E.O. Wilson (1998) popularized the idea of consilience, that is, the unity of science. Originally proposed by William Whewell, consilience seeks to unify knowledge “by the linking of facts and fact-based theories across disciplines to create a common groundwork of explanation”. Wilson believes that the unity of knowledge will be achieved through the application of the methods used in the natural sciences to the social sciences. Consilience is envisioned to overcome the traditional Diltheyan separation of Natur- and Geisteswissenschaft. Wilson’s own theory of eu-sociality and his intellectual ventures into socio-biology are examples of this unificatory approach to human knowledge.
It would be unfair to say that the social sciences primarily provide idiosyncratic, historically oriented accounts of unique events, while the natural sciences provide nomothetic explanations of phenomena that are in principle replicable. History, the Geisteswissenschaft par excellence, tends to favour an interpretive approach to, and idiosyncratic accounts of, past events. It often focuses on the subjective beliefs and psychology of great men and it often adopts a narrative approach relying on tropes (Carr 1961; Collingwood 1946; White 1987). This in itself does not make the writing of history an illegitimate intellectual pursuit (Evans 1997). History as a discipline is of course much varied in terms of its methods and analytical foci. History can be and is written from a macro, a comparative and even a quantitative perspective (e.g. Écoles des Annales). Macro history focuses not only on large, impersonal structures, but it is also closer to the idea of consilience, as it draws on insights, fact, models and theories from other scientific disciplines. The account offered by Jared Diamond (1997) of how and why the West conquered the world is a case in point. Applying models and theories used in one field in another field of scientific enquiry can be intellectually and epistemically fruitful.
But what exactly are models and theories? The term model is often used very loosely and is seldom defined with much precision (Achinstein 1965). First, model can refer to a metaphor. The metaphor can be of a mechanical or biological nature. Eisenhower’s domino theory would be a case in point. The relationship between the entities posited in the model (dominoes) and the real-life entities it seeks to represent (states) is very loose, as is the posited mechanism (dominoes falling). Secondly, model can refer to an analogy. Analogies, compared to the vaguer model-as-metaphor, postulate correspondences in need of empirical support. States in the international system behave like firms in a market. States like firms strive to survive competition. The correspondence relationship between the entity posited the real-life entity, including the underlying processes, is tighter than in the case of a metaphor. Third, model can also refer to carefully a constructed system that is based on a small set of parsimonious and simple premises. The model may or may not represent the relevant process, but it postulates an empirically verifiable or falsifiable relationship between different variables (e.g. money supply and inflation). Observable implications can be derived from the model. And fourth, model can refer to a theory. A theory can be thought of as a model of a higher degree of generality. It also has a deductive structure. But empirical assumptions need to be added in order to arrive at a specific interpretation and to derive empirical implications. Game theory, for example, is such a general theory. The theory is general. Specifying the empirical assumptions about players’ payoff structure determine what kind of game is being played. In other words, the link between the elements contained in a metaphor or analogy and the world are far looser and far less specific and more difficult to operationalize than in the case of a model or a theory. This lack of specificity makes it easier to abuse metaphors.
A metaphor is best thought of as a half-way house between a model and a simple image. The domino theory is a metaphor of a ‘mechanical’ process, but it is also an image. By comparison, ‘iron curtain’ or ‘cold war’ is an image that fails to represent a mechanism or process and fails to postulate a relationship or mechanism. The image ‘cold war’ does suggest non-military conflict, but this is as far the image goes. Models-as-metaphors and models-analogies typically are richer and more specific in terms of their implications. This lack of specificity explains why especially images but also metaphors and analogies are often (ab)used to ‘sell’ policies. Political marketing and propaganda after all seek to appeal to simplicity, make use of repetition and exploit well-known cognitive biases as well as the affective meaning of words and of the images words evoke (e.g. ‘axis of evil’ as in WWII axis powers and metaphysical evil). In particular, the lack of specificity makes it more difficult to ‘falsify’ images, not least because they often operate at an affective rather than analytical level. Framing is an important dimension of political mass communication/ propaganda (Wehling 2016).
What do models do? What should they do? Ideally, models describe, explain or interpret, and predict. Description requires the selection of important or relevant analytic and/ or explanatory aspects of the phenomena under investigation. Explanation requires a causal account, whether in counterfactual or probabilistic terms or in terms of a mechanism or a process, while interpretation typically seeks to understand a phenomenon by identifying the subjective reasons, meanings and/ or beliefs that account for a phenomenon (aka thick description). Last but not least, models – or at least explanatory as opposed to interpretive models – should allow to forecast future events, at least in principle. Generally speaking, explanatory and interpretive models value different epistemic goals differently (e.g. in-depth understanding vs accurate predicting) and they often ask different questions. Why did statesman X decide to attack country Y under circumstances Z versus why do states decide to attack other states in general or given circumstances Z?
Consilience suggests that it can be insightful to apply ‘models’ that have proven successful in one scientific field in another. Analogies, metaphors, models and theories from one field can generate interesting insights and new ways of approaching and explaining an issue in an unrelated field. That this can be politically controversial is illustrated by E.O. Wilson’s experience after he proposed to apply insights derived from studying ants to human societies. Somebody poured water over him at a conference.
Game theory has been profitably applied in the realm of international politics as well as in a host of other fields such as evolutionary biology and economics. Game theory is concerned with the logic of interactive, rational behaviour and decision-making. The prisoner’s dilemma, for example, suggests that states will find it difficult to avoid arms races because they are worried about defection. As a matter fact, the security dilemma, the most important assumption of Realist IR theory, is conceptually based on the logic characterizing the prisoner’s dilemma. Similarly, the game of chicken expounds the logic of state behaviour in a situation like the Cuban missile crisis. More generally, game theory has a great deal of interesting things to say about conflict and cooperation in international politics (Schelling 1966; Kaplan 1991). In the economic sphere, game theory offers similar insights into the logic that prevents the provision of public goods (Olson 1965), causes environmental degradation and the over-exploitation of non-renewable resources (Hardin 1968) as well accounts strategic stability in the nuclear realm (Schelling 1966). Game theory also provides an interesting account of how cooperation may emerge in the first place, including in the international state system (Axelrod 1989).
Cognitive psychology may help explain irrationally seeming foreign policy decisions (Jaeger 2020). Research into cognitive biases has found that individuals do not always act in accordance with what the standard rationality prescribes (Kahneman 2011). This has led to the behavioural revolution in economics and finance. The sunk cost effect may explain why armed conflicts last longer than they rationally should, as the side (or sides) that has (have) incurred costs (irrationally) assign greater importance to the losses already incurred (retrospective) than (rationally) focus on the prospective losses and gains. Naturally if armed conflict is viewed as just one iteration of a sequential game and reputational gains (willingness to incur losses) strengthens the state’s future power and bargaining position, then it may make sense to incur additional prospective losses on top of retrospective losses, even in an unwinnable war.
Combining game theory and cognitive biases may help explain otherwise puzzling foreign policy decisions. Experiments have demonstrated that people do not act rationally in ultimatum and dictator games, suggesting that to a certain degree altruistic and cooperative behaviour is hardwired. It would clearly be going too far to suggest that Germany’s failure to exploit its military advantage in Dunkirk in 1940 was due to some residual altruism or a worry to be seen as non-cooperative. (Interestingly, historians continue to debate what exactly led to the decision to halt the final attack on the British Expeditionary Force.) However, cognitive psychology rather than cool calculation may on occasion help explain the failure of states to act in accordance with the standards of rationality. Marrying game theory and cognitive psychology can be a rich source of new hypotheses and potential explanations. Once again, failing to impose a harsh peace treaty in order to eliminate future military threats for good may be due to rational calculation or it may be due to cognitive biases (Vienna 1815 vs Versailles 1919).
The theory of evolution posits variation, inheritance, differential survival/ reproduction and adaptation as its fundamental features. States may not reproduce in a strict sense, but they certainly face selective and competitive pressures. Being Costa Rica (that is, a state without armed forces) may have worked for Costa Rica in the Central American context. It has not worked for Poland in the inter-European state system of the late 18th or the smaller German states in the late 19th century. Maybe the microeconomic analogy/ model would suffice to capture the competition/ survival aspects of international politics (Waltz 1979). However, neo-realism has little to say about variation and adaptation. Admittedly, adaptation in the case of states in the international system is probably better seen as Lamarckian rather than Darwinian. However, it is clear that domestic reform that enhances the international power of state can be interpreted as an adaptation. Prussia’s relative greater economic efficiency and investment in its military capabilities are an adaptation of sorts that allowed Prussia to survive and later thrive in the European state system. Prussian reforms following the defeat against Napoleon was similarly important. Absent these reforms, Prussia as an initially small and resource-poor state might otherwise have been absorbed by Austria. 18th century Poland failed to adapt and disappeared from the European map for more than a century (Jaeger 2019). Granted, theory of Darwinian evolution is more complex and does not translate one-to-one to international politics. But the interplay of domestic change and international pressure provides offer a richer and less reductionist account of international politics than Waltz’s more parsimonious approach.
Geopolitics – combining physical and sometimes human geography and international politics – fell somewhat out of favour following WWII, mainly ecause it was associated with the militarily defeated and morally bankrupt losers of WWII. Nonetheless, geography has always been indisputably relevant to international politics and neither Haushofer nor Ratzel nor Lebensraum rhetoric has changed this fact. Often seen as a classic left-wing trope, resource competition remains as relevant as ever in international politics, as it does within states (Diamond 2012). To the extent that geography shapes filters the international pressures a state faces, it can also help explain the evolution of domestic institutions and culture. Prussia may well have been “an army with a state” (Voltaire) and in part this is explained by Prussia’s precarious geographic position. The reason why neither the US nor the UK ever saw the emergence of an overbearing military bureaucracy and/ or a militaristic culture is to a significant extent due to their ability to rely on a navy rather than a mass army to ensure their territorial integrity – the military-industrial complex withstanding.
Path dependency is a model that is widely used in historical sociology. It posits that the scope of change is limited due to institutional and/ or cognitive factors (Pierson 2011; Mahoney & Rueschemeyer 2003). The underlying assumption is similar to the notion of mutation in evolutionary biology. Mutation does take place, and while it may not always be gradual, successful mutations tend to be limited in terms magnitude. Similarly, economic development is usually considered to be path-dependent. Economies don’t go, can’t go from being very poor to hyper-developed in a decade. Economies move up a technological ladder, even if some do so remarkably quickly (e.g. Korea). Their development path is constrained. Similarly, the history, culture and domestic-institutional characteristics tend to limit the path available. Historical intuitionalism and path dependency do allow for the possibility of radical change, but this is the exception rather than the rule.
The concept of the balance-of-power has long been central to the study of international politics. The meaning of the concept varies significantly, even more so than the meaning of model (Little 2007). In its simplest form, the metaphor (or analogy?) implies that an equilibrium is necessary to ensure the stability of the international state system. Stability does not imply absence of war, but instead refers to a situation where the emergence of a hegemon turns a competitive state system into a hierarchical system, often in the form of a formal (or informal) empire. In order to maintain an equilibrium, adjustments need to be made constantly. The state system will tend towards stability due to the self-interested behaviour of states. It is largely self-regulating, invoking concepts from cybernetics (Wiener 2018). The equilibrium concept is also widely used in classical economics where exogenous shocks leads to changes in supply and demand, often via the price mechanism.
Less of a model and more of a method, Bayesian statistics was used successfully during WWII to crack the German enigma code, accurately estimate German tank production (compared to intelligence estimates) and predict the location of German submarines in the Atlantic. If it is true that Stalin refused to believe that the USSR was about to be invaded in the spring of 1941, it is clear that had he had a good grasp of Bayesian probability, he ought to have changed his mind over the course of 1941 given the need to update his prior.
Systems’ theory and especially the theory of complex systems can be applied to international politics in very interesting ways. Complex systems are characterized, among other things, by (1) non-linearity, (2) adaptation and (3) feedback loops. Non-linearitiesas well as tight coupling are characteristics of complext systems. Small mistakes/ insignificantly seeming actions can rapidly cascade through the system due to a lack of redundancy and lead to catastrophic outcomes (Perrow 1984). The alliance system on the eve of WWI can be regarded as such an interconnected system with insufficient redundancy. What appeared to be a minor tussle between Austria-Hungary and Serbia (butterfly) and a limited commitment (Germany’s blank cheque or unlimited support for Austria-Hungary) led to the WWI, the death of millions of soldiers and civilians (Spanish flu), the fall of monarchies, the break-up of empires etc. The notion of unintended consequences is not unique to complex system, but it happens to be one of their key features. Although it is difficult to establish in retrospect with much certainty, it is quite possible that most major decision-makers wanted to avoid war. Admittedly, historians continue to disagree of who wanted what and when (Fischer 1967; Clark 2012).
Long story short. International politics is not just a question of the level of analysis (Singer 1960; Waltz 1959). It is not just a question of whether the international system determines tate behaviour or whether individuals determine or whether it is individuals in the face of the constraints imposed by the international system. Other models can be fruitfully brought to bear on various aspects of international politics. Intuition pumps (Dennett 2014), tools for thinking (Nisbett 2015) and many-model thinking (Page 2018) offers a variety of ‘models’ and frameworks and tools to analyse, interpret and/ or explain international politics. From a pragmatic point of view, models extraneous to International Relations offer new ways of tackling interesting questions in international politics. The extent to which this offers better explanations or provides better predictions is an empirical issue. Ideally what starts as an intuition (Dennett) is transformed into a metaphor or analogy and then into a logically cohesive model or even an even more general theory. Consilience, even if it will never be realized, is an epistemically exciting and worthwhile endeavor.