Emergence and Convergence: Qualitative Novelty and the Unity of Knowledge (review)

Early in the twentieth century a consensus developed in science and philosophy of science that ontological emergence (wholes have properties not reducible to, or derivable from, the properties of their parts) is untenable. In the latter part of the twentieth century, dating from the 1960s - around the time Ilya Prigogine's work on non-equilibrium thermodynamic dissipative systems gained prominence - an increasing chorus of scientists and philosophers of science have questioned this consensus. The current debate on emergence as an alternative to reductionism - in particular micro-reductionism - is lively, disciplined, and rooted in contemporary science and the mathematical models used to describe complex, and often non-linear, systems (e.g., atmospheric turbulence). As with most contentious and exciting debates, there is a sense of breaking new ground, but there are also a lot of specious arguments and excessive claims.

By contrast with this engaging and far from settled debate, Mario Bunge's book is stale: a return to old arguments and old attitudes. As such, it contributes little to the current debate. Instead, he refuses even to engage in the debate by defining emergence in a way that empties it of the crucial feature relevant to the current debate. Bunge states: 'My own definition of emergence is this. To say that P is an emergent property of a system of kind K is short for "P is a global [or collective or non-distributive] property of a system of kind K, none of whose components or precursors processes P."'

It is not that advocates of ontological emergence reject this feature; they simply consider it to lack the essential feature of meaningful emergence, namely, that P cannot be reduced to, or deduced from, properties of the components or precursors. Bunge summarily dismisses this definition: 'The standard dictionary definition of "emergence" mistakenly identifies it with the impossibility of understanding a whole through an analysis into its components and their interactions' (emphasis added).

The dictionary definition, however, captures the way the word has been, and is, used by most people; philosophers distinguish between epistemological and ontological emergence but recognize that the contentious meaning of 'emergence' is ontological emergence. Bunge proffered his definition earlier (1977), but a lot has happened in science and philosophy of science during the almost thirty years since. That he chooses to address a concept of emergence that lacks an essential feature of the concept as currently hotly debated is not surprising. Bunge's views are rooted in a logical empiricist's conception of science, as the following makes clear: 'Some philosophers such as Ernest Nagel (1961) and Carl G. Hempel (1965) have rightly rejected the holist construal of emergence as an ontological category, as being imprecise. They have admitted "emergence" only as an epistemological category equivalent to "unexplainable [or unpredictable] by means of contemporary theories" - the way Broad had proposed.' Nagel and Hempel were central figures in the logical empiricist movement - a movement that, although impressive half a century ago, has been rejected by a majority of philosophers, historians, psychologists, and sociologists of science as well as scientists.

It is difficult to comprehend how a book on emergence published in 2003 could fail to take account of the writings of Sunny Auyang. It is also difficult to understand how someone could, anytime after the early 1980s, claim, 'The centrepiece of human socio-biology is its hypothesis of kin selection, according to which we are "designed" by natural selection so as to behave altruistically towards those who share most of our genes with us - that is, our kin, in particular child, parent or sibling.' W.D. Hamilton's concept of inclusive fitness (a concept shaped a decade before E.O. Wilson's Sociobiology: The New Synthesis) allowed a compelling population genetic explanation of biological altruism in organisms like ants and termites - social insects. If human sociobiology has a cornerstone, it is reciprocal altruism, game theory, or mechanisms of culture - gene interaction. These are vastly more important than kin selection (inclusive fitness). There...