The topic of heuristics is attracting increased attention in U.S. sociology, as evidenced by a growing number of works on this topic (e.g. Becker, 1998; Abbott, 2004; Martin, 2014; Swedberg, 2014). The term “heuristics” itself is situated somewhere between the following two poles in social science discourse: rules of thumb on the one hand and discovery on the other. Its meaning as discovery was the original, dating to Antiquity. The notion of heuristics as rules of thumb, which people use in situations of uncertainty, was introduced some decades ago by Amos Tversky and Daniel Kahneman (Kahneman, Slovic & Tversky, 1982; for a related approach, see Simon, 1989).

While our own preference is for heuristics understood as discovery, what we will argue in this note is that there also exists a third meaning, which is not part of the semantic history of heuristics but which is important for understanding the current role of heuristics in social science. According to this meaning, heuristics do not only consist of “tricks” and “moves” and “good advice” that sociologists follow, in the hope of making a discovery. Heuristics in sociology is also part of something larger, namely a new and developing area of knowledge. And this new area of knowledge throws in its turn a new light on heuristics: not only on what it is but on what it can become.

Today's attitude towards heuristics among sociologists, we suggest, is part of a larger and more important development in social science which is still taking shape. During the last ten or so years a new field of knowledge has slowly begun to open up; and the knowledge of tricks, moves and advice is part of this field. This new field of knowledge is theorizing. Theorizing is about sociologists becoming aware of what they are actually doing when they work with theory, and also being aware of how they can use this knowledge to shape their work.

Social scientists have concerned themselves with what they are doing and with the forces that shape their behavior for a long time. This concern has crystallized into different forms of knowledge, one of which is the sociology of knowledge or, more precisely, the sociology of sociology. Reflexivity and the sociology of ideas are more recent, related approaches. All of these express first and foremost an awareness that the works of sociologists is not the result of objective consciousness, but instead deeply influenced by their authors' upbringing and related social forces. You grow up in a specific family, in a specific class; you work in a specific profession, and so on. All of these factors will influence how a sociologist handles research, including theory.

This type of knowledge, however, is general in nature and blunt; it lacks the specificity needed when dealing with complex phenomena such as thinking and language. Knowing that someone is a member of a certain family, a certain class and so on does not tell very much about what will happens in the head of that person (even if has its obvious uses). The errors will also typically be of the oversocialized type.

This type of knowledge fails in another way as well: it cannot be translated into a positive program for how to do research, including how to do theory. In this respect reflexivity and related perspectives are mute and of less value.

But we know that there also exist books and articles that tell social scientists how to do research with some precision. There are plenty of these on topics such as sociological methods and theory construction; and they view as their primary task to spell out how you should go about things when you do social science, including theory. In doing so, they are at the same time both helpful and insufficient.

They are helpful in that they provide answers to question such as the following: how is a model built, a survey conducted, a theory constructed? Their drawbacks, especially when it comes to heuristics in the sense of discovery, are equally obvious. They mainly focus on the context of justification and have little if anything to say about the context of discovery.

To this should be added that they do not describe how research is done in a realistic way, with all of its ups and downs, dead ends and inspirational parts, and so on. They present “the story-book version of scientific inquiry,” to cite Merton in Social Theory and Social Action (Merton, 1968, p. 16). Merton also approvingly cites a physicist who once said that scientists are “professionally trained to conceal from themselves their deepest thought” and to “exaggerate unconsciously the rational aspect” of work done in the past (ibidem, pp. 6–7).

It is precisely here that theorizing comes into the picture. Theorizing represents an attempt to portray how things are actually done, and how theory is actually used in research. The search light is directed straight at what the social scientist does for two reasons. First, by proceeding in this way, social scientists will become aware of what they are currently doing; and second, they will also learn what they should be doing.

At this point it is possible to note some interesting parallels between theorizing and moral action; and the reader may suddenly realize that the topic of theorizing is not so different from what constitutes the central topic in Foucault's History of Sexuality (Betta, 2016a; 2016b). This is the creation of the self as an object of knowledge. In the case that Foucault is concerned with there is what he calls “ethical work” (“travail éthique”); and in the case that is discussed in this note, there is a concern with what Merton calls “theory work” (Foucault, 1985, p. 27; Levine, 2006, p. 239).

According to Foucault, ethical work originally took the form of “care of the self” and this happened in ancient Rome. In the important introduction to volume two of The History of Sexuality, where this new Roman form of ethics is discussed, the reader is told the following about the creation of the ethical self as an object of knowledge:

In short, for an action to be “moral”, it must not be reducible to an act or a series of acts conforming to a rule, law, or value. Of course, all moral action involves a relationship with the reality in which it is carried out, and a relationship with the self. The latter is not simply “self-awareness” but self-formation as an “ethical subject”, a process in which the individual delimits that part of himself which will form the object of his moral practice, defines his position relative to the precept he will follow, and decides on a certain mode of being that will serve as his moral goal. And this requires him to act upon himself, to monitor, test, improve, and transform himself. (Foucault, 1985, p. 28; emphasis added)

Foucault then goes on to show how knowledge is created when the actor views himself/herself as an object to work on. He also describes how this type of knowledge becomes a new field or discourse with its own structures and rules.

What links the projects of ethics and theorizing to one another is that they both lead to the creation of a new discourse of knowledge, which not only describes what someone is doing but also what should be done to reach a goal. The actors, to repeat, create this new field through their thinking and the acts entailed. Personal and impersonal rules are created and anchored, first in practices and then in institutions.

To return to heuristics. The bits and pieces of knowledge about heuristics that so far have been collected and held up as useful to the individual sociologist are part of the formation of a much broader and more general area of knowledge. And it is also through this new area of knowledge that the emerging notion of heuristics in social science is best understood.

Again, the discourse is constituted through the theoretical practices of the social scientist. Heuristics joins in this way a set of issues and activities that all are part of theory work, such as how to use analogies, metaphors, induction, deduction, explanation, generalization, and so on. Heuristics interacts with these other techniques or tools of thinking; and it also comes into being with their help, in the concrete acts of research.

The area of theorizing, which is created in this way, seems to be increasingly recognized in today's social science (e.g. Swedberg, 2017). It is developing alongside traditional theory and the history of social science, both of which have atrophied over the last few decades, overshadowed by the quick and brutalist progress of methods in the social sciences.

Heuristics, in other words, is one of the many mental techniques that help to shape the theorizing self, and also allows this self to express itself in thought and practice in the form of research. This is a promising development that will hopefully continue.

But turning into a new field of knowledge, about to be established in social science, theorizing will also have to go through the process of being scrutinized, classified and probably neutralized by the powers that control the discourse in social science. These powers are currently poorly understood even if they are strongly felt.

If this neutralization of theorizing happens or not depends also on how deep the transformation of the self of the social scientist will be. Tricks and moves and good advice touch only the surface, but when heuristics turns into serious work on the theorizing self it reaches considerably deeper. How deep, and what will be found at these depths, we do not know. Research in cognitive science may have to be brought in at this point, and probably also advances from some other sciences.

But in discussing the possible neutralization of theorizing, the reader should also be reminded of the intellectual tradition that Foucault comes from as well as theorizing itself. Another reason for returning to Foucault at this point is the need to say something more about the parallels between ethical work and theorizing.

Foucault was not only influenced by some of the best minds in the great tradition of French social thought, such as Gaston Bachelard and Georges Canguilhem. He was also, from the very beginning of his career, deeply immersed in the work of Immanuel Kant, as exemplified by his early translation and introduction to Kant's Anthropology from a Pragmatic Point of View (Foucault, 2008). It was, however, another work by Kant that fascinated Foucault the most, and which he kept coming back to time and again in his writings and lectures. This is the short article by Kant that can be seen as the first and also the founding document in the history of theorizing: “What is Enlightenment?” (1784).

Here we find, for the first time in modern philosophy, a thoroughly democratic concern with thinking — how to go from obedience and conventional thought to learning how to look at the way in which you think, in order to be able to think for yourself (Selbstdenken, Arendt, 1992, p. 32). Kant outlines many of the obstacles to be overcome in a concrete manner. You must not, most importantly of all, hand over your right to think to somebody else.

The Kantian project of the thinking self, which can be described as persons who act on themselves by teaching themselves how to think, has been carried on and further developed by many philosophers, including Charles Sanders Peirce (1992a; 1992b) and Hannah Arendt (1978; 1992). This is the broader tradition to which theorizing, as well as heuristics, belong, whether they take place inside or outside the university. Its motto was given by Kant and is still valid today: “Have the courage to use your own understanding!”