Chapter 8

IMPLICATIONS FOR THE UNITED STATES

William Bentley
Rita Colwell
George Gamota
Leo Young

OVERVIEW

The attractiveness of the ERATO and to some extent the PRESTO programs stems from the amount and duration of funding and from flexibility. Additionally, these programs require only short research proposals.

ERATO projects differ from U.S. research projects funded by federal agencies in at least two ways: the complete freedom given to a project leader to carry on a large research project without review or interference for a relatively long period of time (five years), and the ability for a team of interdisciplinary researchers from universities, industry, and national labs to work together under one roof with responsibility only to ERATO. Since not all research projects can be like that, then how many should be? If some projects should be conducted in this or a similar manner, then how much funding would be appropriate? How should the topics and the leaders be selected?

These questions do not lend themselves to easy answers, but ERATO deals with the last question in an interesting way. ERATO attempts to attract top-quality researchers by actively searching out top-quality leaders. Not only must such a leader be a scientist or engineer with a proven record in research, but he (or she) must also be able to communicate and inspire others. To find such a person, ERATO does not advertise U.S.-fashion but goes about (somewhat undemocratically by U.S. standards) seeking out such persons and going after them, for years if necessary, and inviting them to submit ideas (short proposals). ERATO officials say that they find these leaders by taking a kind of informal survey among young researchers (such as might have participated in previous ERATO projects). It was not made clear to us just how this is done; rather we got the impression that it is a deliberately informal process. ERATO officials then invite unusual or high-risk proposals from these potential project directors. The one hundred or so short proposals received are narrowed down to about twenty by the JRDC staff. The authors are then invited for an interview with an advisory council which includes outside experts. The experts are given the opportunity to interview but do not pick the winners, since in most cases the winners are selected by JRDC. Some experts expressed a desire to be able to select the interviewees from a larger pool of candidates. This process is repeated annually.

ERATO projects are organized into three groups that are sometimes closely and sometimes loosely coupled, depending on the project director. Research is typically performed in three locations, and these may be quite far apart. As the project leader is usually in great demand, he (we never heard a "she") is always part-time on the project, devoting perhaps 10% or 20% of his time to it. He must give direction and cohesiveness to the project, but he also needs to select a good deputy to manage the science or engineering efforts on a day-to-day basis and he must have a dependable financial officer. The project leader is responsible for hiring these two critical people and the other dozen or more key personnel. As it may take one year, sometimes more, to find and rent a suitable location and fix it up for research, and furthermore, as much of the fifth and last year is taken up by winding down research, disposing of equipment, preparing to move out of the rented facility, and by job hunting for many (it could be one half) of the team members, it follows that the most productive years may be limited to three out of five. Thus ERATO projects are not only temporary research efforts, but, in contrast with U.S. projects, are also temporary stand-alone entities, not embedded in or reporting to organizations like universities or industrial corporations that existed before the project and continue after it is terminated.

ERATO was designed to allow Japanese researchers to work under conditions more like those in the United States, but it does have unique features that vary from the United States (i.e., challenging conditions without job security). Its success is evidenced by the fact that its budget will be increased as part of an economic stimulus package to revive the country's economy. Although designed to meet Japanese conditions and shortcomings in its research programs, the ERATO model has lessons to teach the United States. Here are some suggestions:

Reduce Proposal Writing Efforts

Research is recognized as an investment in the country's future, not as an entitlement. Such investment opportunities abound, but funds are limited. In the United States the frantic need to compete has caused U.S. researchers to spend more and more time on proposal writing, time that is taken away from the research itself. In contrast, in Japan historically there has been too little competition for research funding, a situation that ERATO has gone some way to correct. To alleviate burdensome competition in the United States, some federal agencies are limiting proposal lengths, but due to shrinking research budgets, the competition has become fierce. Thus the two systems are converging in a small way. Other measures could be taken; for example, very short pre-proposals could be used to select qualified researchers, somewhat along the lines of ERATO.

Appoint Part-time Project Leaders

An intriguing possibility is the selection of outstanding, part-time leaders, particularly from industry, to direct and guide even very large research projects. These leaders would have to be very carefully selected and provided with competent full-time deputies to manage the project. University restrictions on the use of outside help for thesis directors and part-time research faculty would have to be eased to stimulate this possibility.

Encourage More Risk-Taking

The idea of giving a few large projects to strong independent leaders without review or interference from the funding agency for a relatively long period (like ERATO's five years) would be novel in the United States, where accountability is emphasized. Accountability favors caution over risk-taking. It is appropriate where quantitative measures apply to predictable outcomes at the conclusion of the work, but progress in research depends on risk-taking; "safe" research that always pans out is not necessarily good research. Success may not be judged for years to come. Risk-taking by responsible, qualified individuals should not only be allowed, it should be encouraged. That is what ERATO aims to do. It does so by limiting the ERATO project leader's accountability to JRDC to fiscal matters, while his scientific "accountability" is left to his peers through his publications and their assessment of his scientific contributions.

Cast a Wider Net

Develop multidisciplinary groups from diverse fields, even when they seem far apart (e.g., medicine and engineering) to cross-fertilize ideas and to form networking that may be productive for a lifetime. (Such groups may be thought of as a form of lateral technology transfer.) The same trend - the formation of multidisciplinary teams from ever more diverse fields - has also been purported to be encouraged in the United States, especially by government agencies, but very few projects actually exist. This would seem to be particularly worthwhile when it involves advanced information technologies (software and communications), in which the United States leads the world.

Create a Real Industry-University-Government Team

Create a few interdisciplinary research projects composed of researchers from industry, universities, and government working together under one umbrella for a specified time period. The umbrella organization could be a professional society or consortium to assure that no single member organization would attempt to control the project and steer it in its own direction. All researchers would be employed for the duration of the project by the umbrella organization. Upon completion, the projects would be disbanded. Liberal unpaid leave policies or sabbaticals would be needed from researchers' home institutions to ensure the success of such a program.

[Previous Chapter][Top of Report][WTEC Welcome Page][Next Section]


Published: September 1996; WTEC Hyper-Librarian