation of the international system corresponds with a normative agenda and a set of means to achieve ends. In US foreign policy, for example, numerous authors have debated the analytical merits of distinct ideational frameworks or paradigms, from isolationism and imperialism to the four ‘traditions’ of US foreign policy.Footnote 109 Grouping specific ideas into such internally coherent paradigms is useful analytically, for example, to identify longer-term trends in grand strategy, and conceptually convenient to order a complex array of different beliefs.
However, the empirical role of these paradigms for the foreign policy process may be overstated. For one, ideas also matter in shorter-term decision-making and for the choice of means and tools of foreign policy.Footnote 110 The nature of such decision-making tends to make clear-cut answers derived from broader ideological frameworks difficult. More importantly, it is my contention that ideational competition characterises any process concerned with the diagnosis and deliberation of systemic conditions and alternative policy options, no matter what political party, ideological affiliation, or part of the bureaucracy decision-makers adhere to. For example, assertive policy ideas are found across party lines, and ideas of restraint are equally represented across both the partisan as well as the theoretical aisle.
Moreover, then, the variable of ideas can be used to deconstruct notions of ideological or paradigmatic consistency and tease out constitutive parts rooted in individual beliefs. Strategic culture or national identity may then be understood not as holistic things that propel a state's behaviour in unexpected directions, but as contested arrangements of different competing and even contradictory ideas influencing decision-making processes.Footnote 111 Methodologically, the intervening causal impact of ideas as externalised beliefs may be more easily detectable in shorter-term foreign policy decision-making given that information and time may be particularly scarce resources, although this may ultimately be an empirical question. Still, investigating ideas used in foreign policy deliberations may allow a glimpse at the micro-foundations of broader and more durable discourses on long-term foreign policy positioning.
Four pathways of decision-making
How can these insights be used to inform a neoclassical realist theory of foreign policy choice? It has been suggested that decision-makers are tasked with distilling an appropriate foreign policy from systemic conditions through the deliberation of competing ideas. More so when systemic stimuli are ambiguous, this process is difficult and time-intensive, and the ideas decision-makers carry into deliberations will contradict each other. Deep ideational competition can then become a problem for decision-makers who face a predicament between contradicting ideas all the while systemic stimuli make it necessary to act.Footnote 112 What types of decisional dynamics could be expected empirically?
First, decision-makers may converge, through deliberation of their different ideas, on a common position. Per Irving Janis, group members may be motivated to create and maintain consensus, which increases the likelihood of convergence.Footnote 113 Also, the likelihood of convergence may depend on the homogeneity of the group.Footnote 114 Ultimately, which ideas prevail in this process may depend on their internal validity, on who holds them, on the degree of their pervasiveness, or on their plausibility in the judgement of the other group members.Footnote 115
Second, depending on decisional hierarchies and leadership characteristics (for example, whether leaders are strategists or opportunists, or interested in the problem; how leaders view their advisers), leadership may help break through ideational competition by imposing a decision.Footnote 116 It seems likely that heads of government as (frequently) experienced statespersons have both the intuition and conviction in their own ideas and competence to decide which course of action is suited to the scenario at hand. However, they then ‘own’ the decision, especially if made against the advice of trusted advisers and experts, which may be politically risky.
Third, absent agreement or decisive leadership, ideational competition may be left to linger, and decisions are delayed.Footnote 117 Given ambiguous systemic conditions as well as intense ideational competition over the right course of action, not doing anything may even be a sensible approach. However, inaction can also be politically problematic because of not only a potentially worsening scenario, but also a delay of costs and domestic backlash.Footnote 118 Especially during moments of (perceived) crisis, with suddenly raised stakes, the impulse to act may be difficult to overcome.
Therefore, fourth, decision-makers may seek to extricate themselves from ideational competition by employing faux compromise or incremental solutions.Footnote 119 In this case, ideational divisions in the executive decision-making bodies are patched over with policies that pander to multiple ideas at the same time. This fourth alternative outcome is: (1) more likely the more ideas conflict (that is, the lower the ideological homogeneity within an administration and, correspondingly, the lower the likelihood of convergence); (2) more likely the more leaders rely on advisers for decision-making; and (3) more likely the more intense external pressures to act quickly exist.Footnote 120 For this last reason, this pathway seems more likely for crisis-type decision-making than for longer-term grand strategic adjustment.
Ultimately, the relative prevalence of the four paths as well as the scope of the suggested decisional dynamic (in type, that is, short-term/longer-term policy, and spread, that is, regarding domestic institutional arrangements, coalitional governments, autocracies, etc.) must be established empirically. As it does not rely on the presence of specific ideas about interests and appropriate response strategies, but rather on their exchange and potential contradiction, the suggested framework has the potential to be applicable more widely.
Finally, can such a framework still be (neoclassical) realist? The above discussion of intervening ideas and ideational competition has relied to a considerable degree on liberal, cognitive, and constructivist scholarship. Beyond the utility a refinement of the conceptual relationship between interests, beliefs, and ideas might bring for those frameworks, this raises paradigmatic questions as to the adherence of such a framework to neoclassical realist assumptions. Per these assumptions, ideas cannot become independently causal, or even generative of interests.Footnote 121 For (neoclassical) realists, ideas may matter in the foreign policy process, but if they matter too much, states misjudge invariant elements of an objective material reality, which should result in some form of systemic punishment.Footnote 122
This is because the anarchical structure of the international system imposes inescapable constraints upon states.Footnote 123 These conditions are not strictly determinative, in that they leave space for ideas and domestic political processes to intervene in the causal path towards foreign policy choice. This does not mean that they are weak or unimportant. States that consistently fail to adhere to them will find themselves at a disadvantage and ultimately select themselves out of the system.Footnote 124 Per the above framework, ideas allow decision-makers to deliberate the nature of national interests and make decisions about the costs and benefits of alternative policy sets given a geostrategic context defined by anarchy and the relative distribution of capabilities. If ideas that contradict given state interests permanently captured the foreign policy process, or their competition became so pronounced as to produce persistent stasis, neoclassical realists would expect (at least over the longer term) foreign policy failures that can negatively affect relative power and ultimately even risk state survival.Footnote 125 In addition to demonstrating how and where externalised ideas matter causally in the foreign policy process, then, the task for neoclassical realists employing such a framework would be to demonstrate empirically that when (bad) ideas interfere substantially and continuously in decision-making, the system punishes these states.Footnote 126
Conclusion
I strive to make two contributions in this article: first, to demonstrate how neoclassical realists can employ ideational factors in the analysis of foreign policy decision-making processes while upholding paradigmatic commitments. Second, to this end, to disentangle a complicated conceptual landscape and rearrange beliefs, ideas, and language in a way that renders these concepts useful and operationalisable for positivist-minded scholars more broadly.
The successful integration of ideational factors in positivist frameworks of inquiry into foreign policy and state behaviour hinges on a careful conceptualisation and operationalisation of ideas as distinct from, yet conceptually linked with, interests, beliefs, culture, and language. To this end, I investigate the use of ideational variables in NCR. NCR served here as a hard case for such conceptual integration because, at least in its mainstream (notwithstanding more interpretivist and even post-positivist contributions), the approach remains committed to materialism and a limited rationality assumption in addition to (soft) positivism and Humean causality. As suggested, this expands on Barkin's discussion of the compatibility of realist variants with constructivist concepts,Footnote 127 and contributes to further reflection on the theoretical assumptions, differences and overlap between IR paradigms.
In NCR, the use of ideational factors as intervening variables has been challenging for two reasons: One, NCR per its paradigmatic assumptions requires materially given interests to be independently causal for foreign policy outcomes. It thus necessitates a clarification of ideation's causal role; more precisely, beliefs or ideas cannot be generative of a state's interests in realist approaches. At the same time, they need to be linked to external, systemic factors (independent variables) and foreign policy choices or outcomes (dependent variables) in the causal path. Neoclassical realists (and other rationalists) have argued that interests of the state, grounded in the relative material capabilities of an anarchic, self-help environment, can in principle be known as facts, decision-makers face limitations of time and of the extent of information they can process. Here, ideation helps decision-makers fill gaps of information and knowledge of the material world.
Two, following on from this insight, neoclassical realists (just as constructivists and institutionalists before them) have encountered difficulties regarding the conceptual distinction between individual, ‘mentalistic’ beliefs and social ideas, and their respective impact on foreign policy decisions. I suggest that beliefs and ideas are conceptually distinct, but causally connected. Individual beliefs help decision-makers appraise the international environment. However, to matter in foreign policy deliberations, these beliefs must have interpersonal relevance. To that end, decision-makers feed their individually held beliefs into foreign policy deliberations, where they argue with and try to persuade other decision-makers.
More generally, this conceptual twist addresses theoretical and methodological problems similarly encountered by other positivist-minded approaches to the study of world politics: either the conceptually flawed equation of internal, mentalistic beliefs and social ideas, or the circularity inherent in establishing the causal effect of ideation through its assumed effect on the dependent variable. It operationalises ideas directly as traceable in communicative conduits such as speeches or meeting minutes. As suggested, it highlights contradiction and argumentation in decision-making processes, and thus complements the focus on ideational coherence and continuity often espoused in neoclassical realist approaches on grand strategic adjustment. More specifically, it allows neoclassical realists to investigate the interplay of interests and ideation in foreign policy in a way that upholds (soft) positivist commitments, and usefully supplements materialist and rationality assumptions conventionally embraced in the approach.
Acknowledgments
I thank the editors of the Review of International Studies and three anonymous reviewers for their careful reading and detailed recommendations. For valuable comments on previous versions of this article, I am grateful to Martin Bayly, Andrew Delatolla, Toby Dodge, Anissa Haddadi, Liane Hartnett, Nicholas Kitchen, Steven Lobell, Aaron McKeil, James Strong, and Luca Tardelli.
