There at the beginning: we’re still “emerging,” maybe forever

Introduction

I had the honor of serving as Editor of Policy Sciences twice: from 1974 to 1976 and then again in 1991 for little over a year. My reflections focus on the earlier period. I was only “pinch hitting” in 1991 to help the transition from Bill Ascher to Doug Torgerson, and so can contribute little from that particular time as compared to my initial editorial experience.

“At the beginning,” so to speak, was a time of great enthusiasm and high expectations for policy analysis. Based on genuine successes during World War II, and even before during the Great Depression, of using a range of different academic disciplines and intellectual traditions to focus on, grapple with, and actually figure out many pressing and complex problems, interdisciplinary studies and analyses had emerged to some prominence by the late 1960s and early 1970s. Operations research, applied mathematics and economics, many of the “hard” sciences such as physics and chemistry, and even “softer” ones such as psychology, psychiatry, sociology, and political science each could point with some pride to useful contributions to solutions to crush poverty and more importantly to win the war.

Efforts to gather together many of the best and brightest of that era in the immediate post-World War II period occurred. Similar efforts to provide intellectual coherence and guidance for future developments did as well. Neither of these activities would, or in retrospect even could, proceed without challenge, as you will discover in the following brief recollections and personal reflections. I must stress that my standpoint, perspective, and experience are singular. Others involved at the time will obviously be willing to report things differently.

Systems analysis

The Air Force took the lead in the immediate post-war period to retain, collect, and capitalize on the diverse intellectual skills and talents revealed during the war itself. John von Neumann, Herman Kahn, John Williams, Albert Wohlstetter, Robert McNamara, Charles Hitch, and many others were either recruited to Project Air Force or strongly influenced by it and then to its fuller exposition in the form of The RAND Corporation—all this beginning in 1947 and continuing onward. Systems analysis was an over-arching label to describe the multidisciplinary, multi-method, problem-oriented approach that had emerged from many different war-time experiences. Logistics, input–output economics, weapons effectiveness studies, program and project planning—as well as cost-effectiveness studies of complex operations and systems, modeling, gaming, and simulation to enable hard thinking about how and when weapons (especially nuclear ones) could be made, deployed, and used, as well as many other more specific challenges came to the fore as The RAND Corporation took shape (Smith 1966).

Ed Quade, the first editor of Policy Sciences, and himself a founding member of Project Air Force and then a stalwart at RAND, including as Head of its Mathematics Department, defines systems analysis thusly:

[A] systematic approach to helping a decision maker choose a course of action by investigating his full problem, searching out objectives and alternatives, and comparing them in light of their consequences, using an appropriate framework—in so far as possible analytic—to bring expert judgment and intuition to bear on the problem (Quade and Boucher 1968: 2).¹

For systems analysts, in these terms, work was a process that included the conceptual or formulative phase, a research phase, an analytic or evaluative phase, an interpretative or judgmental phase, and verification or a scientific phase (Quade and Boucher 1968: 33–53). The need to decompose or segment different kinds of intellectual activity is evident, although recognition that much of the fundamental work in this area had already been done and done more completely by Harold D. Lasswell and several others was not (Lasswell and Lerner 1951). Opportunities to enrich systems analysis, and also operations research, by incorporating the human centered and valued-based features of the policy sciences existed but were never incorporated, thus leaving the former to depend on rationalistic and reductionist premises and on solution goals of optimality and efficiency. The solution preference expressed in the policy sciences, in contrast, is for “increasing human dignity through inclusion in all the value-shaping and—sharing processes and institutions, existing and conceivable.” People > Systems, simply put. (Brewer and deLeon 1983: 154, and Tables 6-1, 6-2.)

The possible blending of the policy sciences with systems analysis continued to be “just within reach” but never successfully achieved—although RAND and systems analysis proved to be a supportive host for the journal Policy Sciences in the 1970s.²

Changes in the landscape

Other developments at this time featured various efforts to reform and rebrand more conventional disciplinary programs and institutions as means to increase their relevance for the growing array of new policies associated with Presidents Kennedy and Johnson. Public administration became government or public management, political science became policy studies, or public policy, and area studies morphed into different versions of development. Harvard, the University of Michigan, UC Berkeley, and Duke (among others) all made significant and lasting contributions. Several new professional journals and book series emerged as well.

RAND itself was also confronting significant pressure to change following the revelations exposed by RAND employee Daniel Ellsberg’s making the Pentagon Papers public in 1971. A traditional dependence on military customers and problems yielded to a more varied portfolio of domestic initiatives, e.g., health, housing, education, demography. Early engagement with many of the new clientele also revealed numerous opportunities to bring systems analysis and systems analysts into new arenas. The market for “policy analysts” seemed ready-made for high-level professional training efforts to fill it. Universities were building on their strengths, and RAND decided to build on its own, distinctive ones to compete. And so began the new RAND Graduate Institute (RGI) and with it the new journal Policy Sciences.

The Policy Sciences Emerge: To Nurture and Structure a Discipline

I participated directly in both of these initiatives. RAND invited me to be a consultant in 1969 as I was putting the finishing touches on my Yale Ph.D. I enjoyed the people and the setting very much and so accepted an offer of full-time employment in the fall of 1970. This also allowed me to join forces with Martin Shubik, one of my dissertation advisors who took a year’s sabbatical leave in 1970–1971. Our work on models, simulations, and games resulted in numerous RAND publications as well as several books together (Brewer and Shubik 1979, contains a complete bibliography).

Harold D. Lasswell was another of my thesis advisors who also served as a consultant to RAND beginning sometime in the late 1950s or early 1960s. Indeed, he was asked to assume major responsibility as the first Chair of the RGI Academic Advisory Board sometime in late 1969.³ We interacted often as a consequence. A consistent topic of concern was how to offset the domination of the curriculum by economists, applied mathematicians, operations researchers, and engineers—a real possibility since the RGI program in the beginning relied almost entirely on resident RAND analysts for instruction.⁴ As the RGI grew over the years since 1970, it has changed its name twice and included outside faculty, especially from UCLA.⁵ Its systems analysis, operations research, applied economics characteristics have, as Harold’s concern foreshadowed, come to pass. The rational, reductionist perspectives and methods dominate to the detriment of the humanistic ones given heavy weight in the policy sciences. In 1971, however, hope and optimism ruled and so he convinced me to create and teach one of the core, required courses: “The Scope of the Policy Sciences.” A year or so later, he also convinced me to assume editorial responsibility from Ed Quade, who was set to retire from RAND at the end of 1973.

At the first class session, I encountered a serious communication problem—not something rare when the policy sciences are involved. The decision process model’s terms did not map very well on the then prevailing systems analysis and workaday policy analysis concepts. Intelligence, Promotion, Prescription, Invocation, Application, Termination, and Appraisal are individually and collectively precise and logically interrelated—as is the case for virtually all of Lasswell’s conceptual and theoretical works. Precision and logic aside, these terms did not correspond very well with functional labels then in use. Intelligence certainly initiates problem definition and opens up a range of actions to find data and information with which inventive ways might deal with that problem. Intelligence had fallen far from favor, however, largely as a result of its abuse during the Vietnam War era—a particularly sensitive topic at RAND around this time. A different label was needed to avoid dragging along unwanted political and symbolic baggage. Invocation and Application, so said my students, seemed an awful lot like implementation, which was just then coming into vogue as a result of a book of the same name recently published by Pressman and Wildavsky (1973).⁶ Evaluation was a collective and catch-all label applied to an emerging and burgeoning field of evaluation research. In addition, Assistant Secretaries for Planning and Evaluation were popping up in government departments and agencies all over the place. Appraisal didn’t work nearly as well as evaluation to capture the realities of both research and operational activities. The concept of Termination was well received by the class, but why it came before appraisal/evaluation proved a logical problem for the class—and me, after I thought about it.

What resulted from several thoughtful and productive conversations with Lasswell and the students in the RGI “Politics” course is the six-element Invention/initiation, Estimation, Selection, Implementation, Evaluation, Termination scheme later developed at length for the textbook, The Foundations of Policy Analysis (Brewer and deLeon 1983).⁷ I was truly concerned that Harold would understand and approve, and he did. “Functional equivalents are sometimes required,” he said when I asked. I wish that I had been more careful explaining all this to others, especially with respect to the complexity of usual interactions among the different decision functions rather than any wooden inevitability of a step-by-step process. The functions are not actually “stages,” despite being so characterized in several critiques. As Susan Clark puts it, “Because these are functions or activities of decision making and not stages, they are often carried out simultaneously, rather than sequentially, and they are often mixed together in complex ways.”(Clark 2011: 59).⁸

Prior to taking over as editor of Policy Sciences, I thought it would be important to align, if not even implicate, the policy sciences discipline with RAND, where it was viewed suspiciously—if anyone even cared at all. The decision process modifications created for the first RGI policy sciences course (locally known as “the politics course”) served as the topical structure in “The Policy Sciences Emerge: To Nurture and Structure a Discipline,” a RAND publication first published under this title as RAND P-5206 in April 1974. With the same title and only slight editorial modification, the RAND paper was then published in Policy Sciences (Brewer 1974).

Twenty-three years after Dan Lerner and Harold published The Policy Sciences, where everything essential to characterize and motivate the policy sciences was laid out, the emergence of the policy analysis craze and movement along with the RAND Corporation’s commitment to graduate education in policy analysis seemed auspicious. The basic problem, as I understood it as I took over as second editor, was “to sharpen the identifications, expectations, and demands of individuals beginning to call themselves ‘policy scientists’.” (Brewer 1974: 239). The prior 23 years had not produced many significant “wins” for the policy sciences for several different reasons: entrenched disciplinary interests, limited financial support for “risky” creative ventures and experiments, intellectual and practical threats to customary perspectives and practices, and simple lack of imagination in many different quarters. Besides, taking the policy sciences seriously almost without exception means accepting considerable responsibilities and attendant additional efforts to do it well.

Besides a simple recitation of the modified decision process framework, I tried to identify some emerging areas of policy concern, based largely on the experiences from the “Faculty Seminar on the Study of the Future” that Harold organized and ran at Yale in 1968–1970.⁹ Genetics, especially human-caused genetic modifications, women’s rights, higher education (funding and also international aspects), termination of various institutions—especially dysfunctional political ones, nuclear war—with a focus on the Communications, Command, Control, and Intelligence (C³I) aspects of it, and energy futures and crises. My hope was to stimulate problem-oriented efforts which then might be featured in subsequent issues of the journal. To capitalize on work I knew was nearing fruition at RAND and elsewhere related to several of the decision functions, I committed to three special issues devoted to termination, implementation, and initiation, and most of these worked out subsequently and reasonably well.

Another publication effort nurtured by RAND was the Policy Sciences Book Series, under the general editorial direction of Yehezkel Dror, an Israeli political scientist/public administration scholar at the Hebrew University. Dror was invited to RAND on a consulting arrangement during the 1968–1970 period with specific tasks for the book series, joining the board of Policy Sciences to promote international interest in it, and to generally participate in RAND’s new teaching programs at RGI. His 1968 book, Public Policy-making Re-examined: Design for Policy Sciences, was within the revisionist camp of those seeking makeovers in political science and public administration. It did not demonstrate a deep or careful understanding of the policy sciences discipline (Dror 1968). The book series was with one standout exception, similarly limited. Of the initial six books published in 1971, two were by Dror (including a revision of his 1968 Public Policy-making Re-examined), an historical rendition of the policy evolution of ballistic missile defense, an early effort to link education assisted by emergent and limited computers to organizational theory, and a case study of how policy evaluation research was being put into practice in several anti-poverty and equal-opportunity programs. The standout exception was Harold’s A Preview of Policy Sciences, perhaps one of the most accessible and concise summaries of the discipline ever made. (Lasswell 1971).

Moving on

A wonderful professional opportunity to return to Yale and become one of the founding faculty members of a “radically different” management school came my way in early 1974. I accepted Yale’s offer to join The School of Organization and Management and also did two additional things more or less simultaneously: I took an invitation to spend the 1975–1975 academic year at the Center for Advanced Study in Palo Alto and also convinced RAND that I could discharge editorial duties while making the slow-motion move back to New Haven.¹⁰ Thanks to outstanding administrative help from RAND’s Janet deLand and several staff editors, as well as continuing financial assistance for the journal itself, this proved an acceptable solution for the 1974–1976 end of my term. It eventually turned out that I was able to take the journal to Yale for my final year before having to turn it over to Tom Anton, then a senior faculty member of the Institute for Public Policy Studies at the University of Michigan.

On reflection, the idea of “emerging” is perfectly suitable for our shared enterprise. The project will never be finished or complete. Human foibles that manifest in institutional difficulties and limited resources are not going to vanish either. But, as I tell my students (and myself) always remember why we are doing this: To improve human dignity. Of course, the work is never going to be done, but that doesn’t mean it’s any less essential to try.