Main Menu · Search ·Current Issue ·Contact ·Archives ·Centennial ·Letters to the Editor ·FAQs
| Books: A Discipline Dissected | Coming Soon to Your TV |
| Open Book - The Faculty: Are They Dutiful? | Music: Satire in 4/4 Time |
| Off the Shelf | Chapter & Verse |
|
I have to admit that I had certain misgivings when I picked up this book.
For one thing, the sheer heft is a bit intimidating--800-plus pages of text.
More important, I have become somewhat leery of 
books about the social structure of science. All too often they involve flights
of philosophical fancy about scientific theories or events of which the author
has almost no understanding. When I read these books, I see nothing in them
that resembles the day-to-day existence of the working scientist. In fact, I
often ask myself, "Is this how an atom would feel if it could read a physics
textbook?"
I wasn't more than a dozen pages into Image and Logic when I realized that such fears were groundless. This is a major work by a man who clearly believes that if he is going to interpret a particular theory or experiment, his first job is to get his hands dirty and understand exactly what the theory or experiment is. It's a refreshing change to find an author who can converse with equal facility on the details of the design of bubble chambers and the anthropological analysis of different subcultures within a scientific discipline.
Galison's goal is to understand the workings of the scientific enterprise, and his vehicle is a detailed study of a particularly important period in physics--the time from (roughly) the turn of the century to the early 1980s. He looks at what he calls "microphysics," which is the study of the basic structure of the material world. During this period, the subject matter of microphysics evolved from atoms to the atomic nucleus to the elementary particles that make up the nucleus (and later to quarks, but that development is largely outside Galison's time frame).
His thesis is subtle, complex, and many-faceted, impossible to summarize in a single sentence. In fact, the book is best described in terms of an analogy that Galison uses to describe the structure of science itself, the analogy of a steel cable. Unlike a chain, which proceeds linearly from one link to the next, a cable is made of many interwoven strands, each of which contributes to the final product. And just as an understanding of the cable involves an understanding of each strand and also of its relation to all the others, understanding science involves seeing the complex interplay among the many different elements that make it up.
The link between theory and experiment in science has long been a subject
of study (and controversy). Theory and experiment are clearly two of the strands
that make up science, but to these Galison adds a somewhat surprising third--the
strand of instrumentation. He argues that the physical production of the measuring
instrument, the building of a reliable recorder of nature, is a separate and
coequal part of science, and as such is as much influenced by the surrounding
culture as are theoretical ideas and experimental design. 
Looking at science as, at least in part, the consequence of these three strands already has important implications for the way we think about scientific change. It is popular these days to think about science as proceeding through "paradigm shifts"--sudden discontinuous changes in our ideas about the structure of the universe. What Galison points out is that major discontinuities in one strand, such as the development of quantum mechanics by theorists, don't correspond in time to discontinuities in the other strands. The reality of scientific change is a lot more complex than many writers have imagined.
It is fitting that a good deal of the book is devoted to a detailed discussion of the development of the instruments that scientists have used to probe the ultimate structure of matter. Within this strand, Galison defines two basic traditions. In one, the goal is to provide a direct visualization of the collisions of elementary particles. Devices called the cloud chamber and bubble chamber (which I'll talk about in a moment) produced stunning pictures of these collisions that still fill our science textbooks. In the other tradition, the goal is to measure the presence of the particles without actually trying to get a picture of what they're doing. The familiar "click" of the Geiger counter when a radioactive particle goes through it is an example of this sort of instrumentation. Calling these the "image" and "logic" traditions respectively (hence the title of his book), Galison goes to great lengths to describe how the two performed a kind of stately waltz through the twentieth century, first one leading, then the other, until they finally coalesced (more or less) in the 1970s and 1980s.
What impressed me most about Galison's discussion of the instrumental strand was the tremendous breadth and depth of scholarship that he brings to this study. His coverage of the image tradition, for example, begins in late Victorian England with a device called the cloud chamber: a cylinder with a piston at the bottom. When the piston was pulled down, water vapor in the chamber condensed into a cloud. The Victorian scientists who developed this device weren't primarily interested in studying the constituents of matter. Instead, they were trying to reproduce, in miniature, the clouds and storms that can be seen in nature. In a sense, they were in the romantic tradition, trying to re-create nature rather than dissect it.
As it happened, however, the droplets in the cloud chamber tended to form preferentially around atoms that had been disrupted by the passage of high-energy particles from space--what we call cosmic rays. Eventually, the original purpose of the device was lost as physicists began photographing the tracks of droplets that traced out the paths of elementary particles.
After World War II, the cloud chamber was replaced by a similar device called the bubble chamber, in which bubbles forming in very cold liquid hydrogen took the place of the water droplets. The bubble chamber, together with the substitution of particle accelerator beams for cosmic rays, is what drove elementary particle physics through the 1960s and 1970s. (Incidentally, the shift from cloud to bubble chambers is one of those discontinuities in the instrumental strand that wasn't accompanied by any particular change in the theoretical one.)
I can remember some of the events Galison describes from the early part of my career, so there was a strong personal (as well as intellectual) resonance for me in this book. I was particularly taken by his detailed and careful description of how developments in instruments changed the way scientists worked, and even the definition of what a scientist is. The classical definition of an experiment involves a single person, perhaps with a few students, building an apparatus, gathering and analyzing data, and, finally, interpreting and publishing the results of the experiment. As microphysics progressed from nuclei to particles to quarks, however, the instruments became progressively more complex and difficult to build and operate. Little by little, each of the classical functions of the scientist was handed over to other groups, until in the modern experimental collaboration it is often difficult to tell the difference among a physicist, an engineer, and a computer programmer.
Galison traces this transition in some detail. A cloud chamber, for example, could be built and operated by a single physicist, often working in isolation at a mountaintop observatory. As time went on, however, and the number and complexity of the photographs grew, the physicist running the experiment could no longer examine each one in detail. Instead, a group of people known as "scanners" (usually women) was brought in to do this job. Since image-tradition experiments often relied on a single "golden event" to establish the existence of a new kind of particle or a new interaction, the question of authorship rose immediately. Who gets credit for the discovery, the scientist in charge or the person who actually identifies the new event? (As a matter of historical fact, early cloud-chamber papers often identified the scanner by name, a practice that was quickly dropped in favor of a policy of anonymity. Today the issue is moot because the analogous job is done by machine.)
After World War II, a new corporate way of doing experiments emerged, driven by the experience and prestige of the "Los Alamos Men." The production protocols, chains of command, and safety standards of wartime research began to make their way into the physics laboratory. At Berkeley, for example, Luis Alvarez headed a group of 30 to 60 people who built the first large bubble chambers and accomplished much of the exploration of the new world of elementary particles (an accomplishment for which Alvarez was awarded the Nobel Prize). I can remember being warned as a graduate student during this period to avoid going into experimental research because "you'll just be a cog in the machine."
Today, this trend has accelerated, as detectors costing tens of millions of dollars and requiring hundreds of scientists to build and operate have come into vogue. Galison paints what (to me, at least) is a chilling picture of life in these groups, with "spokespersons" elected by the collaboration at large and detailed lists of criteria for who will and will not be included in a list of authors on published papers. An example: on one project, an author had to have worked "at least one half of the average number of shifts per person during the relevant run." What a long way from the lone scientist on the mountaintop!
As befits a world that has become so complex, Galison calls on the techniques of the anthropologist to analyze the way scientists communicate with each other across the different strands that make up their discipline. In what may well be the most important contribution of the book, he borrows from cultural anthropology to introduce the notion of a "trading zone." In trading zones, people of different cultures develop pidgin languages to carry out their business. It is not necessary that speakers of pidgin languages share deep philosophical outlooks, but only that they be able to converse to get the job done. Over time, pidgin languages may disappear (if the community of interests between the two groups evaporates) or grow into the full-fledged languages we call creoles (if that community of interests grows). In the same way, Galison argues, engineers, physicists, and computer scientists involved in building a particular instrument will develop the language needed to talk to each other even though a specific term ("electron," for example) may mean something rather different in the culture of each group. In this scheme, it is the pidgin languages developed in trading zones that cement the different strands of the scientific enterprise together.
Image and Logic is not a book to be read for an evening's diversion. It is a multifaceted, complex, and richly articulated look at the enterprise we call science. But in the end, I think, only an analysis as complex as this one can hope to capture the reality of a human activity as varied as science.