This year, I'm sharing an office with an econometrician on Mondays and with a geographer on Fridays (you don't want to go into the subtleties of the French educational system). We're discussing the content of our research and the strengths and weaknesses in our respective methodologies, and, of course, joking and complaining about the sociology of our communities.

Except in those F... research moments. Those moments when, after months waiting for your data, weeks cleaning them up, tedious hours programming, you fail to reject the null-hypothesis (I haven't yet figure out on which basis a geographer feels he's having a F... moment). Those F.... moments are inevitably followed by the look. My colleagues, no matter what their specialization is, raise their eyes from the screen, and stare at me with that mute disbelief: “you have the same position as I do, and you spend your days assembling chronologies and writing down timelines. So unfair.”

Now, this is unfair.

First, archive-focused historians of economics have their deal of F.... moments. Those moments when:

1. after months grant writing, getting half the money you need (the next year), fishing for a not-too expensive overseas flight and a cheap hotel (because, of course, the archive center is located in the middle of Beverly Hills), you open your 50^th box of archives and realize that the conference proceedings you need have been lost forever. Or misfiled.

2. or the archivist proudly informs you, on your last day, that only one box that you ordered has not reached the reading room: “correspondance : L.” Well, that's sad, because you flew these 10 000kms with the very purpose of checking the correspondance between your subject and Lucas. Maybe there isn't one after all. You will never know. Unless you write down another research project, ask for money....

3. or after hundreds of documents dealing with flight booking, edition contracts, accounts, recommendations letters, schedules, you eventually stumble on this short telegram “really enjoyed the thorough discussion we had last week at your place on the policy implications of rational expectations. An essential subject.” Indeed...

4. or you eventually unearth that uncensored debate on the relative merits of adaptative, rational and implicit expectations. Dated 1998. While the debates were raging at Carnegie in the sixties. Ex post

5. or you spot that 1933 written exchange on economists' value-neutrality between your subject and his (future ex) best friend and colleague. At least it seems. Because it's written in Swedish.

6. Or you have set to write the history of urban economics because you believed you would find out that urban economists in the sixties and seventies were all influenced by the representation of cities in DC Comics books (and that you would consequently show maps of Gotham city and chuncks of Batman movies during you next talk). Then you realize that you're simply going to try to understand which methodological and theoretical evolutions led to Mill's aggregative model of resources allocation in a metropolitain area and Meyer, Kain and Wohl's The Urban Transportation Problem.

7. Edit. And yesterday morning F-moment, when the archivist tells you that the boxes of very-important-and-busy-Nobel-Prize economist has just arrived, but that you need to get the permission of the great man in person before seeing them. Send him an email?

Second, assembling a chronology is far less easy than it seems. It's a preliminary to writing down the history of something (MIT/ Carnegie/ public economics/ the notion of expectations/ growth theory/ randomized trial experiments/ policy evaluation/ economics and biology, etc.). Yet, at this early stage of your research, the boundaries of your object are not set. Thus, you don't know what to include in your chronology. In some case, you don't even know what to look at.

Take the history of economics at MIT/Chicago/Carnegie ect. What do you include? The research of the department of economics? But why exclude other departments, schools, and the wealth of Institutes created in each university after the war? Growth theory and development economics were researched at the Center for International Studies as much as at the econ department at MIT. Chicago price theory was applied at the Business School, where Stigler was recruited in 1958, and at the Law School, the realm of Director, Coase and Posner. There were huge tensions between economists from the department (Friedman) and the Cowles (Marschak, Koopmans) on the design of macroeconomic models and their testing, and the writing history of general equilibirium requires bringing into the picture the Chicago department of mathematics as well. Also, how many characters should you include? Focusing on a priori key players is biased and restrictive (Chicago= Friedman+Stigler+Becker, MIT=Samuelson+Solow+Modigliani), it leaves out other crucial characters who haven't been canonized the same way: for instance Director, Mincer or Ted Schultz at Chicago, Kindleberger, Frank Fisher or Ed Kuh at MIT. Yet, if you consider the research of, say each generation of 20 faculty and 25 graduate students who made MIT in the forties, sixties and seventies, you may end up with an untractable story. Unless you use quantitative methods. Finally, why focus exclusively on research? Teaching, organizing curicula, advising goverments, wiriting journal chronicles, looking for funds matter.

Now, take the history of one of those fields which institutionalized in the seventies: public economics. What are the lower and upper time boundaries of such a project? And given that the use of the word “public economics” might not be stabilized even yet, which work do you include? All that is labeled “public finance”? But then, how do you treat stabilization policies, included in the field by Musgrave in his Theory of Public Finance (1959), then excluded from the practices of public economists in the seventies? And what if “public economics” emerged in the sixties as a notion with altogether different meaning and coverage in European and American communities ?

Finally, take “rational expectations.” The definition issue surface in the panel discussion held of the 50^th anniversary of Muth's paper. After Kevin Hoover had outlined the emergence of the word in economics publications, Lucas argues against including such work in the story: “Muth’s idea of rational expectations...no one had said anything like this before. It was his simultaneous determination of people’s behavior which produced a time series and a time series which led people to form expectations that affected their behavior. It’s that simultaneity that neither the statisticians nor the economists of the day had . . . and Jack had it.” Yet, during the discussions, many definitions of rational expectations surface, and the possibility that it was an idea used under a different name in game theory, and that will still be used in economics under a different name in the future is raised.

The scope of a historical narrative depends on the chronology outlined. Yet what the chronology looks like is predicated upon the boundaries of the historical object considered. A chronology, which looks as a raw history building block, is thus always emminently subjective.