Keith Parsons’ Drawing Out Leviathan provides a critique of the social constructivist interpretation of science by focusing on paleontological controversies over the dinosaurs. He concedes that social constructivists are correct about nonrational social and cultural factors playing a more significant role in the resolution of scientific disputes than traditionally acknowledged. But he argues that careful case studies of specific scientific controversies over the dinosaurs reveal that reason and evidence also play central roles. Parsons concludes by sketching an account of science that he believes will accommodate important insights of the constructivists while preserving the rationality and objectivity of science.

The first three chapters, which develop case studies from the paleontology of the dinosaurs, provide fascinating reading. Chapter 1 traces the history of the debate over matching the proper skull with the skeleton of Carnegie Museum's famous Apatosaurus (formerly known as “Brontosaurus”). It is now generally conceded that the skull that graced the museum's specimen for forty-five years belonged to a dinosaur (a Camarasaurus) that is not closely related to Apatosaurus. Parsons persuasively argues that reason and evidence played key roles at every stage of the debate, including the mounting of the wrong skull in 1934! Paleontologist Robert Bakker's controversial work on the “hot-blooded dinosaurs” is the focus of Chapter 2. Parsons reviews Bakker's impressive array of arguments for dinosaur endothermy, along with the critical responses from the paleontological community. He argues that the skepticism still evinced by many paleontologists towards Bakker's work is grounded in rational assessments of the evidence rather than the prejudice of an old guard faced with radical new ideas. In Chapter 3, Parsons evaluates David Raup's claim to have undergone a Kuhnian-style conversion to the Alvarez (meteorite-impact) hypothesis for the extinction of the dinosaurs. He contends that Raup's change of mind was actually motivated by rational and evidential considerations widely adopted in the natural sciences.

Chapters 4–6 critically evaluate the social constructivist approach to science. Parsons focuses on the work of three major figures, Bruno Latour, W.J.T. Mitchell, and Adrian Desmond. The evolution of Latour's thought from his infamous 1986 Laboratory life to the present is analyzed in Chapter 4. Parsons’ main argument against Latour involves an inference to the best explanation. He unfavorably compares Latour's constructivism with Marcello Pera's modest account of scientific rationality (The Discourses of Science), arguing that the latter provides a much better analysis of the case studies introduced in the first three chapters; Pera, who accepts some constructivist claims, argues that scientific controversies are ultimately decided by the strongest rational argument, although such arguments are not logically conclusive. Chapter 5 reviews W. J. T. Mitchell's account of the dinosaurs, in The Last Dinosaur Book, as hybrid objects where fact and fiction are inextricably intertwined. Parsons argues that Mitchell's account contains confusions and fails to show that myths, metaphors, and images involving dinosaurs have influenced paleontological research and theorizing in any significant way. Adrian Desmond's account of the Huxley-Owen controversy over birds and dinosaurs (The Hot-Blooded Dinosaurs: A Revolution in Paleontology) is the topic of the following chapter. Desmond argues that Huxley's view, which is now accepted by the paleontological community, was just as socially constructed as Owens’. Unfortunately, Parsons’ response doesn't directly address the specifics of Desmond's analysis. After conceding that Desmond's case is “powerful” (131), he launches into a long discussion of scientific progress, contending that social constructivists cannot explain it. The main burden of his argument is to show that paleontological methods have “advanced” since the time of Huxley and Owens. He argues that although there isn't a universally applicable scientific method, every scientific discipline utilizes a disparate variety of “lower-level” methods, some of which are accepted by closely related and even unrelated disciplines. Parsons’ contends that the adoption of such methods is rationally warranted. But his arguments are hard to pin down, presumably because it is difficult to argue for the rationality of a specific method without appealing to some sort of universally applicable account of what constitutes a rational scientific method.

The last chapter is devoted to transcending the science wars by reconciling differences between social constructivists and their scientific and philosophical opponents. It is doubtful, however, that any social constructivists will be appeased. Among other things, Parsons portrays them as confusing the rationality of science with an algorithmic model of theory choice. He sketches an alternative “dialogical” account of theory-change (closely based on the work of Richard Berstein) in which nonalgorithmic reasoning from empirical evidence is sufficient to rationally persuade but not to rationally compel. As earlier, his account suffers from a lack of attention to what constitutes nonalgorithmic rationality; reference to “practically learned skills of judgment and deliberation” (170) is philosophically unsatisfying.

In conclusion, Parsons’ case studies provide fascinating reading, and his arguments against constructivist interpretations of them are compelling. For these reasons, many scholars will find value in this work. However, the philosophical part of his discussion is somewhat disappointing. He exhibits an uncritical, almost reverent, attitude towards science, never delving very deeply into the really thorny philosophical problems about scientific rationality, progress, and objectivity.