Hugh Desmond

Biologists explain organisms’ behavior not only as having been programmed by genes and shaped by natural selection, but also as the result of an organism’s agency: the capacity to react to environmental changes in goal-driven ways. The use of such ‘agential explanations’ reopens old questions about how justified it is to ascribe agency to entities like bacteria or plants that obviously lack rationality and even a nervous system. Is organismic agency genuinely ‘real’ or is it just a useful fiction? In this paper we focus on two questions: whether agential explanations are to be interpreted ontically, and whether they can be reduced to non-agential explanations (thereby dispensing with agency). The Kantian approach we identify interprets agential explanations non-ontically, yet holds agency to be indispensable. Attributing agency to organisms is not to be taken literally in the way we attribute physical properties such as mass or acceleration, but nor is it a mere heuristic or predictive tool. Rather, it is an inevitable consequence of our own rational capacity: as long as we are rational agents ourselves, we cannot avoid seeing agency in organisms.

If evolutionary history were to be replayed from the beginning, what would be the same, and what would be likely different? Would there be a human-like species, multicellularity, or even DNA? There is a great variety in the answers biologists give to this question, despite having the same access to empirical data and biological theory. For instance, Stephen J. Gould has claimed that evolutionary history is radically contingent, while Conway Morris holds that it converges onto specific biological structures that are favored by natural selection. Others such have proposed that evolutionary history is characterized by an inexorable increase in complexity, while others see it as an evolutionary arms race. In this dissertation I investigate the fundaments underlying claims biologists make about contingency and directionality in evolutionary history as a whole. The topics of convergence and contingency have received attention from philosophers of biology in recent years, but the foundations of interpretations of evolutionary history as a whole remains a relatively neglected field. Hence the primary objective of this dissertation was not to defend this or that account, but rather to show a method by which these fundamental issues can be identified and analyzed constructively. The dissertation is organized into two parts, each dedicated to a single problem. The first part concerns the problem of ‘description dependence’: claims about the contingency of evolutionary outcomes depend on how these outcomes and the evolutionary process itself are described. I set out to map the different ways in which the contingency of outcomes changes as the phenomena are described in more and in less detail, and as broader or narrower subsets of evolutionary history are taken into account. That such an analysis is useful to pursue, I attempt to show by applying it to two of the most prominent interpretations of evolutionary history, those of Gould and Conway Morris. According to how their claims about evolutionary history are analyzed, one can arrive at opposing conclusions about the contingency of evolutionary outcomes. The second part concerns the problem of ‘causal complexity’: evolutionary history is a complex mess of unrelated causal processes, and for every generalization there is an exception. This raises the question whether non-speculative generalizations over evolutionary history as a whole are even possible. Limiting the scope of the investigation to the mechanism of natural selection alone, I consider first whether and how natural selection may be expected to give rise to trends at all. Some philosophers reject that natural selection is a cause at all, and that all evolution is simply an accumulation of births and deaths. If any trend occurs at all, it is due to a confluence of unrelated causal processes that could easily not have occurred. I argue against this by showing that these philosophers overlook the issue of time-scale: causal processes may make a difference for reproductive outcome at a time-scale of a single generation without them making a difference at the time-scale of multiple generations. With this distinction in mind, one can argue that natural selection causes a population to tend towards equilibrium. If a yet longer time-scale is taken – not that of multiple generations in a single environment, but that of many species and genuses across many environments – the challenge of causal complexity becomes much more difficult to overcome. Drawing on the phenomenon of phenotypic plasticity and niche construction, I argue that selection for plasticity ‘feeds’ on this complexity and variability in the environment. The trend in plasticity is unique in this regard, since trends in other types of adaptation are interrupted by the variability of complex environments.

Some of the most significant policy responses to cases of fraudulent and questionable conduct by scientists have been to strengthen professionalism among scientists, whether by codes of conduct, integrity boards, or mandatory research integrity training programs. Yet there has been little systematic discussion about what professionalism in scientific research should mean. In this paper I draw on the sociology of the professions and on data comparing codes of conduct in science to those in the professions, in order to examine what precisely the model of professionalism implies for scientific research. I argue that professionalism, more than any other single organizational logic, is appropriate for scientific research, and that codes of conduct for scientists should strengthen statements concerning scientific autonomy and competence, as well as the scientific service ideal.

Implicit contextual factors mean that the boundary between causal and noncausal explanation is not as neat as one might hope: as the phenomenon to be explained is given descriptions with varying degrees of granularity, the nature of the favored explanation alternates between causal and non-causal. While it is not surprising that different descriptions of the same phenomenon should favor different explanations, it is puzzling why re-describing the phenomenon should make any difference for the causal nature of the favored explanation. I argue that this is a problem for the ontic framework of causal and noncausal explanation, and instead propose a pragmatic-modal account of causal and non-causal explanation. This account has the added advantage of dissolving several important disagreements concerning the status of non-causal explanation.

Many have argued that there is no reason why natural selection should cause directional increases in measures such as body size or complexity across evolutionary history as a whole. In this paper I argue that this conclusion does not hold for selection for adaptations to environmental variability, and that, given the inevitability of environmental variability, trends in adaptations to variability are an expected feature of evolution by natural selection. As a concrete instance of this causal structure, I outline how this may be applied to a trend in phenotypic plasticity.

It is an ongoing controversy whether natural selection is a cause of population change, or a mere statistical description of how individual births and deaths accumulate. In this paper I restate the problem in terms of the reference class problem, and propose how the structure of stable equilibrium can provide a solution in continuity with biological practice. Insofar natural selection can be understood as a tendency towards equilibrium, key statisticalist criticisms are avoided. Further, in a modification of the Newtonian-force analogy, it can be suggested that a better metaphor for natural selection is that of an emergent force, similar in nature to entropic forces: with magnitude and direction, but lacking a spatiotemporal origin or point of application.

Path-dependence offers a promising way of understanding the role historicity plays in explanation, namely, how the past states of a process can matter in the explanation of a given outcome. The two main existing accounts of path-dependence have sought to present it either in terms of dynamic landscapes or branching trees. However, the notions of landscape and tree both have serious limitations and have been criticized. The framework of causal networks is both more fundamental and more general that that of landscapes and trees. Within this framework, I propose that historicity in networks should be understood as symmetry breaking. History matters when an asymmetric bias towards an outcome emerges in a causal network. This permits a quantitative measure for how path-dependence can occur in degrees, and offers suggestive insights into how historicity is intertwined both with causal structure and complexity.