The last few weeks have been dedicated to finishing off my 2000 paper, the grand finale paper of my coursework. Below is an abstract I just submitted to the &HPS4 conference, which is basically a 1000 word summary of this 12,000 word beast. I haven’t posted anything lately, mostly because I’ve been entirely preoccupied with trying to turn this thing into something resembling a coherent whole.
Writing is just agony.
Historicity in Scientific Explanation
John Beatty and Eric Desjardins argue that historicity is a crucial feature of Biology, and that contingency is key to understanding historicity. I will argue that while this is true, mere contingency is not enough to capture the interesting methodological consequences of historicity for Biology. Drawing on the work of historians such as John Harley Warner, Peter Galison and Roger Smith, I try to articulate one of the important features of modern historical narrative, which Beatty and Desjardins’ account of historicity lacks: its detail-oriented context sensitivity. Wilhelm Windelband called this the ideographic character of historical narrative, which stands opposed to the nomothetic or abstractive, law-giving mode typically associated with the physical sciences. I will try to characterize these two explanatory styles, and show how they can enhance Beatty and Desjardins work on historicity.
Beatty and Desjardins follow Steven J. Gould in treating contingency as the “essence” of historicity. To use Desjardins formulation, historicity entails: (1) multiple possible pasts, (2) multiple possible outcomes at a given instant, and (3) a relationship of causal dependence between these two sets. But these three conditions can be trivially satisfied by very simple systems which do not show the rich methodological complexity which Beatty and Desjardins take to be entailed by historicity. Specifically, Beatty argues that historicity entails theoretical pluralism, which itself gives rise to relative significance controversies. The typical “Newtonian” approach to science requires that we eschew the “pomp of superfluous causes”, and try to explain a phenomenon by postulating as few causal processes as possible. In Biology, in contrast, theoretical pluralism is the rule, meaning that multiple causal stories are treated as complementary and overlapping explanations. The style of debate that Biologists typically enter into, Beatty argues, is not over which account is correct, but rather which account is more relevant. This is an interesting and important observation about Biology, but it does not follow from the characterization of historicity above. Very simple quantum mechanical systems, for example, would satisfy Desjardins three criteria, but apparently require no pluralism in their explanation. So something more is needed.
I propose therefore to add a fourth criterion to Desjardins list: (4) the causal process must unfold in a context-sensitive way. That is, for historicity to obtain it must be the case that the causal process in question cannot be explained by abstracting away the larger systems in which it is embedded, or by treating the phenomenon as merely an instance of a category or universal. This is the requirement that history be ideographic. In history, things only ever happen once. To understand an event is to see it in its uniqueness, by situating it in some broader context. The prototypical scientific explanatory style, which Beatty wishes to contrast historicity with, does the opposite. It treats all events as instances of some category – this is the nomothetic style. In terms of the criterion I propose to add to the account of historicity on offer, context sensitivity means ideographic reasoning. If it is not the case that the particularity of an event needs to be considered in order to explain it, then that explanation is context-insensitive, and vice versa. It is crucial to note that not all science is nomothetic, nor has all historical narrative been ideographic. But I hope to show that sensitivity to the difference between the two styles can help elucidate the debate over historicity in Biology.
Context-sensitivity can be formally described using dynamical systems theory, which provides an analogy for when a system can be considered sensitive to its context. Andreas Wagner describes the problem of finding regular, repeatable causes in non-linear dynamical systems. Non-linearity means that effects are not proportional to their causes. Some non-linear systems can be dealt with by finding regions of linearity. To take the simplest case,can be linearized by taking the first derivative, and so a clear proportionality can be described between the values of y and x. But in systems with a great many variables, a change in any of which can produce non-proportional effects throughout the system, it quickly becomes impossible find linearizable regions. When a system is densely non-linear in this way, nomothetic description of the interaction between parts becomes problematic, because laws just are a way of talking about the proportionality of causes and effects. Any notion of causality based on regularities also becomes impossible for the same reason. If we know the value of all of the variables in a non-linear system (as in a simulation, for example) we can give a detailed description of how each variable affected each other. But if the system is non-linear enough, saying what lead to a given change in any one variable will involve describing the entire system. This, I propose, is analogous to ideographic explanation, where a detail-oriented narrative situates some unique fact in its context.
If this analysis is correct, it shows why historicity is particularly important in Biology, where densely non-linear systems are the rule. It also helps to show how historicity is related to relative significance controversies. Because the whole context of an event embedded in a densely non-linear system is relevant to its determination, we can only ever tell part of the story. Biologists (and historians) try to zero in on the most relevant parts, but success can only ever be partial. Nomothetic explanations have the virtue of offering the possibility of complete specification of relevance, but only in very special explanatory circumstances will it be possible to give a truly context-insensitive explanation. Attending to the difference between the two helps understand the role of history in Biological explanation.