This document is downloaded from DR‑NTU (https://dr.ntu.edu.sg) Nanyang Technological University, Singapore.

Review of : Joseph November (2012) Biomedical computing : digitizing life in the United States

Stevens, Hallam

2013

Stevens, H. (2013). Review of : Joseph November (2012) Biomedical computing : digitizing life in the United States, Isis, 104(1), 186‑187. https://hdl.handle.net/10356/100750 https://doi.org/10.1086/670905

© 2013 The History of Science Society. This paper was published in Isis and is made available as an electronic reprint (preprint) with permission of The History of Science Society. The published version is available at: [http://dx.doi.org/10.1086/670905]. One print or electronic copy may be made for personal use only. Systematic or multiple reproduction, distribution to multiple locations via electronic or other means, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper is prohibited and is subject to penalties under law.

Downloaded on 02 Oct 2021 13:19:42 SGT 186 BOOK REVIEWS—ISIS, 104 : 1 (2013) signed to challenge underlying political reasons Her tightly written book plays with and show- for mistrust of the “establishment,” including cases big ideas of scientific authority, AIDS the scientific community. In the second half of counterknowledges, and political leadership. the book, Nattrass turns to AIDS denialism, The cumulative effect is a book rich with insight which she suggests has three major characteris- that should find a wide audience among scholars tics: skepticism toward the science of HIV of health politics and historians of science. pathogenesis and treatment; neglect of advances in LUCAS RICHERT antiretroviral treatment; and support of alternative, unproven therapies. She capably captures the inter- connecting networks, institutions, and key personal- Joseph November. Biomedical Computing: ities that propagate AIDS denialism. In particular, her Digitizing Life in the United States. 344 pp., assiduous chronicling of numerous counternarratives illus., apps., index. Baltimore: Johns Hopkins about AIDS is instructive, as is her identification and University Press, 2012. $60 (cloth). treatment of such figures as Peter Duesberg, Eleni Papadopulos-Eleopulos, and Roberto Giraldo, Twenty-first-century biology depends deeply among others. and widely on computers. Biologists use comput- Two core strengths of The AIDS Conspiracy ers for collecting, storing, managing, and analyzing are the explicit focus on political leadership and data and for operating instruments, simulations, visu- the structural approach offered by Nattrass. alization, communicating, and presenting their work. First, her handling of South African President It is hard for most contemporary biologists to imag- Thabo Mbeki and his AIDS policy illustrates a ine their lab or their work without these machines. complicated relationship among scientific ex- But all this is a relatively recent development. perts, interest groups, and policymakers. When For decades after digital electronic computers he took seriously the claims that HIV science were developed in the 1940s most biologists was fundamentally flawed and initiated a Pres- continued to believe that these calculating ma- idential AIDS Advisory Panel incorporating chines were ill suited to the qualitative reason- AIDS denialists, Mbeki rejected the strongest ing needed for good biological work. Somehow, available evidence and acted against the best between the 1960s and the present, biology has interests of his own population. In Nattrass’s “evolved from exemplars of systems that com- view, it was his leadership that was flawed, not puters could not describe into exemplars of sys- the science: Mbeki’s willingness to promote sci- tems that computers could indeed describe” (p. entific debate was reckless and he ought to have 7). How did this transformation take place? And deferred to experts. Second, Nattrass charts the what are the consequences of biology’s and group dynamics of the AIDS denialist commu- medicine’s new dependence on computers? nity and classifies four separate and significant Joseph November’s book offers an account of players: cultropreneur, hero scientist, living the earliest attempts to deploy electronic com- icon, and praise-singer. While the “cultropre- puters in biology and medicine between the neur” offers alternative cures, the “hero scien- 1940s and the 1970s. This approach promises tist” lends credence to the rejection of mainline the reader important insights into the complex medications and the science underlying them. and multiple origins of fields such as bioinfor- “Living icons,” for their part, stand as living matics, computational biology, and medical in- examples of the concept, whereas “praise- formatics. November claims not only that at- singers” trumpet the positive message of the tempts at computerization in the 1950s and AIDS denialist movement to the broader public. 1960s altered the practices of biomedicine (al- It is worth noting, on the one hand, that this though not as deeply as some enthusiasts categorization is explicitly Nattrass’s and that hoped), but also that biomedical computing had she is not revealing an evident framework of an important influence on the development of actors; on the other hand, this taxonomy of in- computing practices. dividuals provides a means to confront AIDS The story of biomedical computing begins denialism more effectively and thereby redress with operations research (OR). During World the long-run consequences of ignoring the best War II, operations researchers devised ways of scientific evidence on HIV/AIDS. reducing the messy problems of warfare to simple The AIDS Conspiracy takes aim at relativist models and equations that could be analyzed and philosophical stances, antiscience conspiracy computed (using either human or, as they became mongers, practitioners of pseudoscience, and available, electronic computers). Many of those who misguided policymakers. Nattrass’s qualitative introduced computers into biology—including John and quantitative arguments are robust and per- Kendrew, who applied computers to protein crystal- suasive, while her tone is cool and measured. lographic data, and Robert S. Ledley, who attempted

This content downloaded from 155.69.24.171 on Thu, 31 Aug 2017 01:24:15 UTC All use subject to http://about.jstor.org/terms BOOK REVIEWS—ISIS, 104 : 1 (2013) 187

(unsuccessfully) with George Gamow to crack the difficulties for November’s narrative. There is genetic code using a computer—were trained in OR. still a wide gap between the efforts of the NIH, This legacy meant that the National Institutes of Stanford, and the LINC and the massive, ubiq- Health’s push to computerize biology and med- uitous, DNA-centered biological computing of icine in the 1960s (largely driven by Ledley and the 1990s and 2000s. Even in 1980, at the end of his colleague Lee B. Lusted) began as an at- the period on which November focuses, com- tempt to mathematize biology, to find problems puters remained rare in biology. What happened that would yield data that could be analyzed to computing in biomedicine in the 1980s and with the help of computers. The computer was a 1990s was different in scale and kind from what calculator or a tabulating machine, and deploy- went before. November’s book doesn’t show us ing it in biology meant transforming biology, how we got from one to the other. The stories along the lines of high-energy physics, into Biomedical Computing tells are crucial: their something resembling a large engineering or importance, however, is not in telling us about industrial project. how computers have succeeded in colonizing One of the more successful and unique proj- today’s biology but, rather, in helping us to ects to emerge from these 1960s intersections understand why and how computers failed to between computing and biomedicine was the find a significant place in biomedicine before LINC (Laboratory Instrument Computer). De- 1980. signed by Wesley Clark under the sponsorship HALLAM STEVENS of MIT’s military-funded Lincoln Lab, the LINC was a small (by 1960s standards) machine that could be owned, programmed, and adapted Duncan Wilson. Tissue Culture in Science and for use in individual laboratories. The NIH had Society: The Public Life of a Biological Tech- committed itself to funding large, centralized nique in Twentieth Century Britain. x ϩ 183 computing resources for biologists; Clark’s vi- pp., illus., bibl., index. New York: Palgrave sion was of a computer as a convenient tool Macmillan, 2011. $80 (cloth). equipped with graphical capabilities, fully at the disposal of one team. The LINC, November Another title for this book could simply have argues, proved popular and influential, not been “Tissue as Culture,” as its main idea is that merely among biologists but also on computer tissue culture—the art of keeping human or an- designers. Its small size, graphics capability, imal cells alive outside their original organ- and real-time user control anticipated many of isms—is a prime site for investigating the en- the features of 1980s personal computers. gagement between science and culture and the Another contact zone between computer sci- blurring of the line between the two. entists and biomedical researchers emerged at Tissue Culture in Science and Society offers a Stanford University in the 1970s. There the series of case studies that follow the develop- NIH sponsored the application of computing, ment of practices and attitudes to tissue culture and particularly artificial intelligence, to a in the twentieth century from its inception in range of projects, including the search for life 1907 to current discussions. The main focus is on Mars, the prediction of chemical structures on the public reception of tissue culture, and the from mass spectrograph data (DENDRAL), cases are often chosen not just because of their and the planning of molecular genetics exper- centrality to scientific development but also for iments (MOLGEN). Ultimately, however, their role in public debates about tissue culture. Stanford’s efforts at computerizing biomedi- The book offers only a little framing, but it cine experienced as many failures as suc- explicitly presents itself as a British version of cesses. Indeed, many of Stanford’s projects Hannah Landecker’s Culturing Life: How Cells would strike present-day biologists as “quixotic Became Technologies (Harvard, 2007). There is or downright bizarre” (p. 221). not, however, just a difference between British It was not just Stanford: many of the NIH’s and American case histories; this is a very dif- efforts toward biomedical computing also met ferent project. Where Landecker is interested in with fierce resistance or insurmountable diffi- general changes in the biological sciences in the culties. The hoped-for transformation of biology direction of greater plasticity and manipulabil- did not take place (and hospitals and physicians ity, Duncan Wilson is interested in the connec- proved even more resistant). Even the LINC, tions between science and the public and in ultimately cut off from military funding and demonstrating that these interactions are histor- “cast out” of MIT, was marginal to mainstream ically contingent. computer development in the 1960s and 1970s. The first half of the book follows the Strange- The limited lifespan of such endeavors raises ways Research Laboratory in Cambridge in its

This content downloaded from 155.69.24.171 on Thu, 31 Aug 2017 01:24:15 UTC All use subject to http://about.jstor.org/terms