American Scientist

Investigative Pathways: Patterns and Stages in the Careers of Experimental Scientists. Frederic Lawrence Holmes. xxii + 225 pp. Yale University Press, 2004. $35.

The late Larry Holmes was, to put it simply, our best historian of experiment. His death last year was a major loss to the history-of-science community. Using published sources, laboratory notebooks, correspondence and, where possible, personal interviews, he gave us extraordinary accounts of the experimental work of Antoine Lavoisier, Claude Bernard, Seymour Benzer and Hans Krebs, and of what has been called "the most beautiful experiment in biology"—Matthew Meselson and Franklin Stahl's demonstration of how DNA replicates. Now we have the excellent short volume Investigative Pathways (published posthumously), which provides us with generalities about the experimental life that Holmes gleaned from those studies. These conclusions are amply supported with details from his earlier works and also serve as a very nice introduction to them. Here Holmes examines experimental careers from the broadest scale—what he calls "investigative pathways"—down to the fine structure of the "Aha!" or "Eureka!" moments. This is a book that I will use in my own work.

Although Holmes focuses on those who have made significant contributions, he suggests that his observations apply to almost all scientists. This claim is supported by the work of the other historians he cites and by Jed Z. Buchwald's afterword to the book; also, my own experience in science confirms it.

Holmes notes that even important scientists take a considerable period of time to acquire mastery in a particular fairly narrow area of research:

The older heroic image of the great scientist as one who possesses from the beginning more profound, unerring insight than those already active in the field, who takes up from an entirely original point of view problems that have stalled his predecessors, or that they have failed to notice as problems, can no longer be taken seriously.

Because of the time needed to develop expertise, scientists tend to continue working in a single area for a substantial length of time, perhaps even throughout their lives. Hans Krebs, for example, spent almost his entire career working on intermediary metabolism. Holmes points out that this concentration on a single relatively narrow area of research is a very effective strategy, allowing scientists to recycle their expertise.

At the much shorter time scale, Holmes discusses momentary flashes of insight or discovery. Although both scientists and historians often give accounts that support the existence of such dramatic moments, he shows that what may appear to be a sudden break from the past sometimes turns out on closer examination to be a progression of small steps. He cites the work of Howard Gruber, who believes that memory simplifies the past, conflating the single most powerful of a series of similar experiences with that sequence. Holmes notes that Seymour Benzer, in an account written 11 years after the event, remembers his work on the mapping of the T4 bacteriophage as a sudden insight: "To me the significance of this result was now obvious at once." However, when Holmes and Benzer examined Benzer's laboratory records for that period, they found that the series of events leading to that insight were spread over a period of 15 months. Holmes also reports that Krebs denied that any of his discoveries took place in a short period of time. However,  Holmes presents other instances in which a discovery does seem to have occurred within a very short span—for example, James D. Watson's insight concerning the structure of DNA. He concludes that in fact, the Eureka view and the notion that creative thinking is a slower growth process are not opposed and mutually exclusive ideas, and that we must keep this in mind when studying the fine structure of investigative pathways.

Holmes also discusses several recent trends in the study of science. Commenting on the problem of presenting the complexity of the technical content and argument of science (something that has disappeared from much of contemporary history of science) and on the view that science is merely a social construction, Holmes remarks, "If we are to truly understand experimental practice, however, we cannot evade that problem, or be satisfied with surrogate solutions that would tell us that even the knowledge pursued or acquired by . . . towering intellects is merely constituted by social processes." Commenting on the current view that science is purely local, he refers to his own histories of experimenters:

Each story is unique, local and personal. To that extent I agree with the constructionist viewpoint that all scientific knowledge originates in local, contingent circumstances.

Told at a local level alone, however, such stories have little meaning. Each of these stories is situated within successively larger stories. Each episode is connected to everything that the experimenter had previously experienced. . . . None operated, even for a day or a moment, in isolation from the contemporary state of the field in which he practiced. None of the moves that we can follow him making from experiment to experiment could have led to recognized achievements had his thoughts and actions been based purely on local considerations. Each story is, therefore, at once personal and unique, and representative of features common to scientific research in general. Each story is a small piece both in the individual quest for a meaningful, well-spent life, and in the collective advance of science.

A book review as short as this one can provide only brief glimpses of the richness, complexity and nuance of Holmes's observations on experimental practice. I strongly urge you to read this truly excellent book in its entirety.—Allan Franklin, Physics, University of Colorado