Carl Gillett is convinced that current discussions about reduction and emergence, especially in philosophy, remain mired in an early twentieth-century understanding of what is at stake. That dispute was about the scope of compositional explanations across the sciences. Except for residual worries about, for example, phenomenal consciousness, the sciences throughout the twentieth century settled that dispute. “Everyday” reductionism won; all the entities, properties, relations, and processes in the world either are fully composed entities or are the most basic constituents that make things up. A new reductionism-emergentism dispute, however, is now raging in the sciences. It is over the nature of composition itself and the determinative relations between (some) composed entities and the powers of their parts. Philosophers have mostly missed this new dispute. Their current consensus, that “reductionism” is dead and “nonreductive physicalism” triumphant, results from two errors. The first is a widespread but mistaken acceptance of a “semantic” account of reduction(ism), rooted in logical positivism but still holding sway. The second is a “metaphysics for science” practice, as opposed to a “metaphysics of science” one more common among “working scientists” (Gillett’s term). By following Gillett in adopting the latter and clearly articulating the complex nature of “compositional explanation” at work throughout the sciences, we learn that there are currently three “live” positions within the contemporary scientific reduction-emergence dispute, two reductionist and one emergentist. As yet no one of these positions can claim a single successful case from science that fully meets Gillett’s explicit “threshold” for empirical adequacy (although his sympathies are with scientific emergentists).

(Disclaimer: This reviewer is doubly a target in Gillett’s book, first in chap. 1 for my unwillingness to take metaphysics seriously in advocating reductionism in scientific practice and later, in chap. 10, for advocating an unsuccessful simple fundamentalist “local” scientific reduction of psychology to molecular neuroscience.)

Gillett begins with a detailed account of ‘composition’ in science, drawn from the widespread “astonishingly successful” practice across the sciences of providing ‘compositional explanations’. He draws his initial account from two paradigmatic cases: the explanation of the diamond and its properties in terms of the bonding and alignment of its carbon atoms and the explanation of the ion channel and its properties in terms of its protein subunits. He spells out 15 criteria any successful account of scientific composition must meet and defends the inclusion of each one. He argues that his account meets all 15, while alternative philosophical accounts (“functionalist” accounts like Jaegwon Kim’s, “manipulability” accounts like Carl Craver’s) fail to meet some. Armed with this account, Gillett then articulates scientific reductionists’ metaphysics of science and then scientific emergentists’ alternative. In each case he carefully explains key contrasts between his accounts and ones currently in philosophical favor. He stresses numerous important consequences of his newly grounded accounts. Scientific reductionism as articulated by Gillett easily accommodates multiple realization, the existence of composed kinds (a “collectivist ontology”) and their properties, relations, and processes, and a host of other “semantic antireductionist” features; similarly for his scientific emergentism, which advocates a kind of “strong emergence” widely thought by current philosophical orthodoxy—incorrectly, according to Gillett—to be incoherent. On Gillett’s reconstruction, the reductionist’s argument from composition is invalid. It ignores the logically consistent possibility of accepting both compositionality of all entities and strong emergence, as some scientific emergentists do.

In the end, what is at stake between the competing “live” views is not a difference over the comprehensive scope of compositional explanations. Both scientific reductionists and emergentists acknowledge that point. Rather, two differences remain: acceptance of a “simple” versus a “complex” account of aggregation and whether or not determinative (not causal!) higher-level entities exist. Simple fundamentalist scientific reductionists alone accept the simple view of aggregation: contributions of powers by the components in complex collectives are continuous with their contributions in simple collectives and the contribution of powers by components is determined solely by other components. All scientific emergentists and conditioned fundamentalist scientific reductionists, on the other hand, accept the conditioned view of aggregation: contributions of powers by the components in complex collectives are discontinuous with their contributions in simple collectives, and components contribute differential powers. On the second difference, both kinds of scientific reductionists deny determinative higher-level (collectivist) entities: components alone determine all existing entities, properties, relations, and processes. Scientific emergentists alone accept that fully composed higher-level kinds nevertheless “machretically” determine (not cause!) parts’ activities in (some) collectives: collectives nonproductively determine some of the powers of the components through “role-shaping” or “role-constraining,” but not through “joint role-filling.” Gillett provides a useful diagram in chapter 8 that summarizes the live positions and their key differences.

There is much in this book for metaphysicians—of science, for science, or otherwise—to sink their teeth into. Gillett is careful about stating and defending criteria for the correctness of his accounts, not only for scientific componentry (noted above) but for many of the notions developed throughout the book. He offers both detailed verbal explanations and precise definitions of his key concepts. Unlike many metaphysical treatises, he even offers a seven-page glossary for his key theoretical terms. His figures, while not professionally drawn, are often useful and give the cozy appearance of impromptu whiteboard illustrations. (However, some could have used more extensive labeling and captioning.) And his many swipes at contemporary philosophical/metaphysical orthodoxy stand out and certainly will not be ignored.

There are also features that will cause philosophers of science of an antimetaphysical bent to throw up their hands in exasperation. Although Gillett writes clearly, the book is a long slough that requires a systematic, concerted effort to work through. Many of his precise definitions of single concepts are multiple lines in length and thus are difficult to dissect. Those less sympathetic to metaphysics may well find some of Gillett’s hair-splitting distinctions not worth the effort. One example is his distinction between ‘determination’ and ‘cause’. Yet this distinction figures centrally in his arguments to dodge standard challenges to emergence based on the causal completeness of physics. Like many metaphysicians, Gillett uses unfamiliar terms for key concepts (e.g., ‘machresis’, ‘coadunative’) and employs acronyms (e.g., ‘MEAT’ for ‘Meta-Explanatory Adequacy Test’). Be advised to bookmark his glossary as you read. While he continuously talks about the contemporary scientific reductionism-emergentism debate raging among “working scientists,” most of the detailed quotations come from only two physicists, Steven Weinberg for reductionism and Robert Laughlin for emergentism. Most of Gillett’s descriptions of the empirical cases for the live views in the scientific reductionism-emergentism debate (in chap. 10) are lacking in much real scientific detail. And his account of “threshold,” the minimum amount of empirical data needed to establish a “local” argument for one of the competing live views, seems to require nothing less for fulfillment than a complete account from the constituent-level science. One doubts seriously whether any advocate of scientific reductionism or emergence is under the illusion that he or she has met that standard. Such views are usually advocated as best guesses, as predictions about where the empirical evidence appears to be headed.

One claim in Gillett’s book sure to provoke philosophers who regularly interact professionally with scientists occurs at the book’s end, where Gillett insists that his research has now set the stage for guiding future empirical work toward resolving this new reductionist-emergentist “battle of the ages.” One of the first things a philosopher has to accept when he or she engages professionally with practicing scientists and their day-to-day work is that scientists do not take themselves to be in the business of doing the work that philosophers think is most relevant or important. If one turns to scientists hoping to find such work, one is sure to be quickly disappointed and frustrated. Scientists work on issues that scientists deem important and relevant. That is not to say that philosophers cannot learn a lot that is of philosophical interest by working with scientists on scientific endeavors. But what philosophers end up learning invariably is not what they wandered in at first to find. No matter how strongly Gillett and others crave to do “metaphysics of science” rather than “for science,” merely “doing metaphysics” according to the strictures of contemporary philosophy imposes an alien practice on science. In addition to the philosophically humbling quote from Bertrand Russell with which Gillett closes the book, another from Wilfrid Sellars would do him (and his fellow metaphysicians of science) well to consider: “does the reader not recognize Jones as Man himself in the middle of his journey from the grunts and groans of the cave to the subtle and polydimensional discourse of the drawing room, the laboratory, and the study … and of the philosophers who, in their efforts to break out of discourse to an arche beyond discourse, have provided the most curious dimension of all” (“Empiricism and the Philosophy of Mind,” in Minnesota Studies in the Philosophy of Science, vol. 1, ed. H. Feigl and M. Scriven [Minneapolis: University of Minnesota Press, 1956], 328).