From instrumental to cognitive rationality: the role of reasons

Sead Zimeri

Do normative reasons play any explanatory role in social sciences? Instrumental rationality particularly in the form of Rational Choice Theory (RCT) has been the dominant form of rationality in social sciences, particularly in Economy, but has it been able to articulate and explain normative claims, beliefs and actions that are guided by considerations other than those guided by self-interest? There has been a normative turn in social sciences that has challenged some of the core principles (postulates) of the RCT and has provided a broader conception of rationality that is inclusive of normative claims. Considering that my background is practical philosophy which deals with normative issues I find it interesting and want to explore some of the overlapping points between social sciences and practical philosophy. In certain respect, they overlap and in others there seems to be no overlap whatsoever. A dialogue is possible but on what premises and on whose terrain? Normative or cognitive rationality provides the terrain as that is the element that is common to both disciplines, each approaching it from their own research positions.

I therefore find the normative turn that social theory took after being subjected to critiques from a wide range of scholarly standpoints particularly instructive and useful to my own research and to the possibility of finding congruent or overlapping points between social sciences and philosophy, which is neither science nor art.

Reason and rationality are not only the point of overlap between social sciences and philosophy but also probably the point of division of social sciences and the natural sciences. That is an interesting topic to explore and to see whether social sciences are closer to humanities and hermeneutics or to the hard sciences. I will simply assume a Weberian approach of sorts that social sciences are interpretive sciences that explain meaningful human action. Having said that I should begin by giving a brief overview as to what is interesting about social sciences normative turn.

According to Kalleberg (2010) there is a widespread scientistic prejudice among social scientists that normative issues should be kept outside scholarly discourse because they cannot be settled rationally with reasons. The reason why many social scientists maintain this is because of the empiricist assumption that, as a science, sociology is an empirical and not a normative endeavour (Kalleberg, 2009). This is the “core intuition” of the empiricist view of science, that science is value-free, and that it seeks to objectively explain the natural and the social world (Benton & Craib, 2011). The distinction between fact and value is held as the core methodological characteristic of what science is and seeks to do, and if social sciences in any way want to emulate the success of the natural sciences they too must uphold the distinction. But as Merton (1932) has shown science is not free of values, that science as the pursuit of objective knowledge about the world itself implies value commitments, and he specifies four such values as indispensable to scientific endeavour itself, as the ethos of science, namely, universalism, communism, disinterestedness and organized scepticism.

This however does not touch the problem of what is at issue in the normative turn of social sciences in the second half of the 20th century. A scientist as a citizen can of course approve or disapprove of normative practices in her field, but she may do that only as a citizen, not as a scientist. Kalleberg, drawing on a host of other social scientists and philosophers (Boudon, Dahl, Habermas, Rawls, Sen, Dworkin) argues that non-cognitivism in sociology is untenable and seeks to develop and defend a cognitivist platform for normative claims.

What is rationality? It “essentially refers to the capability of the actors to criticize or defend claims with reason. Both descriptive and prescriptive claims can be accepted or rejected on the basis of the intersubjective influence of the better reasons” (Kalleberg, 253).

In his article on “Beyond Rational Choice Theory” (2003) Boudon argues that due to a narrow definition of rationality RCT fails signally in explaining positive non-trivial beliefs as well as normative nonconsequential beliefs. Boudon has no problem admitting that RCT is the adequate framework of many successful explanations, as an example of which Boudon mentions the collapse of the Soviet Union in the early 1990. Applying the RCT can help explain why the Soviet Empire collapsed when it did, and why, so abruptly. In a nutshell, after the Second World War the Western World and the Soviet Empire became involved in arms race, which presented a “prisoner’s dilemma” structure. After decades-long the PD structure was destroyed. “It was destroyed by the threat developed by the U.S. President Reagan of reaching a new threshold in the arms race by developing the SDI project, the so-called Star Wars” (p.4). Economically the project was so expensive that the Soviet could not follow the project thus losing its status as a superpower. Here RCT helps explain why the Soviet Empire collapsed, and the explanation works because it “provides a convincing explanation” (p. 5).

According to Boudon RCT has six postulates:

P1: any social phenomenon is the effect of individual decision (individualism)
P2: an action can be understood, in principle at least (understanding)
P3: any action is caused by reasons in the mind of the individuals (rationality)
P4: these reasons derive from considerations by the actors of the consequences of their actions (instrumentalism)
P5: actors are concerned mainly with the consequences to themselves (egoism)
P6: actors can distinguish cost-benefits of alternative actions and choose the one with the most favourable balance (maximization/optimization) (p. 3-4)

RCT model is a powerful theory which “produced a substantial number of genuinely scientific contributions”, but nonetheless it is powerless when confronted with many other phenomena, such as, for example, the “voting paradox”. According to the RCT the effect of a single vote is so small that rational actors should refrain from voting. Because it assumes that individual action is instrumental it seems to be the case that RCT cannot explain many beliefs and actions that are guided by consideration that have nothing to do with their own interest, or with the consequences of their actions or reactions. It operates with an instrumental notion of rationality which cannot explain normative behaviours which are guided by reasons that sometimes even go against one’s own self-interests.

From a philosophical point of view there are several criticisms that can be made of the RCT model, none of which would be dismissive of its broad range of possible applications or declare the theory null and void. A philosophical criticism recognises that RCT is indispensable to understanding many of the actions that humans as social actors undertake. There is no better theory to explain, for instance, traffic lights and why people or drivers stop at the red light and drive through at the green light. It is because they think in terms of the RCT postulates. Inconvenient as it may be, we are ready to accept those inconveniences because of the benefit it accrues. The benefits/advantages are far greater than disadvantages to the person observing the rules of the traffic lights.

But is this the only plausible explanation that can be mustered to explain human social action? RCT helps to explain a lot but not everything. The reason why it does not explain everything can be understood through some examples, but before we go there it is important for the clarification of this methodology to show where its weaknesses lie. RCT operates with an instrumentalist notion of rationality, which means that it explains all human behaviour in terms of cost-benefits, in terms of maximisation of benefits to themselves. The notion of rationality it operates with is instrumental in the sense that it subordinates all human action to factors that are in the last instance means to satisfying by maximizing certain egoistic aim. If an action does not satisfy these postulates it then appears as if the actor is acting irrationally, that is against one own self-interest. The RCT thus operates with a narrow definition of rationality and as a result it not only is incapable of explaining rationally a manifold of human actions that do not fall under its postulates but it declares those actions to be irrational. So, while one can generally accept all of the six postulates that define the RCT, none of them can be accepted without qualifications. Since there is no space to subject each one of these postulates to a philosophical critique, I will focus only on the postulate of rationality, while accepting others as they stand. In other words, generally the RCT is sound, it only gets into difficulties when it claims that all human action can be explained in terms of its narrow definition of rationality.

RCT in other words is either too wide or too narrow. Too wide if by that we mean that it misses the mark of complex human interactions that are not guided by instrumental reason, that is by maximising or satisfying egoistic ends. Human beings do operate with other reasons, reasons that sometimes explicitly go against what from an RCT perspective appears as irrational. One follows moral commands fully aware that by following them one sacrifices one’s own self-interests in the pursuit of wider social interest, peace for example, although an anarchic situation may provide ideal circumstances for making great profits; or follows a practical norm that is morally binding despite the fact of the norm’s demandingness in that self-interest be put aside for the greater common good.

In this case, it is the greater good that functions as a normatively binding reason for the subject or the social actor to act in a non-egotistical way, to give priority not to his egoistic desires or reasons for maximising the personal benefits but for maximizing the benefit of the whole society, working to realize a certain shared ideal of a community, or simply to bring about an order where previously disorder, from the perspective of engaged agent, reigned. This goes to show that there are not only different language games or, more precisely, different language games, different perspectives, are different realities. What from one perspective appears as a problem from another perspective may appear as a solution to another problem. Each perspective is thus a different way of trying to make the world intelligible: “If for example, I become converted to Christianity, I do not see the world as I saw it before: whereas before I might have seen people who were miserable, I now see people who refuse to allow Jesus to touch them. Whereas I would have avoided such people I know believe it is my duty to bring the Word to them.” (Benton, T. & Craib, I. 2011, p.95)

What the example of conversion so clearly shows is that in changing religion, people also change their rationale or their reasons for actions, that what was previously rational to do no longer appears so. It may have been that they had good reason to avoid those people, but essentially for very different reasons: it is one action seen in at least two different perspectives that are not reducible to a purely instrumental rationality operative in RCT. To a non-Christian certain class of people might have appeared miserable not because they did not follow Christianity but because they were doing certain actions that prevented them from achieving happiness, they were, let us assume, acting irrationally according to the understanding of irrationality that emerges from RCT. But once he converts, the previous reason ceases to be a reason, it no longer carries the power to convince the actor to act or abstain from acting based on its directives. But the actor has not ceased to be rational: what we have now is simply the change of perspectives and these perspectives come along with their own internal reasons, or a rationality peculiar to their own world-view.

Since human beings as social actors engage in all kinds of social activities it is not rational to dismiss their actions as irrational when they are capable of providing cognitive or moral reasons for them. Human actors usually are capable of providing reasons for their actions even if those reasons may appear to us odd or bizarre. Nonetheless from the perspective of their worldviews or theories that underpin their reasons, those reasons carry normative weight that cannot be dismissed without distorting the explanation that is given to their actions. (See Habermas, 1984, p.15).

Boudon succinctly summarises the weaknesses of RCT in the following way: RCT has nothing to tell us about the phenomena on which actors place their noncommonplace beliefs. Since all behaviour involves beliefs and to account for those behaviours it is crucial that we account for beliefs on which the behaviour rest, but RCT is not capable of doing just that because it operates with an instrumental notion of rationality. What is needed is a cognitive notion of rationality. The second type of phenomena which RCT cannot account for and therefore it is a weakness which fails the theory are those characterized by the fact that “actors are following non-consequentialist prescriptive beliefs.” RCT can account for consequentialist beliefs and has no trouble explaining them, i.e., the example of the traffic lights. But RCT is mute when it comes to normative beliefs that cannot readily be explained in consequentialist terms (Boudon, 9). It is also powerless in the face of the phenomena involving the behaviour by individuals whom we cannot in any sensible way assume to be dictated by self-interest. Although these phenomena defy RCT and they weaken its claim to be the only viable theory explaining actors’ social behaviour. We get interested and worked up for issues that do not affect us directly, yet we are ready to mobilize ourselves to bring about the desired change.

What all these have in common is that they show with remarkable clarity the weaknesses of the RCT without therefore undermining its explanatory power in other aspects. According to Boudon, what we need is a broader theory of rationality that is capable of accounting for these phenomena within sociology and social sciences generally. It is exactly this theory that Boudon and others aim to provide in order to supplement RCT.

This brings us closer to Habermas’ theory of communicative rationality that he developed in the Theory of Communicative Action, volume one (1984). According to Habermas rationality is wider than instrumental rationality. Incidentally this is not an original point developed by Habermas. Weber held similar position. Weber argued that the true object of sociology, and of social sciences generally, is the meaningful, rational, social action, as opposed to behaviour. There is purpose to human social action, as opposed to physical actions that are done without purpose. Weber distinguishes different types of social meaningful actions. The first is traditional action, an action which is done and justified because it was always done so before. Tradition provides a satisfactory explanation as to why a certain action is being carried out. The second type of meaningful action is the affective action, i.e., an action based on emotion. This type of action can be more or less rational or irrational depending on how we manage the emotional response to outside stimuli. The third type of action is the value-action. We act on the bases of values we have chosen but these values themselves cannot be rationally justified. Values can, however, be rational within the system of values that we uphold. Each system of values comes with a number of norms that regulate the behaviour of its followers, and it is reasonable to assume that one can and does act rationally or irrationally within that system of values. The fourth and the final type is the practical action oriented toward achieving something in this world. These are all meaningful social actions but there are also two types of understanding, what Weber calls “verstehen”: observational and explanatory. Observational understating is descriptive understanding, and explanatory understanding involves understanding the reason behind a certain action. What all types of meaningful social actions have in common is that they are all based on some sort of reason for action which explains the behaviour of the social actor in different settings. Weber considerably enlarges the notion of rationality to include all purposeful or meaningful action. Weber recognizes other types of action that could be undertaken in pursuit of values that do not bring measurable benefit to the actor. (Benton 2011, 87).

A more coherent notion of rationality is provided by Jürgen Habermas who has argued “for a wider concept of rationality connected with ancient conceptions of logos.” He understands rationality to be “a disposition of speaking and acting subjects that is expressed in modes of behaviour for which there are good reasons or grounds.” (Habermas, 22). So long as claims are open to criticism and can be defended against criticism, that is so long as the claims raise validity claims they can be considered rational regardless of the fact that they are not narrowly defined as rational by the set criteria of RCT. Habermas has developed in detail the rationality of communicative reason which I cannot reproduce here, but in this context it was important to mention how rationality is wider that it is assumed by RCT and that a conception of rationality based on reason accords finally more readily with practical philosophy which studies normative reasons that we have for doing certain actions that may or may not accord with the conception of the rationality inherent to RCT but which is in accord or overlaps with the cognitive and communicative rationality that is being developed in social sciences by the likes of Weber, Habermas and Boudon. Philosophy overlaps and can potentially contribute or help in clarifying the role normative reasons play in motivating human action. Cognitive rationality seems particularly promising in light of the fact that it is concerned with reasons that actors see as valid for their behaviour – I did so because I believed that X is true, likely, plausible, etc. Cognitivist model is drawn from the rational choice theory by lifting the restriction that the reason of social actors should always be the cost-benefit type. What the enlarged concept of rationality shows is that humans are more complex in their reasoning processes than it is assumed by the RCT, and that the range of reasons that motivate human behaviour and also explain it or cause it need not be, nor it is irrational to assume that if they do not follow the RCT model they somehow fall to qualify as rational explanations.


Benton, T. & Craib, I. (2011). Philosophy of Social Science: The Philosophical Foundation of Social Thought. Palgrave Macmillan.
Boudon, Raymond (1996). “The Cognitivst Model: A Generalized Rational-Choice Model” in Rationality and Society 8 (2): 123-150.
Boudon, Raymond (2003). “Beyond Rational Choice Theory” in Annual Review of Sociology, Vol 29, pp. 1-21.
Habermas, Jürgen (1984). The Theory of Communicative Action: Reason and the Rationalization of Society. Vol 1. Translated by Thomas McCarthy. Polity Press.
Kalleberg, Ragnvald (2009): “Can Normative Disputes be Settled Rationally? On Sociology as a Normative Discipline” in Mohamed Cherkaoui and Peter Hamilton: Raymond Boudon: A Life in Sociology. Essays in honour of Raymond Boudon.The Bardwell Press, pp. 251-269.
Kalleberg, Ragnvald (2010): “Ethos of Science and the Ethos of Democracy” in Craig Calhoun (ed.): Robert K. Merton: Sociology of Science and Sociology as Science. Columbia University Press, pp. 182-213.
Merton, Robert K. (1942/1973): “The Normative Structure of Science” in Merton, R.K. and Norman W. Storer: The Sociology of Science, Theoretical and Empirical Investigations. Chicago: The Chicago University Press.
Weber, Max (2009). “Science as a Vocation” in H. H. Gerth & C. Wright Mills (eds): From Max Weber: Essays in Sociology. Routledge.


About albphilosopher

Sead Zimeri has studied Philosophy, Islamic Philosophy and Religion, International Politics and Psychoanalysis. He is currently the project coordinator of "Islam and the Liberal Society" at the Liberalt Laboratorium (LibLab) thin tank in Oslo, Norway.
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s