This volume had its origin in a conference held at Tilburg University, July 1-3, 1991. It was supported by the Tilburg University Department of Philosophy, center and the Royal Dutch Academy of Sciences KNAW.

It was the intention to bring together economists interested in the strategical backgrounds of research, economic methodologists dealing with the practical problems of the economist’s work and philosophers of science. The conference was meant to be an exchange of thoughts on the idealizational nature of economic theories.

Without much difficulty, the area of economic scientific activity can be split up into a limited set of well known and accepted research lines, to which the vast majority of contemporary economic publications contributes. What one needs to know in order to be a successful participant of some research program is usually the literature published in the past, the way in which the authors in this line of publications formulate their problems, and the kind of solutions they would like to have for their problems.

It is clear what one should do if one is to become an expert in a specific research line, and, once you are an expert, the agenda of theoretical and technical jobs is clearly visible and well specified. Internal discussions in a research line proceed orderly way, and on a technical level. It is clear in that context how one should defend and criticize and propose solutions to problems.

This comfortable order breaks down as soon as discussion takes place between experts of different research lines. The technically refined discussions are set aside and the much rougher “economists’ debate” sets in.

Research lines seem like flowers competing for the insects’ attention. It is difficult for an economic scientist to know what to do as long as he is still in the air, his choice to land somewhere may even seem arbitrary, but as soon as he lands, he knows where to go.

What is shared by those who choose to participate in developing a specific line of research, distinguishing them from others, is (apart from a set of economic departments, regular conferences and periodicals) an idealization: an abstract view on real world economies that highlights some of the aspects observed, and neglects others. Central tasks of research are analysis (of the mathematical properties of the abstract model) and concretization (the successive dropping of the idealizing assumptions). Questions for this conference were: Why choose one idealization as a starting point rather than another? What is the list of tasks once a choice is made for some specific idealization? Can there be an orderly discussion between adherents of different research lines, using different idealizations, and not necessarily being in the same stage of concretization? These questions were to be dealt with in a specific way, preferably by analyzing well specified examples from the real world of working economists.

This volume includes revised versions of papers presented at the conference plus a couple prepared since. It addresses three main topics, the first of which is logical analysis of idealizational structures in economics. An idealization is a simplified representation of the object of inquiry. The idealization abstracts from many, if not most of the known properties of the object. It is a false image, and those who use it as an idealization know that it is a false image.

The idealization is a small world in itself. It is described by concepts and statements. You can say false and true things about it. You can neatly model it with sets and functions. You can do formal research into its mathematical properties: some theorems hold in it, others do not. If you drop some idealizing assumptions, allowing more possibilities, different types of “species” into your ideal world, you will loose some theorems, some will be modified, others retained.

Questions: What is the structure of the idealizations that one finds in economics? For what reasons do economists regard any particular idealization as a good starting point for research? How are idealizations handled in the course of developing and refining a theory? What is the list of unsatisfactory features of an idealization that are to be improved upon?

The reader will find these topics addressed in the contributions below by Cartwright, Hoover, Walliser, Janssen, Garcia de la Sienra, Maki, Diederich and Birner.

The second topic of the volume is historical case studies dealing with specific research lines. The beginning of research in new directions is usually marked by a publication (sometimes more than one) that, once research develops, becomes a “classic” in the new field that it created. Very often, its criticism of contemporary rival approaches, its principles, and the research program implicitly advocated by its basic ideas seem obvious, if not trivial to readers in later stages of the development of the research line. It becomes harder and harder to understand what problems the earliest readers had in acquiring the new line of thought. But at the time an article that is to become a “classic” was written, its points were not trivial at all. If the paper succeeds in its task of redirecting research, its points remain clear, but their original problematic character to those who have a different frame of mind, becomes more and more obscure. As one of the results, the history of the field before the “classic” publication now starts to be looked upon as a “prehistory” in which authors (now starting to count as “precursors”) held bits and pieces of its idealizational ideas in rudimentary form. Idealization has done its work, including in such cases the reinterpretation of the history of the field.

Questions: What was the criticism of publications that later became “classic” of older idealizing representations of the relevant economic process? What did the old idealization look like through the eyes of those preferring the new? What did the new idealization look like through the eyes of the old? What was the counter-criticism that the new approach had to overcome? How did it succeed, and in so doing, make older idealizations fade?

These processes of transition from one idealization to another can be observed in the large (the “revolutionary” periods of economics), and in the small (changes of approach within specific fields). Aspects of these problems can be found in the contributions here, by cools, hamminga and kuipers, hausman, keuzenkamp and jorland.

The third topic is logical analysis of the relationships between idealizations prevalent in different research lines. We often find adherents of rival theories opposed to each other, both claiming that the other’s theory is simply a special case. Logical analysis of the relationship between the rival idealizations is something that an economist debating over the borders of his own research line does all the time. But surely it is the most difficult theoretical task an economist can assume, and a satisfactory result may not even exist in some cases. “Abstraction” means “forgetting about some things” for the time being, and analyzing the relationships between two idealizations means comparing two different sets of things forgotten for the time being, and the effect of this on theory.

It would be nice if we could assume that adherents of rival idealizations at least agreed on the complete list of relevant properties of the object of inquiry, and thus only disagreed about which of these could safely be “assumed away,” as a first, second, or third approximation. But the real world of working economists seems to be more complicated.

Questions: How to model rival idealizations in such a way that rival economists would at least agree how they should be compared and what is the result of such a comparison? And, if this remote goal cannot be reached, then what is the list of obstacles? These problems are touched on in many of the contributions here, but especially exercise balzer, cook and hendry, and nowakowa and nowak.

This gives a rough framework for approaching the individual papers, though it is hardly a proper introduction. Rather than forcing what is in fact a very wide-ranging set of papers into a mold into which they do not readily fit, we have preferred to devote the space that would otherwise have formed an introduction to sketching the models of idealization espoused by a group of well known economic methodologists: this will serve as historical background material to the papers.

Bert Hamminga and Neil B. De Marchi