By Aiden Singh
It is the main organ of the human nervous system. It regulates our sleep patterns, produces our vision, and ensures the proper functioning of our vital organs. And, most importantly to social scientists, it is responsible for our decision-making and social interactions. Yet the human brain – the organ that is ultimately responsible for the development of organised government, the market mechanism, and our legal systems – is not a topic of interest for most political scientists, economists, and legal scholars. And it features nowhere in standard textbooks on government, economics, finance, or the law.
The social science discipline that makes the boldest claim to “understanding” human decision-making without incorporating neuroscientific and psychological evidence is undoubtedly economics. And its tools and methodologies have been adopted by neighbouring disciplines such as legal studies and political science. To see how economists justify their claim of “understanding” human decision-making without ever making reference to the brain or mind, consider my experience in a game theory course I took during my first term at the London School of Economics (LSE).
Throughout the duration of the course, I wondered if the professor actually believed that decision-makers perform the elaborate computations the course conjectured they do. Towards the end of the term, I got my answer. The class was presented with the example of football players attempting penalty kicks. The problem at hand was understanding when the kicker would kick the ball towards the left of the goal and when he would kick it towards the right of the goal. And similarly, we wanted to understand when the goalie would dive left and when he would dive right. We were presented with a study of real-world penalty kicks that suggested that both professional kickers and professional goalies made the decision on which way to go “as-if” they were playing the game in line with the optimal strategies suggested by game-theoretic models. That is, the players may not be computing their optimal strategies in the way that the models suggest. But their years of experience playing football and honing their skills have resulted in them being able to play “as-if” they are. Thus, the professor’s conclusion was that game-theoretic modelling is justified by the possibility that it may be capable of making accurate predictions about the real world.
This “as-if” approach to the study of society was famously articulated by Milton Friedman in 1953. Friedman asks us to consider the problem of trying to predict the shots taken by an expert billiards player. He points out that we can make accurate predictions about these shots by assuming that the player knows the complicated mathematical formulas that give the optimal directions of travel, can accurately estimate the angles describing the location of the balls merely by eyeing the table, can make lightning speed calculations using the mathematical formulas, and can make the balls travel in the direction suggested by the formulas. Of course, it is not actually possible for players to do all this. But, we can model the shots made by expert billiards players “as-if” they can. For if their play does not correspond closely with the mathematically optimal strategy, then they cannot truly be professional billiards players.
However, while the “as-if” approach may be capable of making accurate predictions about the shots taken by expert billiards players, it does not tell us anything about how these actors are actually making decisions. That is, it does not inform us of the underlying neurological and psychological processes that give rise to such decisions: it does not help us understand. Moreover, “as-if” models lack applicability to real-world problems that are more complex than billiards or penalty kicking. For example, an “as-if” model of financial markets which assumes that human beings have perfect capacity to process unlimited information will yield predictions far different from observed real-world outcomes in which cognitive limitations mean that the financial markets are subject to periodic booms and busts. Furthermore, as a result of its inability to help us understand how actors are actually making decisions and its inapplicability to complex real-world problems, the “as-if” approach does not help us formulate public policy and improve society. For example, in the case of financial markets, the assumption that investors are unboundedly rational would lead us to conclude that booms and busts cannot occur. And this result would suggest a public policy of minimal financial regulation. But of course, history has taught us that this is a mistake.
Thus, Friedman’s “as-if” methodology, which has been widely adopted by social scientists, does not assist us in understanding how society works. And by extension, it does not help us formulate public policy. To have a chance of improving society, social scientists must understand how societies function. To understand how societies are established and ultimately sustained or destroyed, social scientists must make reference to how individuals make decisions. And to understand how individuals make decisions, social scientists must make reference to the brain and the mind.
But aside from the problem of many social scientists using “as-if” methodologies, those who avoid this methodology often use overly simplistic approaches to addressing the problem at hand. For example, consider international relations (IR) theory, the branch of political science that purportedly seeks to understand observed interactions between states. It emerged at the conclusion of World War II in an attempt to prevent the recurrence of such a calamity.
However, this discipline suffers from confusion over what constitutes a theory. A theory is an attempt to explain some aspect of the observed world. Thus, given that the aim of international relations theorists is to explain interactions between states, their theories should attempt to explain why states interact with one another in the way that they do. But the various theories of this discipline are not intended to be viewed as such. Rather, IR theorists claim that these different theories serve as “frames” through which to view the world. Thus, international relations students are taught that, to understand the complex interactions between nations, begin by picking your choice from among a set of simplistic “frames”. Then, use this frame to “interpret” the behaviour of a particular country or set of countries in a particular instance. For example, one might use the realist framework and its emphasis on relations of power to “interpret” the imperial behaviour of empires. But one would not be encouraged to use this framework to understand why 51 states came together to form the United Nations, an institution intended to foster international co-operation. For this purpose, one would use an alternative frame which emphasises the role of international institutions, such as liberalism.
This seems an odd way of attempting to understand the world. It is analogous to telling neuroscience students to pick some narrow framework for viewing the brain and use that framework to try to “interpret” how the brain works. But of course, this is not what neuroscience students are taught. Instead, neuroscience majors take a collection of biology, chemistry, and psychology courses, the sum of which are intended to give the student an understanding of the human nervous system from the molecular to the cognitive level. Similarly, computer science majors are required to take a collection of courses in programming, algorithms, computer architecture, and so on, the sum of which results in an understanding of how a digital computer works.
An idea may be an attempt to explain some real-world phenomenon (i.e. a theory) or a basic conceptual structure that may or may not be useful for “interpreting” real-world phenomena (i.e. a framework). But a theory is not a framework and the frameworks of international relations are not theories. Nor should the construction of frameworks be the aim of the social sciences.
The different “theories” of international relations each focus on some of the factors that drive state behaviour while neglecting others. And later “theories” are developed to bring to the attention of IR theorists some factor that has been left unaccounted for by previous “theories”. Continuing this research paradigm will at some point result in the creation of a “theory” for every factor driving state behaviour so that the sum of these “theories” is simply a list of the different factors driving state behaviour and, in the case of some of these theories (e.g. realism, liberalism), a description of what international relations may look like if this was the only factor driving state behaviour. In a research paradigm in which international relations theorists actually sought to build theories about the real world, all of these factors would form the starting point for the creation of a theory rather than a subset of these factors serving as the starting point for the creation of a “theory” that is also a “framework”.
But were IR theorists to attempt this, they would quickly come to realise that the factors that shape state behaviour are too numerous; that a theory that explains the behaviour of every state in every circumstance does not exist. They would find that emphasis should instead be placed on detailed investigation of the particular circumstances and historical path-dependencies of each state. In short, they would move away from the construction of “theories” to case-by-case analyses that take into account geographic, economic, historical, cultural, ideological, psychological, and other considerations. And they would come to the realisation that the relations between states cannot be understood as international relations theorists, but only as social scientists.
Thus, the curriculum of those students who wish to study the complex interactions between states must be reformed. Like the neuroscience and computer science students who take courses in a collection of topics which help them understand the totality of their subject matter, the curriculum of international relations students should consist of a collection of courses in history, economics, geography, and so on that allow them to understand the behaviour of the specific countries in which they are interested.
Moreover, the problems in IR theory pervade the social sciences. Many of the questions social science considers cannot be answered as economists, political scientists, or legal scholars, but only as social scientists. How do Homo sapiens make decisions about consumption vs. saving and who to vote for? Why do financial markets behave as they do? And which constitutional arrangement of government achieves the best outcomes? All of these questions can only be answered by transcending the disciplinary boundaries imposed by current university social science curricula.
But worse than my intellectual disagreements with many social science academics about approaches and methodologies is my realisation that some social scientists simply do not care whether their research deals with the real world or not. In a recent course on decision-making, I brought to the attention of the class that experimental studies had suggested that much of what we were learning in the course was not an accurate representation of the real world. To my surprise, the professor agreed: he did not contest the validity of the experimental studies. I then asked whether the nonconformity of the real world to our academic world was an indictment of real-world actors whose behaviour does not conform to our academic models or an indictment of the academics whose models do not conform to the real world. He responded, to my concurrence, that it was an indictment of the academics who constructed models that bore no resemblance to reality. He did not deny that the approach taken by researchers in his discipline may be overly simplistic. Nor did he try to justify the approach via the “as-if” argument. But just as swiftly as I had brought up my concerns about the usefulness of what we were being taught, and despite the professor’s admission that what we were being taught did not seem to bear any resemblance to reality, another student interjected with a question about the mechanics of the problem we had been discussing. Upon answering the student’s question, the instructor then carried on lecturing about the very same models whose usefulness he admitted left much to be desired. The fact that both I and the instructor agreed that what was being taught may not be useful nor accurate had no impact on the content of the course. I was expected to learn the material simply because it was being taught; because I was to be examined on it. Several weeks later into the course, the professor made recommendations to the class about areas in his field that were relatively unexplored and “required” more research, after having stated several weeks earlier that the entire field bore no resemblance to reality. Thus, the social sciences appear to suffer not only from methodological deficiencies, but, in some cases, from having forgotten the aim of social science research altogether.
Rerum Cognoscere Causas. To Know the Causes of Things. That is the motto of the LSE, one that appropriately sums up why researchers should conduct research, why teachers should teach, and why students should study. But I have come to find that most social scientists fall short of this goal. The divorce from the real world. The straitjacket of using narrowly defined academic disciplines to study the complex interdisciplinary problems of the real world. And the indifference of some researchers towards whether their research actually explains real-world phenomena. These are the reasons why I have come to the realisation that I will not discover how the world works through a university social science education. And that is why I am leaving the LSE.